Tune that dial - biblecast.net.br
free supplement
DECEMBER 2006
£3.80
www.elektor.com
Tune that dial
shortwave receiver
with DDS
2.4 GHz WLAN antennas
in practice
Unusual Christmas presents
gadgets & wannahaves
Extra! free 24 page supplement
The i-TRIXX Collection
R50
Spy radio stations • Flowcode v. 3 is here
Boxing up prototypes
•
IR Romote Control
Extender MKII
KC-5432 £7.25 + post & packing
Operate your DVD player or digital decoder
using its remote control from another room. It
picks up the signal from the remote control and
sends it via a 2-wire cable to an infrared LED
located close to the device. This improved
model features fast data transfer, capable of
transmitting Foxtel digital remote control signals
using the Pace 400 series decoder. Kit supplied
with case, screen printed front panel, PCB with
overlay and all electronic
components.
Im
prove
Model! d
Requires 9VDC
wall adaptor
(Maplin #GS74R
£9.99)
Battery Zapper MKII
KC-5427 £29.00 + post & packing
This kit attacks a common cause of failure in wet
lead acid cell batteries: sulphation. The circuit
produces short bursts of high level energy to
reverse the damaging sulphation effect. This new
improved unit features a battery health checker
with LED indicator, new circuit protection against
badly sulphated batteries, test points for a DMM
and connection for a battery
charger. Kit includes case with
Improved
screen printed lid, PCB with
Model!
overlay, all electronic
components and clear
English instructions.
Suitable for 6, 12 and 24V
batteries
• Powered by the battery
itself
py
m
REE co
For a F catalogue fro )
e
m
g
le
a
rob
410+ p
rickly p
of our
’s no p
e
t
it
(I
s
a
b
li
e
a
ue
our w
Austr
o
t
n
catalog
/
o
k
log
ics.co.u
n
o
r
t
c
£ Stg
ycarele
www.ja All prices in
Galactic Voice Kit
KC-5431 £13.25 + post & packing
Be the envy of everyone at the
next Interplanetary
Conference for Evil Beings
with this galactic voice
simulator kit. Effect and
depth controls allow you to
vary the effect to simulate
everything from the
metallically-challenged C-3PO,
to the hysterical ranting of Daleks hell-bent on
exterminating anything not nailed down. The kit
includes PCB with overlay, enclosure, speaker and all
components. For those who really need to get out of
the house a lot more. Take me to your leader.
• Requires 9V battery
High Range Adjustable
Switch with LCD
KC-5376 £22.75 + post & packing
This temperature switch can be set anywhere up
to 1200ºC, so it’s extremely versatile. The relay
can be used to trigger an extra thermo fan on an
intercooler, mount a sensor near your turbo
manifold and trigger water spray cooling, or a
simple alarm to warn you of overheating. The LCD,
which can easily be dash mounted, displays the
temperature constantly. Kit supplied with solder
masked PCB with overlay, LCD panel, temperature
probe and all electronic components.
n the
sed i oys
u
s
A ach B
d
Be hit ‘Goo
c
’
i
s
s
s
n
cla ibratio
V
Requires 9-12VDC
wall adaptor
(Maplin #JC91Y £14.99)
Smart Fuel Mixture Display Kit
KC-5374 £8.95 + post & packing
This kit features auto dimming for night driving,
emergency lean-out alarm, better circuit
protection, and a ‘dancing’ display which functions
when the ECU is operating in closed loop. Kit
supplied with PCB and all electronic components.
• Car must be fitted with air flow and
EGO sensors (standard on all EFI
systems) for full
functionality.
Recommended box
UB3 (HB-6013)
£1.40 each
Log on to
www.jaycarelectronics.co.uk/catalogue
for your FREE catalogue!
0800 032 7241
Magnetic Cartridge Pre-amp
KC-5433 £11.75 + post & packing
This kit is used to amplify the 3-4mV signals from
a phono cartridge to line level, so you can use your
turntable with the CD or tuner inputs on your Hi-Fi
amplifier - most modern amps don't include a
phono input any more. Dust off the old LP
collection or use it to record your LPs on to CD.
The design is suitable for 12" LPs, and also allows
for RIAA equalisation of all the really old 78s.
Please note that the input sensitivity of this design
means it's only suitable for moving-magnet, not
moving-coil cartridges. Kit
includes PCB with overlay and all
electronic components.
• Requires 12VAC power
Universal High Energy Ignition Kit
KC-5419 £27.75 + post & packing
A high energy 0.9ms spark burns fuel faster and
more efficiently to give you more power! This
versatile kit can be connected to conventional
points, twin points or reluctor ignition
systems. Kit supplied with
diecast case, PCB
and all electronic
components.
We stock
an extensive
range of quality
automotive
kits
Theremin Synthesiser MKII
KC-5426 £43.50 + post & packing
By moving your hand between the metal
antennae, create unusual sound effects! The
Theremin MkII improves on its predecessor by
allowing adjustments to the tonal quality by
providing a better waveform. With a multitude of
controls, this instrument's musical potential is only
limited by the skill and imagination of its player.
Kit includes stand, PCB with overlay, machined
case with silkscreen
printed lid, loudspeaker,
Improved
pitch antennae, all
Model!
specified electronic
components
and clear English
instructions.
POST AND PACKING CHARGES:
Cost
Order Value
Cost Order Value
£200 - £499.99 £30
£20 - £49.99 £5
£40
£50 - £99.99 £10 £500+
£100 - £199.99 £20
Max weight 12lb (5kg). Heavier
parcels POA. Minimum order £20.
All prices
in £ Stg
(Monday - Friday 09.00 to 17.30 GMT + 10 hours only).
For those who want to write: 100 Silverwater Rd
Free
Silverwater NSW 2128 Sydney AUSTRALIA
410+ page
Catalogue
Intelligent Turbo Timer Kit
KC-5383 £14.75 + post & packing
This great module uses input from an airflow,
oxygen, or MAP sensor to determine how hard
the car has been driven. It then uses this
information to calculate how long the car needs to
idle, reducing unnecessary idle time. The
sensitivity and maximum idle time are both
adjustable, so you can be sure your turbo
will cool properly. Kit supplied with
PCB, and all electronic
components.
Recommended
box UB3 (HB-6013)
£1.40 each
PC Oscilloscopes & Analyzers
DSO Test Instrument Software for BitScope Mixed Signal Oscilloscopes
DSO
2.0
4 Channel BitScope
Digital Storage Oscilloscope
�
Up to 4 analog channels using industry standard
probes or POD connected analog inputs.
Mixed Signal Oscilloscope
�
Capture and display up to 4 analog and 8 logic
channels with sophisticated cross-triggers.
Spectrum Analyzer
�
Integrated real-time spectrum analyzer for each
analog channel with concurrent waveform display.
Logic Analyzer
�
8 logic, External Trigger and special purpose
inputs to capture digital signals down to 25nS.
2 Channel BitScope
Pocket Analyzer
BitScope DSO Software for Windows and Linux
BitScope DSO is fast and intuitive multi-channel test and measurement software for your
PC or notebook. Whether it's a digital scope, spectrum analyzer, mixed signal scope,
logic analyzer, waveform generator or data recorder, BitScope DSO supports them all.
Capture deep buffer one-shots or display waveforms live just like an analog scope.
Comprehensive test instrument integration means you can view the same data in
different ways simultaneously at the click of a button.
DSO may even be used stand-alone to share data with colleagues, students or
customers. Waveforms may be exported as portable image files or live captures replayed
on other PCs as if a BitScope was locally connected.
BitScope DSO supports all current BitScope models, auto-configures when it connects
and can manage multiple BitScopes concurrently. No manual setup is normally required.
Data export is available for use with third party software tools and BitScope's networked
data acquisition capabilities are fully supported.
Data Recorder
�
Record anything the DSO can capture and
display. Supports live data replay image export.
Networking
�
Flexible network connectivity supporting
multi-scope operation, remote monitoring and
data acquisition.
Data Export
�
Export data with DSO using portable CSV files or
use libraries to build custom BitScope solutions.
www.bitscope.com
From
the Editor’s desk
With this extra-thick December issue we
close off the publication year 2006. I’ve
a few points to share with you all and I
guess it does no harm to mention them
in this month’s Editorial space — in
random order!
All change to dot com. As you may
have noticed on last month’s front cover,
our website is also available at
www.elektor.com. We are using this
domain name because it is easier to
remember (and type!) than
www.elektor-electronics.co.uk .
However, both websites run 100% in
parallel and the old domain name will
continue to be used.
The i-TRIXX supplement you get free
of charge with this issue is actually
the sixth gift our readers get this year.
Earlier this year we gave away a Visual
Basic booklet (January 2006); a C
booklet and an FPGA poster (March
2006); an RFID card (September
2006); and an E-Simulation DVD (October 2006). The circuits in the 24-page
i-TRIXX supplement are aimed at the
younger generation (11 to 15s) and we
would appreciate if you would let us
know what you think of it.
We’re hiring! Due to our expanding
activities we’re looking for:
• free-lance translators, German-English
and Dutch-English;
• a free-lance News Correspondent (UK
based);
• a free-lance Books Acquisition Manager (UK based).
For details regarding these positions,
please contact me on
editor@elektor.com and
I will put you in touch with the responsible person within our organisation.
The Cumulative Index for Elektor Electronics volume 32 (2006) will be available
as a free pdf download from our website. Like last year, we feel that the three
precious pages taken up by the year
index are better used for projects and
other articles. Readers without internet
access may write or telephone to request
a free copy of the document on paper.
Finally, from all of us here at Elektor, we
thank all our readers, business partners
and advertisers for their continued support during the past year and wish you
all a Merry Christmas and a peaceful
and prosperous 2007!
Jan Buiting, Editor
24 Shortwave
Capture
As a special treat for all radio amateurs we
present a general-coverage AM/FM/SSB receiver with a wide range of features, which uses a
DDS chip in the VFO section and also has a DRM output
that can be fed into a computer. The receiver is controlled
by an 8-bit Atmel RISC processor. The frequency readout is
on a clearly legible 7-segment LED display.
20 Tightly Packed
There are an incredible number of options these days
for the safe ‘packing up’ of electronic circuits. Open
any catalogue from any well-known mail-order company and you will get an impression of the extensive
range on offer. This article gives an overview of the
different types with their particular characteristics and
provides hints as to how you can make a professional
looking front panel yourself.
age
p
4
2
E
E
FR
ment
e
l
p
p
u
s
i-TRIXX
starts
ge
on pa
45
70 WLAN Antenna Design
The domestic use of WLANs has grown rapidly as DSL routers with
built-in wireless Ethernet have become available, and now it is easy to
use a notebook PC to surf
the Internet wirelessly
from the comfort of
one’s sofa. However,
things get trickier if
a reinforced concrete wall stands
in the way, or if a
neighbour happens
to be using the same
frequency...
CONTENTS
Volume 32
December 2006
no. 360
know-how
70 WLAN Antenna Design
76 Where am I —
and Where are the Others?
hands-on
23 Multi-Purpose 3D Milling
Machine (announcement)
24 Shortwave Capture
42 Go with the Flow
79 Design Tips
Client-server quizmaster
Pencil rubber cleans solder pads
84 FPGA Course (7)
88 A New Flowcode
92 Intelligent Voltmeter in a
Plug
94 A Wire with Total Recall
88 A New Flowcode
A new version of
Flowcode for E-blocks
has just been released
— version 3. This is
more than a simple
upgrade: Flowcode
has matured into a
nice if not impressive
development tool.
technology
34 Spy Radio Stations
38 Wireless Key
80 Radio Control using
WLAN ICs
96 Smaller is not Always Better
info & market
6
Colophon
8
Mailbox
10 News & New Products
20 Tightly Packed
108Sneak Preview
infotainment
14 Christmas Presents
100SSB Receiver for 20m and
80m (1987)
101Hexadoku
Subscriptions: Elektor Electronics (Publishing),
Regus Brentford, 1000 Great West Road, Brentford TW8 9HH, England.
Tel. (+44) 208 261 4509, fax: (+44) 208 261 4447
Internet: www.elektor.com
Email: subscriptions@elektor.com
Rates and terms are given on the Subscription Order Form
Volume 32, Number 360, December 2006
ISSN 0268/4519
Elektor Electronics aims at inspiring people to master electronics at any personal
level by presenting construction projects and spotting developments in
electronics and information technology.
Publishers: Elektor Electronics (Publishing), Regus Brentford,
1000 Great West Road, Brentford TW8 9HH, England. Tel. (+44) 208 261 4509,
fax: (+44) 208 261 4447 www.elektor.com
The magazine is available from newsagents, bookshops and electronics retail outlets, or on
subscription. Elektor Electronics is published 11 times a year with a double issue for July & August.
Under the name Elektor and Elektuur, the magazine is also published in French, German and
Dutch. Together with franchised editions the magazine is on circulation in more than 50 countries.
International Editor: Mat Heffels (m.heffels@segment.nl)
Head Office: Segment b.v. P.O. Box 75 NL-6190-AB Beek
Telephone: (+31) 46 4389444, Fax: (+31) 46 4370161
The Netherlands
Distribution: Seymour, 2 East Poultry Street, London EC1A, England
Telephone:+44 207 429 4073
UK Advertising: Huson International Media, Cambridge House, Gogmore Lane,
Chertsey, Surrey KT16 9AP, England.
Telephone: +44 1932 564999, Fax: +44 1932 564998
Email: r.elgar@husonmedia.com
Internet: www.husonmedia.com
Advertising rates and terms available on request.
International Advertising: Frank van de Raadt, address as Head Office
Email: advertenties@elektuur.nl
Advertising rates and terms available on request.
Copyright Notice
Editor: Jan Buiting (editor@elektor.com)
International editorial staff: Harry Baggen, Thijs Beckers, Ernst Krempelsauer,
Jens Nickel, Guy Raedersdorf.
Design staff: Ton Giesberts, Paul Goossens, Luc Lemmens, Christiaan Vossen
Editorial secretariat: Hedwig Hennekens (secretariaat@segment.nl)
Graphic design / DTP: Giel Dols
Managing Director / Publisher: Paul Snakkers
Marketing: Carlo van Nistelrooy
Customer Services: Margriet Debeij (m.debeij@segment.nl)
The circuits described in this magazine are for domestic use only. All drawings, photographs, printed
circuit board layouts, programmed integrated circuits, disks, CD-ROMs, software carriers and article
texts published in our books and magazines (other than third-party advertisements) are copyright
Segment. b.v. and may not be reproduced or transmitted in any form or by any means, including
photocopying, scanning an recording, in whole or in part without prior written permission from
the Publishers. Such written permission must also be obtained before any part of this publication is
stored in a retrieval system of any nature. Patent protection may exist in respect of circuits, devices,
components etc. described in this magazine. The Publisher does not accept responsibility for failing
to identify such patent(s) or other protection. The submission of designs or articles implies permission to the Publishers to alter the text and design, and to use the contents in other Segment publications and activities. The Publishers cannot guarantee to return any material submitted to them.
Disclaimer
Prices and descriptions of publication-related items subject to change. Errors and omissions excluded.
© Segment b.v. 2006
Printed in the Netherlands
elektor electronics - 12/2006
12/2006 - elektor electronics
7
info & market mailbox
Poor man’s VGA Tester
Hi Elektor people — I attach a circuit diagram of a simple
VGA Tester. The circuit is suitable for direct connection
to a VGA display with a resolution of 480×640 pixels
and generates a chessboard pattern. I designed the tester
around a PIC12F508. It contains very few components
and I believe the circuit speaks for itself. Jumper JP1 permits the colour selection between red, green and blue. By
replacing it with three diodes (1N4148), the test picture
goes black and white. The video output level is adjustable
with preset P1.
The software I wrote for the tester is also simplicity itself.
In principle, a loop is executed in which the image is built
up bit by bit. The listing contains information regarding
the pulse timing, which should enable users to adapt the
program to suit other resolutions.
I designed the circuit with the help of MPLAB IDE v7.20
and Eagle 4.16.
Hans Kooij (Netherlands)
bias voltage. I would
expect 1N4148s to work
equally well, though.
RS232 (transmission)
• 2 x 22 pin A1 compatible
slot
Thanks for publishing my design and hope a few readers
benefit from it.
A nice change, I would say,
from all that new fangled stuff
around.
Marcel van de Gevel
(Netherlands)
Thanks Marcel, and our apologies for the errors in reproducing
your design. With over 100 article files being produced in
four languages within a period
of about four weeks, the production of our Summer Circuits
edition is a tour de force where
errors can not be ruled out entirely, particularly when making
the drawings.
Apple-01 Replica
computer
Dear Jan — just tro say that I
built a replica of the 30 year
old Apple-1 computer (see
photo). I was honoured to get
a personal ‘OK’ from Steve
Franz Achatz (Germany)
Free e-SIM DVD
Dear Jan — I believe 754C1
is the answer to Hexadoku,
October 2006 (correct! Ed.).
I was looking for something
to read at the shop, and
found your magazine which
I subscribed to in my student
days, some twenty years ago.
I have already tried some of
the programs on the e-SIM
DVD, thanks for bringing me
back to the days of PCBs and
simulations!
Magnus R. Berg (Norway)
Welcome back Magnus, you’re
in good company here.
Thanks Hans, we agree that your circuit is hard to beat in
terms of component count. The signal is composed entirely by
the PIC and with some dexterity the tester could be built into a
VGA plug. The source codce is available for free downloading
from our website — the file number is 060215-11.zip.
Phono Splitter —
some points to note
Dear Editor — I write to
mention a few small errors
that apparently have kept in
my project ‘Phono Splitter’
published in the July/August
2006 issue.
• Compensation capacitor
C4 should have a value of
47 pF, not 470 pF.
• T1 should be a BC560C
like T2 and T3.
• In my prototype, diodes
D2 and D3 were types
1N4448, mainly because
oif their tighter specifications in respect of forward
Wozniak to reuse his A1 firmware on my replica computer
which I dubbed ‘A-ONE’.
The A-ONE works fine as far
as I can check. Here are some
data:
• 6502 at 1 MHz
• 6821 PIA
• 32 kB RAM
• EPROM with WOZ-Mon
and WOZ BASIC
• GAL for address decoding
etc.
• TINY2313 for PS2
keyboard and RS232
(reception)
• MEGA32 for video and
Pontavi-Thomson Bridge
Dear Jan — there is nothing
new in this life! The above
bridge (Retronics, September
2006, Ed.) is a simplified
version of the Kelvin Bridge
and featured in a few old
books on calibration. I have
a splendid version in a teak
case with a large brass scale,
made by Pye of Cambridge,
and it is accurate to .1%.
I have a collection of over
twenty bridges made
by Sullivan, Cambridge
Instruments, Wayne & Kerr,
Marconi, etc. together with
elektor electronics - 12/2006
RFID Quest extended
To give readers more time to read out the number stored on the free RFID card they got
with their September 2006 issue (using a home built reader unit or one of the designs
published in the same issue), the period for reporting winning card numbers has been
extended to 15 December 2006. See also the RFID pages at www.elektor.com
MailBox Terms
•Publication of reader’s
correspondence is at the
discretion of the Editor.
•Viewpoints expressed by corres­
pondents are not necessarily
galvanometers, standards and
precision potentiometers. The
workmanship is incredibly
high and part of our heritage
we should all be proud of.
John Price (UK)
Dear Editor — I believe the
relatively low output power of the
High-End Power Amplifier from
the March 2005 issue is mostly
owing to the enclosure used. In
other words, if a larger cabinet
is used, more space is available
to step up the power supplied
by the amp. For example, the
transformer voltage can be
increased from 18 V to 25 V.
After rectification, this results in
about 31 VDC. The modifications I carried out at the
component level are:
Power supply:
225 VA toroidal transformer,
2x22 V
C5, C6, etc.: 10,000 µF 35 V
Fuses: 1.5 A
Two NTCs in series with the mains
voltage
Amplifier board:
R18: 10k
R42: 220k
R45: 220Ω
D14;D15: 12V
Heatsink: 2 x Fischer SK 155, 75 mm
(0.9 K/W)
Indicator board:
R16;R17;R33;R34: 330Ω
R5;R22: 820Ω
12/2006 - elektor electronics
R6;R23: 10k
P1 and P2 allow the gain of the indicator board to be adjusted between
9 and about 13.7 times. I selected
the Monacor (Monarch) type UC204/SW case which has a size of
437×82×235 mm. Because of the larger
output power, the indicator board is no
longer required, hence I did not fit the
LEDs on the front panel. Because the
feedback is reduced, I fear the distortion
goes up while damping is reduced.
Lacking high-end test equipment I am
unable to say if my modifications reduce
the amplifier’s performance in any way.
P. Kempenaar (Netherlands)
•Correspondence may be
translated or edited for length,
clarity and style.
•When replying to Mailbox
correspondence,
please quote Issue number.
Indeed we should, John. Thanks
for letting us know about the
origins of the Pontavi-Thomson
bridge. Photograph reproduced
courtesy of Kenyon College,
Ohio (quite a distance from
Kelvin Way Bridge, Glasgow).
More power
from the
High-End Power Amp
those of the Editor or Publisher.
•Please send your MailBox
correspondence to:
editor@elektor-electronics.co.uk
or
Elektor Electronics, The Editor,
1000 Great West Road,
Brentford TW8 9HH, England.
Our audio design specialist Ton
Giesberts confirms that his High-End
Power Amplifier design has potential
for higher output power. The supply
voltage may be increased to 35 V
maximum, but not without major
surgery to the existing design. For
example, a larger heatsink must be
used on the driver stage, and the
resistor with the relay (R45) has to
be adapted, as you have done. The
sensitivity also requires adapting
— it is now fairly low at 1.5 V for
full drive. We confirm that it can
be done by using 10 kΩ for R18,
but stress that the modification
modifies the carefully designed
feedback response, which is likely
to result in instability. This part of the
modification really calls for a redesign. We recommend the use of an
oscilloscope and a protected power
supply if you want to stay on top of any
tendencey to oscillation.
The OPA177 has a maximum supply voltage
spec of 22 V, hence is hard pushed in the
original design already. Zener diodes of at
least 12 V (D14; D15) are recommended at
the suggested supply level of 31 VDC. At
a supply voltage of 35 V, the zener diodes
should be exchanged for 15-V types.
Regarding the output power, at a supply
voltage of 31 V, about 50 watts can be
delivered into 8 ohms, while the minimum
load impedance goes up to about 3 ohms
to keep the power transistors within their
safe operating area.
info & market
news & new products
Single-cell 1-A Li-Ion / Li-polymer charge management controllers
Microchip announces the
MCP73833 and MCP73834 single-cell, high current (1 amp), LiIon/Li-Polymer linear charge-management controllers. These fully
integrated devices combine several key charge-management and
safety features in a single chip for
reliable charging of high-capacity single-cell Li-Ion and Li-Polymer
batteries.
By including a pass transistor, current-sense and reverse-discharge
protection on a single chip, the
MCP73833/4 charge-management controllers eliminate the need
for external components. Multiple
combinations of key charging parameters, including pre-conditioning current threshold and ratio,
charge-termination threshold and
recharge threshold ratio are available, meaning the devices provide
standard product support for a variety of high-current Li-Ion/Li-Polymer charging applications. In addition, with a high charging current of up to 1 A and support for
multiple regulated output voltages
(4.2 V, 4.35 V, 4.4 V and 4.5 V),
the devices can be used with various generations of Lithium battery
technology.
Safety features on the new devices to prevent overcharging and
overheating include charge timers, battery-temperature feedback
and thermal-current regulation. The
charge timer shuts the charger off
if a charge is not terminated before timeout is reached. The battery-temperature feedback reduces
the charge current if the battery’s
temperature reaches the limit of
safety and the thermal-current regulation feature decreases the charge
current if the charge-controller itself
reaches its thermal limits.
Device-specific features include
a power-good output on the
MCP73833 and a timer-enable
input on the MCP73834. Both
devices also offer a low dropout
regulator (LDO) test mode that enables application system test even
in the absence of a battery; and
both feature two status outputs to
provide the user with additional
information about the state of the
charge-controller.
To support development, Microchip
offers the MCP73833 Evaluation
Board (Part # MCP73833EV). The
board is available today at www.
microchipdirect.com.
The MCP73833/4 charge-management controllers are available in 10-pin MSOP and thermally
efficient 3 x 3 mm DFN packages.
They are available for sampling at
sample.microchip.com and for volume ordering at www.microchipdirect.com.
For more information, visit Microchip’s
website at
www.microchip.com/MCP73833.
(067227-VII)
Are you CO sure?
If you have a Carbon Monoxide
(CO) concern and need to know
more, Lascar Electronics’ EL-USBCO carbon monoxide data logger
could help in determining the nature of the problem.
Carbon Monoxide (CO) is a poisonous gas which is both odourless and colourless. It is produced
by equipment/machinery that
isn’t working correctly and can
be found anywhere from construction sites and furnace rooms to office blocks and homes. Lascar’s
EL-USB-CO data logger monitors
and records CO levels in an environment over a period of time.
This can help the user to determine where and when peak levels
of CO occur, allowing corrective
action to be carried out to remedy
the problem.
The EL-USB-CO stores over 32,000
readings and can record CO levels from 0 to 1000 ppm. Setup of
the data logger is completed using the supplied EL-USB software,
with the EL-USB-CO plugging directly into the USB port of a PC.
Here the user can assign the logger a name, choose a sample rate
(from a choice of once every: 10
secs, 30 secs, 1 mins, 5 mins), as
well as determining a high-alarm
level. Once setup is complete the
EL-USB-CO should be left in the
environment where the study is to
take place.
The EL-USB-CO is available for pur-
uring 35mm x 38mm x 2.7mm on
a single-sided PCB.
The DAB signal processing functions and protocol stack are implemented in firmware running on the
Chorus processor, which also runs
the control interface to Naples. In
a master configuration the module
requires a power source, antenna,
LCD and keypad to create a fully
featured digital radio. Alternative-
ly, the module can be controlled
by an existing microcontroller as
a slave module via a serial port
or SCB (serial control bus) compliant device allowing it to be integrated into larger audio systems.
The module also supports various
software features such as DAB dynamic DLS radio service text, 256
kbps decode capacity, stored presets and manual tuning when con-
chase at £49.00 from the Lascar
website (www.lascarelectronics.
com).
(067227-III)
DAB/FM radio module
Frontier Silicon announces the
launch of Naples FS2011, an integrated standalone dual-band
DAB/FM radio module. The unit
is a complete DAB module operating in both master and slave
modes and incorporating Frontier
Silicon’s Apollo RF front-end, Chorus DAB baseband processor and
NXP Semiconductor’s TEA 5764
FM radio IC. The module is meas-
10
figured in system applications.
www.frontier-silicon.com
(067227-I)
elektor electronics - 12/2006
Mid-power 24-Vin maxi modules
Vicor announces the addition of
eight mid-power Maxi DC-DC converters to the 24 Vdc input family:
a 3.3-Vout, 200-W model and 300W models at 5, 12, 15, 24, 28,
36, and 48-Vout. The modules —
which incorporate Vicor’s patented
low-noise Zero-Current and ZeroVoltage Switching (ZCS/ZVS) topology — are appropriate for industrial or process control, distributed
power, medical, ATE, communications, defence, and aerospace applications. With switching frequencies up to 1MHz, the 24Vdc family
provides rapid transient response
well suited for RF applications.
The new products provide design-
ers who do not need the full-power
capability of a 24V Maxi module
with a mid-power option, with all of
the functionality and configurability
of the high power models. In addition, low-noise ZCS/ZVS greatly
reduces the design effort and filtering costs required for power converters to meet agency conducted
emissions requirements.
The modules, which are available
in RoHS compliant models, are a
compact 117 x 56 x 12.7 mm in
size, with a height above board
of 10.9 mm.
With these new models, the 24 Vin
Maxi family now comprises 16
models with output voltages from
3.3 to 48 Vdc and power levels
from 200 to 400 W. The converters operate from 24 V nominal input, with an input range of 18 V
to 36 V. Efficiencies range up to
88% for the higher output voltages. These models are available in
five different environmental grades,
with six different pin options and
three choices of baseplate. They
can be configured in any combination in Vicor’s Custom Module
Design System.
A datasheet is available on:
www.vicorpower.com/library/
technical_documentation/
datasheets/2nd_gen/.
www.vicoreurope.com
(067227-V)
8-bit microcontrollers with integrated Ethernet peripheral
Microchip announces a family
of the world’s smallest 8-bit microcontrollers with an integrated
IEEE 802.3-compliant Ethernet
communications peripheral. The
PIC18F97J60 family is optimized
for embedded applications, and
has an on-chip Medium Access
Controller (MAC) and Physical
Layer Device (PHY).
By integrating a 10BASE-T Ethernet
controller onto a 10 MIPS PIC18
microcontroller with up to 128
kBytes of Flash program memory,
Microchip is providing embedded
systems designers with a simple,
cost-effective single-chip remotecommunication solution for a wide
range of applications. Microchip
also offers a free TCP/IP software
stack to reduce development time.
Ethernet is the leading networking
technology for local area networks
(LANs), and it can be used to connect embedded devices through a
LAN to the Internet. Ethernet’s infrastructure, performance, interoperability, scalability and ease of
development have made it a standard choice for such embedded
communications.
Any embedded application that
requires Ethernet connectivity can
take advantage of the new ninemember PIC18F97J60 microcontroller family. Such applications
can include Industrial Automation
(e.g. industrial control, power-supply monitoring, network/server
12/2006 - elektor electronics
ernet to existing PIC18 designs
with minimal cost and development time.
IEEE 802.3-Compliant: on-chip
10BASE-T MAC and PHY provide
reliable packet-data transmission
and reception.
Dedicated 8-kByte Ethernet Buffer: enables efficient packet storage, retrieval and modification,
and reduces the demand on the
integrated microcontroller.
128 kBytes of Flash and 4
Kbytes SRAM: to accommodate
the TCP/IP stack and Web server,
leaving ample space for application code.
•
•
•
monitoring and environmental
monitoring); Building Automation
(e.g. fire & safety, access control,
security panels, lighting control
and VoIP intercoms).
Key features of the new family
include:
Seamless Migration: add Eth-
•
The PIC18F97J60 PICDEM.net™
2 Development Board (part #
DM163024) has been created
specifically to assist development
with these new integrated devices.
In addition, the latest version of Microchip’s free PIC18 TCP/IP Ethernet Stack can be downloaded at
www.microchip.com/tcpip . The
new family is also supported by
Microchip’s suite of development
tools, including the MPLAB® VDI
Visual Device Initializer, Application Maestro™ software, MPLAB
C18 C compiler and the MPLAB
ICD 2 in-circuit debugger.
The new PIC devices are all offered in RoHS-compliant TQFP
packages.
www.microchip.com/ethernet
(067227-II)
11
Special Christmas Offer
3
Elektor Books/CD-ROMs
for just £ 27.50 | $ 52.50
Offer valid untill 1 January 2007
PC Interfaces
under Windows
Excl. P&P
Build your own High-end
Audio Equipment
Designing Audio Circuits
How does speech, music, or, indeed, any
sound get from the record, CD, or cassette tape to the loudspeaker?
This book endeavours to give
a comprehensible answer.
For those who cannot, or
will not, pay high prices
for high end equipment,
a solution is offered in this
book: build your own at
considerable cost savings.
PC Interfaces can be used for more
than just the printer, mouse, modem
and joystick! While it was relatively
easy to directly access PC interfaces
using a DOS computer, under
Windows things are not all that
350 pages
simple. This book (CD-ROM incl.)
£ 20.75 | US$ 42.00
shows you how it can be done.
262 pages
£ 15.55 | US$ 31.00
265 pages
£ 25.95
US$ 52.00
Lasers: Theory and Practice
Bestseller
Handbook for Sound
Technicians
This book contains chapters on
basic theory; microphones and
musical instruments; various types
of amplifier; loudspeakers; effects
equipment; recording techniques;
lighting equipment; the rehearsal
room; and faultfinding and small
repairs.
276 pages
£ 20.75
US$ 42.00
Save £££’s
A valuable book on the practical use of lasers.
It consists of two main parts. The first deals
with the fundamentals of lasers, including
such topics as types of laser,
modes of laser operation,
wavelengths, chopping,
scanning and applications.
The second part contains a
number of practical circuits
and experiments.
Faultfinding
in Computers
and Digital
Circuits
This book covers faultfinding not just in
microprocessor systems, microcontrollers
and industrial PCs, but also in consumer
items such as personal computers, multimedia devices, digital television and so on.
180 pages
£ 20.75 | US$ 42.00
PC Service and Repair
This book provides the information
you need to be able to deal with
computer system faults whenever
they occur. With the aid of this book,
you can tackle faultfinding at
various levels, ranging from the
replacement of complete cards
or assemblies to the identification of a single faulty component.
479 pages
£ 31.15 | US$ 63.00
make your
choice
CD-ROM
625 pages | £ 31.15 | US$ 63.00
Audio Collection 1 + 2
Two must-haves for the true audio lover.
Each CD-ROM contains no fewer than 75
audio designs from Elektor Electronics. Using
the included Acrobat Reader you are able to
browse the articles on your computer, as well
as print texts, circuit diagrams and PCB layouts.
£ 12.05
US$ 21.25
CD-ROM
Software Toolbox 2
CD-ROM
This CD-ROM contains software tools for, and
information about, microcontrollers. Toolbox 2
gives specific attention to technical documentation about protocols, field buses, as well
as modern information carriers which find
increasing use. The hardware side of things is
not forgotten either.
Digital Circuit Library 1-2-3
Each CD-ROM contains over 300 circuits complete
with diagram and text. Zoom and print module.
Fast search facility. Error-tolerant index search system.
Many printed-circuit board layouts.
£ 12.05 | US$ 21.25
£ 12.05 | US$ 21.25
CD-ROM
Robotics
A large collection of datasheets,
software tools, tips en tricks,
addresses, Internet links to
assorted robot constructions
and general technical information. All aspects of modern
robotics are covered, from
sensors to motors, mechanical
parts to microcontrollers, not
forgetting matching programming tools and libraries for
signal processing.
CD-ROM
ECD Edition 3
Elektor’s Components Database
gives you easy access to design
data for over 5,000 ICs, more
than 35,000 transistors, FETs,
thyristors and triacs, just under
25,000 diodes and 1,800 optocouplers. All databank applications are fully interactive,
allowing the user to add, edit
and complete component data.
£ 14.95 | US$ 26.50
£ 12.05
US$ 21.25
CD-ROM Elektor Electronics
2001 + 2004
These Elektor Electronics annual
CD-ROMs contain all editorial
articles published in Elektor
Electronics magazine volume 2001
and 2004.
The CD-ROMs are packed with
features including a powerful
search engine and the possibility
to edit PCB layouts with a graphics
program, or printing hard copy at
printer resolution.
Offer valid
untill
1 January 2007
£ 16.25 | US$ 28.75
No Order Form here?
No problem, you can still benefit from our offer.
Please contact us:
Elektor Electronics (Publishing)
Regus Brentford | 1000 Great West Road
Brentford TW8 9HH | United Kingdom
Telephone +44 208 261 4509 | Fax +44 208 261 4447
Email: sales@elektor.com
Limited
Stock
info & market
gadgets & musthaves
Christmas Presents
For the electronics hobbyist who has everything already
Do you still need help choosing gifts for others (or for yourself)? We’d like to
lend you a hand with a variegated selection of interesting, fun, handy, gadgetlike or simply unusual gifts. It also includes several items that people with no
special interest in electronics will enjoy.
Hovering football
Time on the fly
The XP3 clock is a programmable
digital clock that displays the time
and messages hovering in the air. A
set of LEDs fitted to a rod that swings
back and forth at a tidy 16-Hz rate
flash on and off at just the right
times to create visible messages.
Each message appears in a different
form. You can program four individual messages, each with a maximum of 200 characters. They can
be combined with the date and time
display in all sorts of ways. There are also several preprogrammed messages, in English of course.
Seen at: www.gadgethouse.nl
We must admit that this ‘hovering
football’ isn’t actually round. However, it’s a nice alternative to the real
thing for a spot of football in the
living room without doing too much
damage to the furniture. This is a
genuine hovercraft in the form of a
flattened ball. It floats on a cushion
of air blown out at the bottom, so it can hover over the
floor and move easily. It doesn’t have any electronics – just
a battery-powered motor.
Seen at: http://smm.de
T-shirt with graphic analyser
You’re bound to attract attention if you wear this T-shirt to a disco or café. An illuminated panel in the
form of a graphic analyser is fitted into the front of the T-shirt. The accompanying microphone and electronics
pick up ambient sound and drive the bars of the graphic display accordingly. The analyser is powered by a detachable battery pack. The T-shirt is available in two sizes. Unfortunately, it can only be washed by hand, so you
have to be prepared to handle it with TLC.
Seen at: www.bestel.nl
Solar-powered headphone radio
With this set of headphones, you have music with you wherever you
go – all without batteries. An FM receiver is built into the ear shells,
and it is powered by a set of rechargeable AAA cells. The cells are
charged by a solar panel fitted on top of the headband. One hour
of sunlight is enough for
1 to 3 hours of pleasant
listening.
The headband, which also
includes an integrated antenna, can easily be adjusted to fit any head size.
Seen at: www.paramountzone.com
14
Mini drum set
Who doesn’t occasionally
get the urge to have a go
at being a drummer? But
how many of us have a
drum set at home? And
of course, you have to remember the neighbours…
Now there’s a solution for
the weekend drummer: an
affordable electronic mini
drum set. Several drums,
and even a cymbal, are located on a surface with the dimensions of a mouse pad. You can drum on these instruments with
your fingers. Of course, it takes a bit of practice to get the hang
of it. There are also several control knobs behind the drums.
This way you don’t have to worry about complaints from the
neighbours – as long as you can resist the temptation to connect the drum set to your stereo system!
Seen at: www.megagadgets.nl
elektor electronics - 12/2006
Flexible keyboard
Talking toilet
paper holder
This gadget is just the
thing for surprising
your friends and acquaintances. It generates a spoken message
each time some pulls
toilet paper off the roll. You can record your own message,
either serious or humorous. For instance, you could use it to
remind your children to wash their hands when they’re finished, or to remind adults to close the lid before they leave.
As for humorous messages, that’s something we’ll leave up
to you! The toilet roll holder has a spring-loaded middle piece, so it can be used with nearly every standard holder.
Seen at: www.bestel.nl
With this unusual keyboard, which is not only flexible and waterproof
but also features trendy blue illumination, you’re immune to just about
everything. It simply shrugs off coffee,
bread crumbs and tobacco residues.
An occasional rinse under the tap is
enough to keep it looking as good as
new. It will last for years if you take
good care of it, and it’s great with
a laptop – just roll it up and pack it
away.
Seen at: www.usbgeek.com
Remote-controlled golf ball
Just when you thought you’d seen it all! You
may wonder how anyone came up with the
idea of a remotely controlled golf ball, but it’s
mainly intended as a sort of practical joke.
Suppose you’re golfing with a friend, and he
suddenly sees his golf ball making strange
lurching motions. You know the answer: you
swapped a remotely controlled ball for his
real ball.
You can use the remote control to cause the
ball to swing to the right or left while it’s rolling. Just the thing
for golfers with a
healthy sense of
humour!
Seen at: www.
iwantoneofthose.com
Mini UFO
If you always wanted to pilot a
UFO, here’s your chance. The XUFO consists of a thin frame with four propellers, along with the
control and receiver electronics you need to operate it. It is gyroscopically stabilised in flight.
The X-UFO is made from ultra-lightweight carbon fibre and EPP
foam. Four LEDs (one red and three blue) not only help you keep
track of the X-UFO, but also give it an unworldly appearance.
The X-UFO comes with a four-channel proportional R/C unit.
Seen at: www.gadgethouse.nl
Illuminated toilet seat
Are you always irritated by the dim
lighting in the WC when you have
to use it in the wee hours? And does
you wife always complain that you
leave the toilet seat up after you use
it at night?
The intelligent LavNav toilet light puts
an end to all these problems. After
you fit this tiny light to the bottom of
the toilet lid, it automatically illuminates the toilet bowl discreetly when
you approach the toilet. And so you don’t forget whether the toilet seat
is raised or lowered, the lamp has two different colours. green means
it’s OK to sit down, while red means watch out, the seat’s still up!
Seen at: www.gadgets.co.uk
Colourful loudspeakers
The Lightwave loudspeakers are small spherical loudspeakers made from transparent plastic that
change colour in rhythm with the music. They also have beat detection.
Listeners can choose from three different colour patterns or a specific constant colour, or they can
let the colours respond to the music.
The speakers are approximately 10 cm in diameter, and the built-in amplifier delivers 5 W
PMPO. They are an ideal complement to an MP3 player or a portable CD player.
Seen at: www.gadgets.uk
12/2006 - elektor electronics
15
info & market
gadgets & musthaves
Shake those numbers!
USB rocket launcher
You’re probably already familiar with pocket torches
that you shake to generate the necessary energy.
The same principle has now been applied
to pocket calculators. The little calculator
shown here has a small tube at the top
with a magnet inside that can move
back and forth. When you shake the
calculator, a coil surrounding the tube
generates enough electrical energy from the
moving magnet to power the calculator for a
short while. That’s naturally something else than a
calculator that runs on solar cells or uses water as a
source of energy.
Seen at: www.gadgets.co.uk
This article is full of all sorts of
gadgets with USB interfaces, but
this is really the best of the lot.
This miniature rocket launcher
has a rocket holder with three
foam-rubber rockets. You can
use the included software (Mac
and PC versions) to aim the rockets horizontally and vertically
and then fire them. The propulsive force is provided by several
springs in the rocket holder, and
the range is approximately 3
metres.
This rocket launcher has become so popular that
hackers have already developed modified software for it. Several successor models have also
been sighted already.
Seen at: www.gadgets.co.uk
Hovering globe
It’s still something special to see
a metal globe hovering in the air
thanks to a magnetic field. Here a
bit of electronics and a coil are used
to attract the globe just enough to
keep it hovering in the air. Various
models are available. Last year we
had a very modern one, and this
time we chose an ‘antique’ model
with a nice 20-cm globe. The base
of the copper-coloured frame houses a full-fledged microcontroller that adjusts the magnetic field 16,000 times per second.
Seen at: www.gadgets.co.uk
Roll-up keyboard
This keyboard (the musical kind) has 49 keys and a USB port (which
is where it draws its operating power) along with another convenient
feature: your can roll it up. That’s something keyboard players who are
familiar with normal ‘hardcase’ keyboards will certainly appreciate.
Especially
since this flexible USB keyboard can hold its own
against its space-gobbling cousins. The keyboard features eight percussion instruments,
demo songs, vibrato and other effects, a
metronome, and much more. You can
also compose your own rhythms if
you want. The beat can be adjusted
from a sedate 40 beats per minute
to a nerve-wracking 208.
Seen at: www.usbgeek.com
USB slippers
If you often sit in front of your computer until the wee
hours, it can get pretty uncomfortable sometimes, especially
during the cold season. However, you can avoid cold feet with
these USB-powered electric slippers. These fluffy slippers can be
connected to any computer with a USB port (PC or Mac) or even a
game console. The heating element is washable, and it warms the
slippers to a maximum temperature of 48 °C. Now that’s comfort!
One caution: make sure your computer has enough USB ports
and a hefty power supply, since otherwise it probably won’t to be
able to handle all these USB gadgets.
Seen at: www.usbgeek.com
Electronic Lederhose
This pair of electronic Lederhose (leather shorts) comes from the
Bavarian clothing manufacturer Lodenfrey. This traditional south-German garment is fitted with an MP3 player with 512 MB
of memory and a built-in (or should
we say sewn-in) control panel. It also
has a handsfree function for your
mobile phone. Now you can slap
your thighs, listen to music and make
phone calls all at the same time.
Seen at: www.lodenfrey.de
16
High-tech ballpoint pen
This is something we all need! Now
you don’t have to chew on your pencil
during a drawn-out press conference.
Instead, you can listen to relaxing
music with this high-tech MP3 ballpoint pen. Thanks to its generous
storage capacity of 512 MB or 1 GB,
the pen is also suitable for long meetings. After you’ve listened to all the
music, you can simply switch on the
built-in FM radio. If somebody happens to say something interesting,
the built-in microphone will pick it
up nicely for you, so
you don’t have to miss
anything. The rechargeable lithium-ion
battery is good for 7
hours of operation.
Seen at: www.usbgeek.com
elektor electronics - 12/2006
Pleasant scents
You know the type: glued to the computer day
and night while churning out code at the rate
of several pages a minute. This talent is often
accompanied by sleep deficiency, personal
hygiene that leaves something to be desired,
and a penetrating musty odour. To help camouflage the aroma of canned cola and
ambulant pizza leftovers, we recommend this
USB-powered aroma dispenser as a suitable
gift. Citronella and anise, camphor and orange-peel
oil are mind-expanding, so they help with debugging.
Seen at: www.usbgeek.com
Heated gloves
Suppose you’re sitting in the Trans-Siberia Express on
your way to an important presentation on drilling for
oil. Just when you want to run through your PowerPoint
presentation again, the train’s heating system fails. No
problem – you simply slip on these USB-powered heated
gloves and carry on. Incidentally, they match nicely with
the USB slippers described in this article. However, in this
case we recommend that you
at least purchase a reserve battery, since otherwise you could
easily find yourself sitting with
cold feet, cold hands, and a
blank screen.
Seen at: www.usbgeek.com
Overclock your
brain
Overclocking CPUs is old
news by now – the new rage
is overclocking the grey matter between your ears. This
gadget provides especially intensive training
of your grey nerve cells. With this device and
a bit of practice, you can significantly boost
the ‘clock frequency’ of your brain. It trains
your assimilation speed. We’ve heard that our
boss has already ordered a thousand or so.
Seen at: http://shop.elv.de
Water-powered calculator
Following in the steps of pocket calculators powered by
AC adapters, batteries and solar cells comes the most
ecologically responsible model yet. It doesn’t run on alcohol, but instead on water, which explains its name: H2O.
It goes for several weeks on just a few drops. When it
doesn’t want to work any more, just fill it up under the tap
and you’re all set for several more weeks. Handy, ecological, and economical – and above all a lot of fun!
Seen at: www.ledindon.com
Binary clock
Seen enough of those omnipresent LCD
clocks that show the time in hours and
minutes with painstaking precision?.
How about a clock that displays the time
using blue LEDs, and what’s more in
binary form? That makes checking the
time a mental exercise, since the six columns correspond to hours, minutes and
seconds, each in the form of tens and
units. The time displayed by the clock in
the photo is 12 hours, 1 minute and 47
seconds. The LED brightness can be set
to three different levels. Of course, as an
electronics whiz you’ll master this new
way of telling time in only a couple of
minutes.
Seen at: http://www.bestel.nl
Cubite speaker and USB hub
We’d be lost without USB in this world, and USB
hubs are at least equally indispensable. The Cubite Speaker USB hub is a stylishly fashioned box
that houses not only an excellent speaker for your
PC, but also a USB hub with ports for your webcam, MP3 player, memory stick and digital camera. The speaker has a large volume knob and two
smaller knobs to adjust the treble and bass. It is
powered from the USB port, so it doesn’t need an
external power supply (or any additional software).
Seen at: www.iwantoneofthose.com
H Racer: a hydrogen-powered car
The first car fully powered by hydrogen: that’s the technology of the future! Unfortunately, this model is rather small,
but the 21st century is still young. The H Racer is a fullfledged demonstration of a hydrogen propulsion system. A
solar cell supplies the energy to produce the hydrogen (H),
and the transparent housing lets you easily see what happens: blue LEDs illuminate the tiny bubbles of oxygen (O2)
that are expelled from the water reservoir.
Seen at: www.latestbuy.com.au
12/2006 - elektor electronics
17
info & market
enclosures
Tightly Packed
Enclosures and front panels
Thijs Beckers
There are an incredible number of options these days for the safe ‘packing up’ of electronic
circuits. Open any catalogue from any well-known mail-order company and you will get an
impression of the extensive range on offer. This article gives an overview of the different
types with their particular characteristics and provides hints as to how you can make a
professional looking front panel yourself.
From an economic and marketing perspective, the enclosure, including the front panel, of commercial equipment
is very important for the manufacturer. Equipment that
does not look attractive will sell poorly, of course, and
is nearly impossible to extol its virtues in advertisements.
So it is logical that much time and effort is spent on the
design of these enclosures. An additional consideration
is the ergonomics of equipment that has many operating
controls. The design of the enclosure is then already taken
into account during the development of the circuit.
This is usually not the case for prototypes, small (handmade) production runs and home-built circuits. Of course,
the marketing aspect is not a consideration here either.
Specific characteristics, such as extra heavy-duty waterproof and explosion resistant boxes, are a little bit over
the top for the average home project. A standard box is
usually good enough. But that does not distract from the
fact that appearance and function certainly also do play
a role in your own circuits. With a little bit of searching
for a nice and appropriate enclosure and a little bit of ef-
18
fort for a front panel you can definitely make a nice looking piece of equipment that would not look out of place
when displayed in the average electronics shop window.
With a clean design and your own logo on the front panel it will look like the real thing.
Making a choice
For prototypes you usually choose from a few types of
enclosures that may optionally conform to some standard, (see inset Industrial Enclosures). For each design you
will look at what type of enclosure suits best. There are a
number of different types of enclosures that may be categorised as follows:
1. 19-inch enclosures, which can be easily built into, or
removed from, a standard rack or industrial box.
2. Enclosures that are deliberately sized for the common
‘Euro format’ PCB.
elektor electronics - 12/2006
3. Console enclosures with a tailored front panel or (sloping) top for control or mixing panels.
4. Small enclosures with a built-in mains plug, which
plugs into a power point just like a mains power adapter.
5. Enclosures for handheld applications, such as
multimeters.
6. Enclosures for DIN rail systems, which are commonly
used in an industrial environment.
A few of the factors that are important for the type of
enclosure are of course the size, safety, construction material, method of mounting and – mainly important in industrial applications – the NEMA and/or IP classification [1].
The size obviously depends on the components that have
to fit in the enclosure, together with the connection and
mounting options, internal and external access, thermal
conditions and potential future extensions. The material
selected needs to be able to withstand the conditions that
the enclosure will be subjected to at the location at which
it will be used. Considerations are corrosion resistance
and rigidity requirements. The NEMA and/or IP classification is an industrial standard for the protective characteristics of an enclosure. Tables 1 and 2 in the inset Industrial Enclosures give an overview of these two common
protection classifications.
Machining
The internal electronics will, in all likelihood, be connected to the outside world via cables and plugs. This
requires a number of plugs and contact points, which usually requires some machining of the enclosure.
In addition, the operating buttons need to be given a
logical position, generally on the front panel. The front
panel usually demands the most attention. To make an
attractive aluminium front panel, complete with labelling, requires a fair amount of equipment (drills, routers,
screen printer, sander, etc.) and the process is not easy.
Also, if you only need to do these things every once in a
while it is not worthwhile to invest in the required equipment. There are a number of companies that specialise
in the manufacture of front panels, such as the German
company Schaeffer [2], the American Internet company
eMachineShop [3], the English company CTL-Components [4], the international company Elma [5] and the
Dutch company Antronics [6]. Schaeffer and eMachineShop even offer their own (free!) software, which makes
it easy to draw your own design. This design can then
be sent to the manufacturer who will then machine and
12/2006 - elektor electronics
screen print the front panel. In this way you can obtain a
professional looking front panel.
Do it yourself
The specialist equipment that is required for this machining has its price, of course. This is often clearly noticed
from the amount of money you have hand over for a custom manufactured front panel. A cheaper solution is the
self-adhesive, transparent film that can be printed with a
laser printer. This film gives only limited protection from
scratches and gives a somewhat dull result. The latter can
be improved with some plastic spray.
In addition to the special film for laser printers there is
also the overhead transparency for inkjet printers. This
Case manufacturers
Manufacturer
Website
ABB
APW
Bernstein
Bopla
Boss
Box
Cooper
Deltron Emcon
Dold
Erni
Eurobox
Fibox
Himel
Hammond Manufactoring
Lawtronics
Monacor
Moeller
OKW
Pactec
Retex
Rittal
Rolec
ROSE Systemtechnik
Sarel
Schroff
Serpac
Spelsberg
TEKO
VERO
Weidmüller
www.abb.nl
www.apw.com
www.bernstein-ag.de
www.bopla.de
www.boss-enclosures.co.uk
www.boxenclosures.com
www.b-line.com
www.deltron-emcon.com
www.dold.com
www.erni.com
www.euroboxenclosures.com
www.fibox.nl
www.himelenclosures.com
www.hammfg.com
www.lawtronics.co.uk
www.monacor.nl
benelux.moeller.net/nl
www.okw.com
www.pactecenclosures.com
www.retex.es
www.rittal.nl
www.rolec.de
www.rose-pw.de
www.sarel.nl
www.schroff.co.uk
www.serpac.com
www.spelsberg.nl
www.tekoenclosures.com
www.vero-electronics.com
www.weidmuller.nl
19
info & market
enclosures
Table 1: Description degrees of protection
to DIN EN 60529 (VDE 0470) (IP-type)
Degrees of protection for people
and solid objects (first number)
Degrees of protection against water
(second number)
0
No protection
No protection
1
Protection against solid objects greater than 50 mm
diameter
Vertically falling drops of water do not cause damage
2
Protection against solid objects greater than 12 mm
diameter
Drops of water with up to 15° from vertical do not
cause any damage
3
Protection against solid objects greater than 2.5 mm
diameter
Drops of water with up to 60° from vertical do not
cause any damage
4
Protection against solid objects greater than 1 mm
diameter
Splashing water from any direction does not cause
any damage.
5
Completely protected against accidental touch, partially against dust
Low pressure water jets from any direction do not
cause any damage
6
Completely protected against accidental touch and
against dust
Strong jets of water do not cause any damage
7
-
Water does not cause any damage if the enclosure is
submerged by 0.15-1 m
8
-
Water does not cause any damage if the enclosure is
submerged by a specified amount
9
-
Water under high pressure from any angle does not
cause any damage
Number
Table 2: NEMA standard for enclosures
20
NEMA
standard
Description
NEMA 1
Enclosures constructed for indoor use to provide a degree of protection to personnel against incidental contact
with the enclosed equipment and to provide a degree of protection against falling dirt.
NEMA 2
As NEMA 1, and to provide a degree of protection against dripping and light splashing of liquids.
NEMA 3
Enclosures constructed for either indoor or outdoor use to provide a degree of protection to personnel
against incidental contact with the enclosed equipment; to provide a degree of protection against falling dirt,
rain, sleet, snow, and windblown dust; and that will be undamaged by the external formation of ice on the
enclosure.
NEMA 3R
As NEMA 3, except no protection against dust
NEMA 3S
As NEMA 3, and in which the external mechanism(s) remain operable when ice laden
NEMA 4
As NEMA 3, and hose-directed water and that will be undamaged by the external formation of ice on the
enclosure
NEMA 4X
As NEMA 4, and protected against corrosion
NEMA 5
As NEMA 2, but protection against airborne dust, lint, fibers
NEMA 6
As NEMA 4, and protected against water during occasional temporary submersion at a limited depth
NEMA 6P
As NEMA 6, but protected against water during prolonged submersion at a limited depth
NEMA 7
Enclosure for indoor use at locations specified as Class I, Group A, B, C and D (refer National Electrical Code,
NEC) withstands pressure caused by internal explosion and prevents the ignition if explosive gasses. Also internal heat does not cause danger to the environment
NEMA 8
As NEMA 7, enclosure submerged in oil
NEMA 9
Enclosures for indoor use at locations specified as Class II, Group E, F en G (refer NEC). Protected against
dust, internal
NEMA 10
over-heating does not create an explosion hazard for surrounding gasses
NEMA 11
Suitable for use in corrosive environments, enclosure submerged in oil
NEMA 12
As NEMA 5, for industrial use
NEMA 12K
As NEMA 12, with knock-outs
NEMA 13
As NEMA 5, also protected against oil and non-corroding cooling fluids
elektor electronics - 12/2006
film is printed on the reverse side. First stick clear, doublesided adhesive foil (available in stationery shops) to the
front panel and then the mirror-printed film. In this way
the ink is also protected against scratches. In colour and
using at least 300 dpi it will also look good and is not
expensive.
Before sticking the film to the front panel, you first have
to make the holes in it. Tip: it is better to deburr the front
side of small holes once the film is in place. The opening
in the film will then have the correct diameter at the same
time. You can easily cut square or large openings in the
film with a sharp knife.
On (ugly) plastic front panels you can use the standard
sticker sheets for the front panel layout. Or alternatively,
you can use standard paper and the above-mentioned
double-sided adhesive foil. The colour of the paper
will obviously determine the colour of the front of the
enclosure.
Tip: a ‘membrane keyboard’ can be made quite easily
with an additional printed film between the front panel
and the adhesive foil. Behind the printed keyboard is an
opening in the front panel that could contain miniature
pushbuttons. The front panel can be made from cheap
materials, of course.
Drilling
To drill small round holes, twist drills are obviously the
best choice. A good drill stand is very handy and a drill
press is ideal. If you mark the locations of the holes with a
centre punch then the tip of the drill will not wander if you
are drilling free-hand. Large, round holes are best made
by first drilling a smaller hole and then enlarging it with a
reamer.
Rectangular openings are a little harder. With simple tools
it is best to proceed as follows: drill, within the cutout, a
hole that’s large enough for the saw blade of a hacksaw.
Now cut the material to within about 0.5 to 1 mm of the
desired opening. Remove the last little bit with a file.
If you are using thin sheet for the enclosure it is better to
use a jigsaw for this, instead of a hacksaw. Sandwich
the thin sheet with thin plywood or something similar, that
you cut at the same time. In this way the thin sheet will not
bend. Leave the plywood in place also when filing.
For very narrow openings (for slide potentiometers, for
example), drill a number of small holes next to each other
(with a good drill press and a short drill these holes can
even overlap). With a little key file you can then remove
the remaining unwanted material.
A nice result
Both in industry as well as for your own applications, the
purpose of the enclosure is mainly to protect the electronics components from permanent damage caused by
moisture, chemicals and dust. At the same time it provides
a protective function for the environment. This can be protection to prevent the accidental touching of hazardous
voltages but also for EMI.
A second good reason to package a circuit is to make it
look more attractive. The average user is not fascinated
with a collection of components on a circuit board. A
good front panel layout will also make it immediately obvious what each operating button does and how to use
the equipment.
In industry there is often great emphasis on the physical
appearance. That is why specifically designed enclosures
are often used, that are evaluated at an early stage of the
Industrial enclosures
The material from which an enclosure is made determines its physical properties for the most part. Plastic is usually chosen for
portable applications, because of its weight and wear resistance. In other applications other factors such as mechanical strength,
electrical and thermal resistance and fire resistance are more important. Tables 1 and 2 show the protection classification that
an enclosure can offer for two common standards.
To provide protection against EMI, the enclosures can be coated with, for example, carbon, aluminium or copper. This is very important, in particular for sensitive electronics, if the circuit is to operate without problems. Up to 1 MHz you can use electronic filters and metal screens. Above that frequency a decent Faraday cage is required. To also keep out magnetic fields, an aluminium
or steel sheet enclosure is inadequate. This application requires special µ-metal.
A number of common materials for plastic enclosures are polycarbonate, polystyrene, polypropylene and ABS. Some enclosures
are transparent for IR. This is very handy for remote controls, for example. And there are many more special properties along
these lines.
If we select a metal enclosure, we can make a distinction between aluminium, galvanised steel sheet, stainless steel and a
number of other materials. A metal enclosure has the advantage that it will largely block electric fields, of course.
12/2006 - elektor electronics
21
info & market
enclosures
overall design. The initial costs of a newly designed enclosure may be quite high, but large numbers reduce the
price per enclosure considerable. Buying large numbers
of standard enclosures would be much more expensive
in the end, because the various holes still need to be
machined.
If you are working with prototypes then the considerations are completely different of course. It is often the case
that the circuit is already finished and you only need an
enclosure that will fit everything. With the directions given
above, it is entirely possible to make a front panel that
does not have to look inferior to that of a professional
piece of equipment.
(060298-I)
All photographs: Conrad Electronics.
Weblinks:
[1] www.nema.org
[2] www.schaeffer-ag.de
[3] www.emachineshop.com
[4] www.ctl-components.com
[5] www.elma.de
[6] www.supermoduul.nl
GLOSSARY
ABS: Acrylonitrile Butadiene-Styrene. Rigid man-made fibre with very little tendency to shrink.
EMI: Electromagnetic Interference. Undesired interference
from electromagnetic fields.
NEMA: National Electronic Manufacturers Association,
American association which represents the designers
and manufacturers of electronic equipment.
IP: Ingress Protection, protection against access or touch.
Stainless Steel: alloy of iron, chromium and carbon
which forms a layer of chromium-oxide on the outside
so that the material does not corrode any further.
µ-metal (mu-metal): Nickel-iron alloy with 2% copper
and molybdenum with a very high magnetic permeability. As a result it is very suitable for blocking magnetic
fields.
Advertisement
22
elektor electronics - 12/2006
next month in elektor
milling
hands-on
electronics
Multi-Purpose
3D Milling Machine
Ready to go kit
specially for Elektor readers
In next month’s issue, Elektor
– in cooperation with a renowned
manufacturer – presents an extraordinary
project many readers have eagerly awaited
for many years: a real 3D milling machine
for home construction from a kit.
taining all metal parts (see photograph), the stepper motors,
two ready-assembled circuit boards for the drive electronics,
a spindle motor with the associated clamping device and a
base plate. The kit comes with special control software that’s
remarkable for its user-friendliness and general structure.
All this is supplied to you no less than 700 pounds (approx. 1,000 euros) cheaper than a comparable instrument.
Thanks to the article in the January 2007 issue and the supplied manual, construction of the Elektor Milling Machine
is straightforward. Our serious advice is not to miss this
unique project in next month’s issue!
(060232-a)
Although we can’t go into
details right now, we also
can’t resist printing a photograph of all metal parts
that go into building
the Elektor 3D Milling Machine. Tell us if
we’re wrong when claiming that this is the first affordable tabletop milling
machine for anyone doing precision mechanical
work, be it for electronics,
electromechanical applications or modelling.
The machine is versatile
in that it can be used not
just for the production of
front panels and 3D models, but also to mill your
own PCBs at home, at incredible precision.
The net X-Y working surface of the milling machine is 300×400 mm,
while the Z (vertical)
range is specified as
100 mm.
A complete kit
The Elektor 3D Milling
Machine comes as a
comprehensive kit con-
12/2006 - elektor electronics
23
hands-on
sw receiver
Shortwave Cap
0-30 MHz SSB/CW/FM/AM/DRM
based on DDS and RISC
Gert Baars
As a special treat for all radio amateurs we present a general-coverage AM/FM/SSB receiver with a
wide range of features, which uses a DDS chip in the VFO section and also has a DRM output that can
be fed into a computer. The receiver is controlled by a modern 8-bit Atmel RISC processor. The frequency
readout is on a clearly legible 7-segment LED display.
This receiver can be seen as the successor to the receiver published in
January 1999. The experience gained
with this predecessor (which, incidentally, has given a lot of listening enjoyment to many constructors), has been
used to develop a more advanced RF
receiver.
The best parts of the original design
have been kept, such as the Intermediate Frequency (IF) and double conversion superheterodyne sections. This
new receiver also has some new functions up its sleeve. For example, there
is a DRM output that can be fed to a PC
for decoding. The tuning resolution has
also been improved to provide better
fine-tuning, for example for SSB reception. This makes the frequency readout
more precise and a BFO is no longer
necessary.
Some thought has also gone into the
aerial input stage, where an active input circuit makes the need for a long
wire aerial superfluous.
We’ve also used a number of contemporary components, such as a DDS
chip that is used as a VFO; the receiver
is controlled by a modern 8-bit Atmel
RISC processor.
In the design our preference went to 7segment LED displays for the frequen-
24
cy readout, which look better and are
easier to read than a standard LCD.
This receiver has three switchable
bandwidths, each of which is optimised for use with one of the possible
reception modes (AM, FM, USB, LSB or
DRM).
The sensitivity and the ability to deal
with large input signals have been improved for the reception bands of this
receiver, largely through the use of a
diode-ring mixer as the first mixer.
The receiver itself generates very little
noise. This can be clearly heard when
you aren’t tuned into a station and connect an aerial: the noise then increases
markedly.
A sensitivity better than 1 µV makes
little sense at lower frequencies
(roughly below 7 MHz), since it isn't the
weak signals that make reception difficult, but rather the strong signals that
swamp the others. A higher gain just
doesn't make sense under these circumstances. Furthermore, you should
find that most signals in that frequency
range are strong enough to be picked
up, even with a telescopic aerial.
In a nutshell, this is a receiver that is
suitable to pick up all broadcast and
amateur bands between 0 and 30 MHz.
The ease of operation, its various functions and its performance are guaran-
teed to provide you with many hours of
listening enjoyment.
Block diagram
The block diagram (Figure 1) starts
with the aerials. The internal aerial is
followed by a high impedance amplifier with adjustable gain. This amplifier isn’t required for an external aerial. Directly after the aerial switch is a
low-pass filter with a corner frequency
of 30MHz. This suppresses any possible image frequencies and other unwanted signals. Following this is the
first mixer. Its purpose is to convert
the range of 0 to 30 MHz into 45 MHz.
For this we require a VFO frequency of
45 to 75 MHz. A first IF of 45 MHz is
a good choice because the first image
frequencies are 90 MHz away, which
make them easy to suppress.
The VFO signal is produced by a DDS
generator. More details on its operation can be found in the DDS RF Signal
Generator article that was published in
the October 2003 issue. The reference
frequency for this DDS is obtained by
multiplying the 10 MHz crystal oscillator frequency by a factor of three to
obtain 30 MHz. Inside the DDS a PLL
multiplies this signal a further 6 times,
so that the internal reference frequency
elektor electronics - 12/2006
ture
becomes 180 MHz. Normally the maximum output frequency of the DDS
should be around 40% of this reference
frequency. But if we add a better bandpass filter at the output it is possible to
increase this figure somewhat.
The DDS can now produce frequencies
in 0.04 Hz steps. The required 100 Hz
resolution therefore isn’t a problem.
The DDS is controlled by a microcontroller that also takes care of driving
the frequency readout, the scanning of
the keypad and a rotary controller for
tuning.
A number of extra I/O lines take
care of the selection of the audio bandwidth and reception mode
(FM/AM/LSB/USB).
After the first mixer is the first IF filter, which has a bandwidth of 15 kHz.
This bandwidth determines the maximum possible bandwidth of the receiver. This filter suppresses the image frequencies of the second mixer. It also
removes any other unwanted by-products from the output of the first mixer.
The signal now arrives at the second
mixer. This converts the signal into an
IF of 455 kHz. It does this using a fixed
local oscillator frequency of 44.545
MHz. A frequency of 455 kHz was chosen because this low frequency can
be easily amplified, and also because
12/2006 - elektor electronics
Specifications
•Double conversion superheterodyne receiver - first IF 45 MHz, second IF 455 kHz
•Microcontroller control of the DDS generator and other functions
•Tuning range of 0 to 30 MHz in steps of 1 kHz or 100 Hz
•DRM output, suitable for connection to a PC soundcard
•Audio bandwidth of 3, 6 or 15 kHz, dependent on the reception mode
•Keypad with 16 keys for inputting the frequency, mode and bandwidth
•Memory for 64 frequencies, including bandwidth and mode
•Synchronous detector for AM
•Quadrature detector for FM
•Product detector for SSB
•Built-in adjustable input amplifier for a telescopic aerial
•Approximate sensitivity of 1 to 2 µV (without preamp)
•Supply voltage of 13 to 15 V, max. 650 mA
25
hands-on
sw receiver
Back in time
In the early days of Elektor Electronics many types of receivers were published on a regular basis. Due to the relatively high cost of semiconductors
in those days these receivers were usually fairly simple designs. As a comparison we’ll show you a shortwave receiver that was published almost 40
years ago in Elektuur magazine. The use of a tunnel diode in this design
is of particular interest because it was an unusual component at the time.
The construction method was nothing like what we expect these days: some
soldering pins and wires were used to connect all the components together
because it was quite difficult at the time to produce a PCB.
(Note: referring article not available in English)
there are many inexpensive but goodquality filters available for this muchused frequency.
The gain of the second mixer can be
adjusted automatically or manually.
To prevent overloading the receiver when very strong SW signals are
picked up via a wire aerial we had to
add a type of AGC circuit. Manually reducing the gain can improve the clarity
of SSB signals in the amateur bands,
because this reduces the QRM.
After buffering the signal it reaches
three switches that select one of three
filters, each with a different bandwidth. These switches are controlled
by the microcontroller.
The three bandwidths in question are
3, 6 and 15 KHz. 3 kHz is suitable for
SSB, 6 kHz for AM and 15 kHz for FM
and DRM.
After another buffer is the IF amplifier. This actually consists of two amplifiers, each with an adjustable gain.
This combination can provide a variable gain up to 80 dB, which is sufficient to keep the output signal level of
the IF amplifier constant for both weak
and strong signals. An AGC circuit is
used to adjust this gain automatically.
This adjustment follows the signal fairly quickly, for example to suppress the
fading of AM signals. But for SSB reception it is made to react more slowly. The fast ‘attack’ and slow ‘decay’
of this AGC circuit improves the audio
quality of SSB and CW signals.
The DRM detector takes the IF signal before it reaches the IF amplifier.
This is because we don’t need much
amplification of the signal for this output. The DRM transmitters are powerful enough to produce a few hundred
mVpp before the IF amplifier. This also
avoids any noise from being introduced
by the extra amplifiers.
The DRM detector consists of a product detector with a local oscillator of
467 kHz. The difference between this
and 455 kHz is 12 kHz, which is filtered before being made available at
the DRM output. The 10 kHz wide signal now covers frequencies between 7
and 17 kHz, which is exactly what is
required by the DRM program, Dream.
An AGC circuit is not really necessary for DRM because the soundcard
and the Dream program can cope with
widely varying signal strengths.
Following the IF amplifier are the AM,
FM and SSB detectors. The AM demodulator used here is a variation on
the well-known synchronous detector.
The principle of operation is that the
AM signal is multiplied by an unmodu-
Internal and external aerial
This receiver has the facility to accept two different aerials, each of which is suitable for different frequency bands.
Since even a suspended 20-metre length of wire has a higher impedance at low frequencies, an external aerial will work better than an internal
one down to about 500 kHz. For lower frequencies it is better to switch to a telescopic aerial, which also has a high impedance, but is only very
lightly loaded by the pre-amplifier.
The external aerial can be a 10 to 20 metre length of wire, suspended between 2 isolators, which in turn should be half a metre away from their
mounting points. These isolators can be easily made from a length of PVC tubing. Two pieces around 20 cm long, with holes drilled at the ends
are perfectly suitable. There is nothing special about the wire used for the aerial, as long as it can withstand strong winds and heavy rain.
We would advise against connecting an aerial longer than about 1 metre to the pre-amplifier. This can result in distortion and other interference. You would also pick up a lot of noise. A telescopic aerial or a 50 cm length of wire works perfectly well.
26
elektor electronics - 12/2006
lated signal with exactly the same frequency and phase. This signal can be
extracted using a PLL with a slow control loop, which keeps the frequency
and phase intact should the AM signal momentarily fade. This results in
less distortion than would otherwise
occur when the signal fades. It is however also possible to extract the mix
signal directly from the input signal as
long as the modulation components
are removed first. This is easily done
with the help of a limiter. This limiter
provides a mix signal with a constant
amplitude. This results in good quality
AM reception and provides better resistance to distortion caused by fading
(which occurs in diode detectors).
We have used a quadrature detector for
FM signal is multiplied by the same
signal, but with a phase shift. This
phase shift is exactly 90 degrees at the
IF, but increases/decreases for an increase/decrease in the input frequency.
The multiplier has a corresponding increase/decrease in its output voltage.
A balanced multiplier is in fact very
similar to an EXOR phase detector.
The SSB detector is a product detector. If we multiply the USB or LSB signal with another signal that differs in
frequency by 1.5 kHz we obtain the
audio signal. Note that the first mixer
creates a mirror image of the spectrum
because we’ve using high-side injection, so USB becomes LSB and vice
versa. The local oscillator used here
is controlled via the microcontroller to
This comes in very useful when receiving SSB and CW signals. The corner
frequency can be adjusted from about
500 to 3500 Hz.
To reduce interference from hum and
other low-frequency noise a high-pass
filter is also included, with a corner frequency of 300 Hz.
The last stage is an audio power amplifier. This has sufficient output power
to drive a 2 W loudspeaker.
Schematic
We should now be familiar with the
general design of the receiver. This
makes it a lot easier to interpret the
circuit diagrams in Figures 2, 3 and
4, and not be alarmed by the large
467 kHz
INT
ANT
44.545 MHz
AM DET
AM
455 kHz
DRM
OUT
NARROW
INT
0 - 30 MHz
MIX-1
0 - 30 MHz
45 MHz
3 kHz
ANT
GAIN
FM DET
IF AMP
455 kHz
MEDIUM
EXT
EXT
ANT
6 kHz
MIX-2
LIMITER
455 kHz
RF
GAIN
90ϒ
AGC
WIDE
45 - 75 MHz
500 - 3500 Hz
FM
SSB DET
15 kHz
SSB
S METER
SSB
10 MHz
6x
DDS
SLOW
300 Hz
LSB/USB
x3
7 SEGMENT DISPLAY
LSB/USB
456.5 kHz / 453.5 kHz
SSB
AM
FM
DISPLAY DRIVER
AF
NARROW
MEDIUM
WIDE
VOLUME
TUNING
KEYPAD
rotary
encoder
030417 - 13
Figure 1. The block diagram for the shortwave receiver is fairly elaborate. At the bottom-left you can see the DDS generator and the microcontroller.
the FM demodulator. This type of detector is well known for the good quality audio it produces. The limiter used
in the AM section is also used to supply a constant amplitude signal to the
FM demodulator. This helps to avoid
distortion caused by very strong input
signals. It also suppresses AM noise
components, which could overload
the detector if their amplitude was too
large. The overall result is a much better signal to noise ratio.
The actual operation of the quadrature
detector is quite straightforward. The
12/2006 - elektor electronics
produce either 456.5 or 453.5 kHz. This
way the selection can be made via the
keypad.
The microcontroller selects the required mode with the help of analogue
switches that follow each of the detectors. Since the microcontroller controls
both the mode and the bandwidth it
can automatically select the appropriate bandwidth for each mode.
After the switches is an adjustable
low-pass filter with a steep cut-off.
This suppresses whistles and other interference caused by nearby stations.
number of components. The blocks
shown in Figure 1 can be found in the
diagrams quite easily.
We start with descriptions of the principal parts in the receiver section of
Figure 2. At the top-left is the pre-amplifier for the telescopic aerial. This
consists of a two-stage configuration
using low-noise DG MOSFETs (T1/T2).
The first stage provides a high-impedance input and some gain, the second
stage functions as a 50 Ω driver. The
gain can be adjusted from about +6 to
–20 dB.
27
hands-on
sw receiver
0V27
100n
EXT
IC11
BF991
BF992
G1
R10
C8
C9
100n
100n
0V28
100n
1
G2
2
100n
S
D
B
(SOT23)
(SOT143)
C46
B
K4
C35
470nH
C14
4
RF
2
C16
150p
3
6
LO
1
150p
L5
L7
470nH
390nH
L4
R11
C17
39p
40p
40p
150nH
0V9
C22
10n
WIDE
MEDIUM
R26
R27
L13
C40
10k
150k
R23
C20
FL2
100n
FL2A
D4
100µH
R14
BA482
SFR455J
BF992
R12
C19
C68
100n
2k2
J310
3V7
L6
C18
100nH
1k8
2k2
IF
330nH
C15
39p
5
C132
100n
R28
SFR455J
FL3
TR1
LMC4202A
D5
2k2
R15
150p
R24
T7
A
1n
D6
2k2
K3
C44
J310
R29
FL4
100n
C41
BA482
SFR455H
C43
150k
L3
100n
R22
10n
T3
R25
3x
100n
C25
T4
C23
FL1
45G15B1
ASK-1
100n
+9V
R30
100k
L2
MIX1
R20
820µH
RF GAIN
(15dB)
1n
C39
470 Ω
330nH
C13
15p
100n
10p
C52
2n2
78L09
C38
R13
C21
P2
100k
C34
C32
40p
C51
330p
C37
D10
47k
L1
C30
56p
40p
C12
R17
C29
470 Ω
C26
15p
D9
L9
1V6
56 Ω
C11
6p8
100µH
2x 1N4148
220k
BFR93A
39p
10p
R19
10k
2p2
C50
IC12
U+
100n
R18
47k
C28
44,545MHz
C36
470n
L8
10k
C31
T5
C24
3
CSB470
NARROW
+9V
100n
390nH
X1
LSB/USB
C33
L10
390k
100n
4k7
100µH
R16
C27
+9V
R21
7
270p
10n
4V5
L11
X2
DRM
OSC
6
C49
E
10µH
10n
5
OUTB
K1
L12
5k6
NE612
OSC
C
C131
C53
R32
4
OUTA
INB
R31
BFR93A
3V8
8
INA
IC1
10n
78L09
R7
C5
C45
100n
4V0
AM
R5
220 Ω
100k
100n
INT
100µH
C48
FM
R3
C3
1V4
U+
BF992
15k
R1
1N4001
SSB
BF991
100n
250mA
2k2
T2
P1
100k
ANT
GAIN
JP1
L14
500mA
R9
2k2
T1
C47
+13V...+15V
K6
D1
BA482
R8
1n
1µ
K7
22k
C4
150p
D3
820µH
1µ
8V6
39 Ω
C1
BA482
2
+5VD
U+
150mA
C10
U+
C6
7V8
D2
1µ
1k
100n
C7
EXT
ANT
L22
R6
270k
1k
R4
56 Ω
R2
C2
220 Ω
K2
A1
BA482
SFR455E
IC13
U+
78L10
4
1µH
2
1n
1V4
+10V
1n
4
MAR8
100n
4V0
C83
1n
T8
U+
78L10
D
G
D
3k3
100µH
C65
M1
150µA
100n
100µH
C72
R53
S-meter
P3
5V1
8
3
C69
IC3
AD603
4
1
COM
100n
7
2
6
4
C70
3k3
10n
100n
+5VD
100n
R42
3V1
R43
C77
3
C92
100n
CSB455
C87
C88
C89
C90
330p
22n
8n2
100n
C94
2x
BF494
C97
T10
1
2k2
R47
2
33p
T11
8
INA
OUTA
INB
NE612
OSC
OSC
6
7
C99
OUTB
R58
3
T15
C103
10k
100n
5
C100
10n
BS170
C101
C102
C104
22n
3n9
100n
1n
+5VD
BF451
U+
78L05
4
IC6
C98
3k3
100µH
100n
R48
IC16
K
100n
C96
100n
100n
L19
C95
100p
L20
C78
100n
A
BS170
R52
7
5k6
100n
35
100p
T13
R57
C80
BS170
C75
330k
C66
27k
KV1235
X3
R56
100n
47µ
R44
5
OUTB
220p
C93
10n
T9
4V4
100n
3V8
R46
C79
+10V R39
R40
6
C86
1V4
C74
100n
10k
OSC
BF494
100n
22k
NE612
OSC
T14
R54
6
R38
C63
D7
2 C73
10n
INB
T12
C91
R51
+5VD
C76
1N4148
7
1
COM
C67
2
C85
NUL
5
IC4
AD603
3V8
4
OUTA
2k5
5k
FS
8
3
5
100n
5V1
3k3
C62
100n
R55
P4
15k
100n
8
INA
+
C56
100n
100n
IC5
10n
BC547B
L17
C71
1
10n
5k6
3k3
R41
G
10k
L16
100n
R50
S
S
C64
C84
R49
3k9
4k7
IC14
+10V
100µH
C82
BS170
J310
R45
R37
C57
100n
C61
3
2k2
Synth.
1
U+
L18
C81
68 Ω
IC2
C60
IC15
78L05
2
R36
K5
+5VD
3
MAR8
220 Ω
100n
1
10k
100n
L15
47k
C55
47k
C54
C109
L21
LMC4100A
100n
1V4
C114
10k
C121
47n
47n
C122
3
47n
R69
47k
6
R34
220k
47n
C120
7
5
LF356
4
6
470µ
C127
IC10
LM380N-8
IC9
3
1
100n
5
4
R63
R65
R68
C124
C125
470n
100n
VOLUME
P6
50k
log
C105
C107
1n
10n
1
4k7
25V
C129
220µ
25V
6
C106
1n
R35
R60
8
INA
OUTA
4
IC7
2
6V0
7
2
100n
COM
3
C117
KV1235
C119
100k
C116
10n
D8
C118
47n
OS
CLK
5
10k
8
OUT
IC8
MAX7400
1
100µH
3V8
R59
1µ
2
SHDN
IN
100n
C128
7
5k6
2
100n
100n
C126
47p
330k
C115
100n
R33
470 Ω
2V5
4
C123
C59
2Ω7
R67
220k
R62
22k
R66
15k
R64
100k
1µ
C58
C110
C108
10n
INB
NE612
OSC
OSC
6
7
OUTB
3
5
C112
T16
D
S
100n
BS170
G
C111
C113
22n
100n
LS1
C130
R61
100n
8Ω
2W
4k7
P5
030417 - 11
28
elektor electronics - 12/2006
3
IC6
+U
IC7
78L05
L1
+U
+5VA
+U
7805
+5VD
IC5
7805
+U
R4
100µH
3k9
R1
T1
C1
C10
C11
C19
C20
40
100n
1
2
100n
3
4
5
6
KEYPAD
7
8
31
10
11
12
K4
13
14
15
ENCODER
DISPLAY
16
17
100n
PC0(A8)
PB1(T1)
PC1(A9)
PB2(AIN0)
PC2(A10)
PB3
PC3(A11)
PB4(SS)
PC4(A12)
PB5(MOSI)
PC5(A13)
PB6(MISO)
PC6(A14)
PB7(SCK)
PC7(A15)
PA0(AD0)
PD1(TXD)
PA1(AD1)
PD2(INT0)
PA2(AD2)
PD3(INT1)
PA3(AD3)
PD4
PA4(AD4)
PD5(OC1A)
PA5(AD5)
PD6(WR)
PA6(AD6)
X1
L4
100µH
FM
23
L5
100µH
AM
LSB/USB
24
L6
100µH
SSB
25
L7
100µH
NARROW
26
L8
100µH
MEDIUM
27
L9
100µH
WIDE
C12
C13
C14
C15
C16
C17
C18
100n
100n
100n
100n
100n
100n
100n
35
C24
34
32
10µH
4x
100n
10µH
7
10k
6
33p
28
18
VINN
D2
1
8
11
D5
VOUTN
D6
VOUTP
D7
DACBP
WCLK
IOUT
FQUD
IC4.D
9
8
1
22p
L14
C31
1µH2
10p
C33
40p
10MHz
15
IC4.C
C30
X2
3x
100n
5
1
220p
6
9
13
14
17
L18
C34
21
22p
RST
VINP
RSET
IOUTB
PGND DGND AGND AGND
4k7
10
1
AD9851BRS
12
C32
IC3
D4
16
R6
REFCLK
D3
7
22
A'
11
D1
2
SSB
23
D0
C3
B'
12
1
10µH
PVCC DVDD AVDD AVDD
FM
WIDE
100n
R9
IC4.E
C29
AM
WIDE
13
L13
C26
FM
MEDIUM
1
C28
AM
MEDIUM
L10
C25
25
SSB
14
C22
100n
IC4
4
IC4 = 74HCU04
C27
L12
33
26
NARROW
100µH
C21
10k
27
LSB/USB
3
IC4.F
3
NARROW
L11
IC4.B
C23
36
4
C'
+5VA
2
1
R8
37
18
8MHz
C
38
X1
33p
IC4.A
1
+5VA
X2
19
10n
220µ
16V
100µH
39
PA7(AD7)
PD7(RD)
L3
22
29
AT90S8515
PD0(RXD)
21
28
OC1B
ICP
C2
100n
C6
100n
K3
IC1
PB0(T0)
C40
100n
C5
100n
30
ALE
20
100n
C4
5
24
10
19
C35
L15
330nH
L17
820nH
10p
Synth.
L19
1µH
330nH
C38
C39
27p
6p8
K6
R11
68 Ω
20
R7
R10
L16
68p
LSB/USB
L20
C36
R12
C37
68p
100nH
68 Ω
22k
K1
C41
RESET
100 Ω
R3
1k
R2
9
100 Ω
BC557B
+5D
4Ω7
100nH
030417 - 12a
Figure 2. Schematic for the receiver section. There is a heap of components, but with the help of the block diagram you should be able to recognise the different sections.
Figure 3. The electronics for the microcontroller board. The wiring diagram at the bottom-left shows how K3 and K4 on the receiver board should be connected to K3 on the microcontroller board.
Figure 4. This part is on the display board: display driver, six-digit 7-segment display, rotary encoder and a keypad with 16 keys.
Two BA482 switching diodes (D2 and
D3) have been used to switch between
the pre-amplifier and the input for an
external aerial. This type of diode has a
very low AC resistance when a certain
forward current flows through it.
From the aerial switch the signal is fed
into a low-pass filter consisting of inductors L1 to L3 and capacitors C11
to C17. These form a Cauer-like filter,
which cuts of steeply at 30 MHz and
which has a fairly flat transfer function
4
from 0 Hz onwards. The input and output impedance of this filter is 50 Ω.
Following this input filter is the first
mixer (MIX1). We have used an ASK-1
diode-ring mixer made by Mini-Circuits
Labs because of its ability to handle
+5VD
+5D
L2
K2
KEYPAD
100μH
100n
220μ
C8
C9
16V
100n
R5
10k
C7
LD6
SC52-11GWA
a
b
c
d
e
f
g
CC
3
CC
dp
LD5
SC52-11GWA
7
a
6
b
4
c
2
d
1
e
9
f
10
g
5
CC
8
3
CC
8
dp
LD4
SC52-11GWA
7
a
6
b
4
c
2
d
1
e
9
f
10
g
5
CC
3
CC
8
dp
LD3
SC52-11GWA
7
a
6
b
4
c
2
d
1
e
9
f
10
g
5
CC
3
CC
8
dp
LD2
SC52-11GWA
7
a
6
b
4
c
2
d
1
e
9
f
10
g
5
CC
3
CC
8
dp
a
6
b
4
c
2
d
1
e
9
f
10
g
5
CC
3
CC
8
dp
18
ISET
LD1
SC52-11GWA
7
DIN
7
14
6
16
4
20
2
23
1
21
9
15
10
17
5
22
24
S1
K5
19
IC2
SEGA
CLK
LOAD
SEGB
SEGC
DIG7
SEGD
DIG6
SEGE
DIG5
SEGF
DIG4
SEGG
DIG3
SEGDP
DIG2
MAX7219
DIG1
DIG0
DOUT
4
1
13
ENCODER
S14
S10
S6
S2
1
2
3
A
S15
S11
S7
S3
4
5
6
B
S16
S12
S8
S4
12
8
5
ENCODER
DISPLAY
10
3
7
6
11
2
9
7
8
9
C
S17
S13
S9
S5
*
0
#
D
030417 - 12c
030417 - 12b
12/2006 - elektor electronics
29
hands-on
sw receiver
large input signals. Eagle-eyed readers
will have noticed that we've swapped
the input (RF) and output (IF) of this
mixer. The mixer still works well in this
configuration, but it has the advantage
that the input frequency can be a lot
lower (almost down to 0 Hz, which is
exactly what we require).
At the output of this mixer is a filter
(FL1) that suppresses unwanted byproducts and the image frequencies
of the second mixer. An LC network at
the output of the filter is used to match
the impedances.
We now come to the second mixer,
which is built around T3. Again we've
chosen a low-noise BF992 DG MOSFET, which has a somewhat better
gain than most of its equivalents.
As far as large signals are concerned,
we don't demand as much from this
mixer as from the first one because
we're only really dealing with one frequency here.
30
The purpose of the second mixer is
to convert the first IF (45 MHz) to the
second IF (455 kHz). Specifically designed crystals with a frequency of
44.545 MHz are available for this.
The local oscillator signal for the first
mixer is supplied by a VFO. This signal
should have a frequency range from 45
to 75 MHz, with a resolution of 100 Hz.
We have chosen an AD9851 DDS made
by Analog Devices (IC3 in Figure 3),
which provides a stable output that
can be accurately programmed in small
frequency steps. The frequency range
is just about wide enough for use in
our application.
At the output of the DDS is a bandpass filter (C34 to C39, L15 to L20) with
a pass-band of 45 to 75 MHz. The impedance of this filter is about 100 Ω.
The impedance is lowered to 50 Ω using resistors R11 and R12. The resulting signal is then fed into a MAR8 (IC2
in Figure 2). This IC (made by Mini-Circuits) is a wide-band amplifier with an
output impedance of 50 Ω. It also has
enough output power to provide a good
signal to the first mixer.
After buffer T4 come the three IF-filters. These can be individually turned
on using BA482 switch diodes (using the same principle as for the aerial switch). In this way the processor
can switch between bandwidths of 3,
6 or 15 kHz. The IF transformer (TR1)
provides a load to the filters that is
equal to their characteristic impedance
(about 2 kΩ).
Following buffer stage T7 we end up
at the IF amplifier consisting of two
AD603 ICs made by Analog Devices
(IC3/IC4). The gain of each IC can be
adjusted over a range of 40 dB using
a control voltage. This gives us a total
AGC range of 80 dB. The AGC voltage
is also used to drive the S-meter (M1)
via T8.
The detector for DRM signals has been
placed straight after the IF filters. The
IF amplifier is therefore not used for
DRM signals, as we saw from the block
diagram.
The DRM signal is decoded using a
product detector that mixes the input signal with a fixed frequency signal. The mixer chosen for this job, an
NE612 (IC1), has an internal Colpitts
oscillator. This can be set up in such
a way that it will oscillate at a slightly lower frequency than the resonant
frequency (470 kHz) of the CSB470 resonator used here. In this case we require 467 kHz, which is used to convert
the 455 kHz signal into an audio signal
with a centre frequency of 12 kHz and
a bandwidth of 10 kHz.
We've now arrived at the detection
stages for AM/FM/SSB. Each of the
three stages has been built around an
NE612 mixer IC.
The SSB detector is a product detector, as is usual for SSB detection. The
internal oscillator of IC5 has been configured in such a way that only one
resonator is required for USB as well
as LSB operation. The trick used here
is that transistors T13 and T14 switch
a trimmer capacitor either in series or
parallel with the resonator.
The quadrature detector for FM demodulation is built around IC6. This
detector is very suitable for demodulating narrow-band FM, which is normally used in the 11-metre CB band.
Because this type of demodulator is
somewhat sensitive to AM components, the signal is first fed through
a limiter. This consists of a push-pull
transistor pair (T10/T11). With the val-
elektor electronics - 12/2006
ues for the base resistors as shown, the
signal fed to the NE612 will be limited
to about 250 mVpp.
The simplest way of demodulating AM
is with an ordinary diode. A disadvantage of this method is that distortion is
introduced when the signal weakens.
This is particularly a nuisance when
the signal fades, which happens fairly
often in the shortwave bands. A synchronous detector (IC7) is more resistant to these occurrences. For this we
use the signal from the limiter in the
FM demodulator as the mix signal.
The amplitude of this signal is virtually constant, even when the signal
strength varies a lot.
There are now three available audio
signals, which can be selected by the
microcontroller with the help of three
BS170 FETs that are used here as analogue switches.
The receiver is equipped with an effective audio filter. This is an adjustable
low-pass filter built around IC8, which
is a switched capacitor filter. The
MAX7400 contains a very steep 8th-order elliptic filter where the corner frequency can be set with a capacitor. If
a varicap (D8) is used for this capacitor
then the corner frequency can easily be
adjusted using a potentiometer.
An active fifth-order high-pass filter
built around opamp IC9 then filters out
all frequencies below 300 Hz, thereby
suppressing hum and other low-frequency noise.
The final stage is an audio power
amplifier, for which we’ve chosen an
LM380N8. This 8-pin IC is capable of
delivering 2.5 W of audio power when
supplied with a high enough voltage.
In this case we supply the IC with the
maximum permitted 15 V, which means
it can output 1.7 Watts into 8 Ω. That
should provide enough volume in even
somewhat noisy environments.
At the heart of the control section is an
AT90S8515, a microcontroller made by
Atmel (IC1, Figure 2). One of its many
tasks is to provide the DDS chip (IC3)
with the correct data. This happens serially, which requires fewer I/O lines.
For each change in frequency the microcontroller sends a string of 40 bits
to the DDS. This happens extremely
quickly so you won’t experience any
delays when changing the frequency.
A rotary encoder (S1 in Figure 4) is
used for tuning purposes, which supplies 24 pulses per turn to the microcontroller. The encoder output is con-
12/2006 - elektor electronics
nected to an interrupt line, so not a
single pulse will be missed by the software. In this way you can tune through
24 kHz per complete turn at 1 kHz resolution, which is sufficient for AM and
FM. The resolution can also be set to
100 Hz (via the D key on the keypad),
which results in 2.4 kHz per turn. This
is more suitable for tuning to a SSB station on the amateur bands. It would be
nice if the encoder could provide more
pulses per turn, so that you could always tune with a resolution of 100 Hz.
In this case you would need to replace
it with an optical encoder.
The keypad is another input device
that is read by the microcontroller. This
consists of 16 keys arranged in a 4x4
matrix. In this configuration you only
need eight I/O lines to scan the keypad. The scanning is fast enough never
to miss a key press.
Apart from the 10 digits there are four
extra keys on this keypad (A, B, C and
D), which are used as function keys.
This reduces the number of switches required on the front panel, which
makes the operation of the receiver
easier and clearer.
Three I/O lines from IC1 provide the
display driver (IC2, a MAX7219) with
data. This display driver can drive a
maximum of eight 7-segment displays.
This IC also has the ability to control
the segments of each digit individually.
In this application we’ve made use of
this to display an L for LSB mode, a U
for USB mode, an F for FM and an A for
AM, followed by the bandwidth of either 3, 6 or 15.
And last but not least, the microcontroller also controls the settings for the
bandwidth and modes. This is implemented very simply using seven I/O
lines, which are set high or low by the
software as required.
The software for the microcontroller
was written in assembly language and
consists of about 1800 lines. The program has 21 subroutines, a reset routine, two interrupt routines (encoder,
timer) and a main program loop.
A programmed microcontroller can be
ordered from Elektor Electronics using
order code 030417-41. In contrast with
other Elektor Electronics projects we
can’t provide you with the source and
hex files.
Each of the boards for the receiver has
been provided with a liberal number of
voltage regulators, suppression chokes
and capacitors. The supply voltage
of 13 to 15 V (0.5 A) is provided by a
mains adapter.
Construction
The electronics for the whole receiver
have been divided across three boards:
a receiver board, a microcontroller
board and a display board. Due to a
lack of space we can only show you
photos of the completed boards in this
article. All PCB layouts, component
overlays and the accompanying parts
lists can be downloaded free of charge
from http://www.elektor.com/ (look in
magazine/December 2006/Shortwave
Capture). Ready-made circuit boards
are available from Elektor’s business
partner ‘The PCBShop’ (Eurocircuits)
In a radio receiver you obviously have
to use several special RF components.
The best places to buy these are specialist electronics firms, such as internationally operating Barend Hendriksen (www.xs4all.nl/~barendh) or HaJé
Electronics (www.haje.nl).
A reasonably experienced electronics
hobbyist shouldn’t have many problems with the construction. There is
only one component that is difficult to
solder: the DDS chip, which comes in
an SSOP package!
There are two parts lists: one for the
receiver board (‑1) and one for the microcontroller and display boards (‑2).
Take care not to mix up the part numbers between ‑1 and ‑2.
The receiver board (030417‑1) has been
made very compact. Because of this,
the space for resistors is smaller than
usual and the leads have to be bent
closer to the body. All SMD FETs and
SMD transistors have to be soldered on
the underside of the board. The MAR8
can be soldered either way up, since
its connections are symmetrical. Do
take care that the lead with the indicator dot is soldered to the correct pad.
We decided to make JP1 (aerial input
selection) a header with a jumper, but
if you intend to change the aerial input
on a regular basis then it will be much
easier if you add another switch to the
front panel instead.
As we mentioned earlier, the soldering of IC3 and its associated components on the microcontroller board requires very delicate work. The decoupling capacitors are also SMD; these
should preferably be 0805 types, but
1206 ones fit as well. You should first
solder two decoupling caps (C24, C27)
underneath (!) L12 and L13.
The headers for K3, K6 and the supply
are soldered onto the solder-side of
the microcontroller board (assuming
that you don’t want to solder the con-
31
hands-on
sw receiver
nection leads straight onto the board).
The voltage regulators (IC5 and IC7)
are also found here. These should be
provided with a heatsink, otherwise
they could overheat when the case is
screwed shut.
On the display board the following
components are mounted on the underside of the board: the rotary encoder
S1 (so the spindle sticks out through
the front), the headers (if required) for
K2, K5 and the supply, and electrolytic
capacitor C8.
It is best to use flexible wire for the
connections between the boards. If
you have plugs at one end it will be
easier if you ever need to separate the
boards in the future. The cables should
be kept away from the receiver board
as much as possible, otherwise you’re
asking for trouble (unwanted oscillations etc.). Route the cables directly
away from the board and only then
bend them towards the other boards.
Keep in mind that the order of the pins
on K3 and K4 on the receiver board is
different to that on K3 on the microcontroller board. We’ve added a wiring di-
32
agram in Figure 3 to remind you. You
should use a short length of thin 50 Ω
coax cable like RG174 for the synthesiser connection.
The supply for boards -1 and -2 loops
from one to the other, which is why
there are two connectors for the power supply. On the receiver board you
could use K6 for the incoming +13 V
for example, and use K7 to link it to
the microcontroller board and display
board. To avoid buying many different
types of plugs we have used only 5 or
8-way types and just removed the unwanted pins when necessary.
The dimensions of the boards are such
that they all fit in a standard case made
by Bopla. The receiver board is mounted
flat on the bottom of the case and the
other two boards are mounted vertically
behind each other (display board at the
front). The spindles of the potentiometers of the receiver board fit through the
holes in the other two boards.
The potentiometers should therefore
have long spindles so they can stick
through the front panel. When you use
the recommended Bopla case, the receiver board can be fitted using a few
M2.5 bolts.
It won't do any harm if you put some
aluminium foil on the bottom inside
the case and connect it to ground. This
provides some extra shielding, which
improves the stability of the receiver.
An unetched piece of PCB is also suitable. We don't want any short-circuits,
so the PCB should be placed with the
copper side down, or if you used aluminium foil you should cover it with insulation tape.
The microcontroller and the display
boards are mounted vertically inside
the case. Although there are slots
on the left and right-hand side of the
case for this purpose, we didn't use
them because the boards wouldn't fit.
Luckily the boards can also fit on either side of the slots. The 7-segment
display ends up just behind the front
panel and there is also enough room for
the microcontroller board this way.
At the back of the case is a connector for the supply, a feed-through for
the telescopic aerial and a hole for a
phono-socket for connecting an external aerial.
elektor electronics - 12/2006
Adjustments
There are nine trim points in this receiver. To start with it is best to set all
trimmers at their halfway position. Before you continue, make sure that an
aerial is connected and selected.
Select the AM mode using the A key
on the keypad. Next you should type
in the frequency of a strong medium
wave transmitter. If you use an internal aerial you should set the gain to
its maximum; this also applies to the
RF gain. Set the audio volume control
to halfway. If necessary, use the rotary encoder until you hear a medium
wave station. First set trimmer C30 in
the collector of the local oscillator of
mers should again be adjusted to get
the maximum S-meter-reading.
In a double conversion superheterodyne receiver there are two oscillators
that determine the accuracy of the reception frequency: the reference oscillator for the DDS and the local oscillator for the second mixer. You will
need a frequency counter to set these
correctly. First type in a frequency of
30.000.0 MHz. Then adjust the trimmer of the 10 MHz reference oscillator
(C30 on the microcontroller board) until the frequency at the output of the
MAR8 (IC2) is 75.000.000 MHz. The
frequency counter should now be connected to the second gate of the second mixer (T3). Use trimmer C26 near
AM station, for example in the medium wave. Then use the B key to select
LSB, and adjust C94 until you have a
beat frequency of 0 Hz. Now use the B
key to select USB, and adjust C93 until
you have a beat frequency of 0 Hz.
The receiver is now ready for use and
the case can be put together.
DRM use
In order to process DRM signals you
have to connect the DRM output of the
receiver to the line input of a soundcard in a PC. Next you have to start a
program called Dream. The bandwidth
of the receiver has to be set to 15 kHz
using the C key. The mode doesn’t mat-
Operation
Overview of all keypad functions:
The display for the frequency readout consists of 6 digits and has
a resolution of 100 Hz. This also means that when you input a frequency via the keypad you have to enter it down to the last 100 Hz.
For most stations you’ll find that you have to input the frequency in
kHz followed by an extra zero. But to receive the DCF time signal on
77.5 kHz you just type in 775 and D.
0 to 9 numerical input of frequency
D ‘Enter’ after inputting the frequency
A AM(6)/FM(15) mode
B LSB(3)/USB(3) mode
C select 3, 6 or 15 kHz bandwidth,
independent of the current mode
D select 100 Hz or 1 kHz tuning resolution
* store the frequency, mode and bandwidth into memory
*mm store into memory entry mm (mm = 00 t/m 63)
# recall the frequency, mode and bandwidth from memory
#mm recall memory entry mm (mm = 00 to 63)
(When the receiver is turned on, memory entry 00 is used by default.)
the second mixer to obtain maximum
volume.
Next you should set the trimmers of the
first IF (C18 and C19) to obtain maximum volume.
The full-scale trimmer (P3) and the
null-trimmer (P4) for the S-meter should
now be adjusted. The rest of the settings can now be completed with the
help of the S-meter reading.
If a very strong signal is received you
should reduce the aerial gain a little.
Adjust the core of the IF transformer
following the IF filters (TR1) to get the
maximum S-meter-reading.
When the sensitivity of the receiver improves you can try tuning to a weaker
station. The previously mentioned trim-
12/2006 - elektor electronics
There are two fixed decimal points on the display. So for channel 14
in the 27 MHz band for example, the display will show 27.125.0.
When the A or B keys are used to change the reception mode, the
display will show the selected mode for several seconds. For example, after pressing the A key it will show A-6, which means AM mode
with a 6 kHz bandwidth. Pressing the A key again will show F-15 (FM
mode, 15 kHz bandwidth).
Pressing the B key changes the display to L-3 (LSB mode, 3 kHz
bandwidth). Pressing the B key again shows U-3 (USB mode, 3 kHz
bandwidth).
Repeatedly pressing the C key doesn’t change the mode, but rather
the bandwidth. Again, the display will show the last selected setting.
When none of the A, B or C keys has been pressed for a period of
2 seconds the display will automatically revert back to the frequency
readout.
the 44.545 MHz crystal to adjust the
frequency to 44.545.000 MHz.
Once all these settings have been completed we can start with adjusting the
FM demodulator. For this we need to
tune in to an FM transmission in the
27 MHz band. We then need to adjust
the core of the IF transformer (L20) in
the FM demodulator. (Make sure that
you have selected the FM mode with
a 15 kHz bandwidth, using the A key.)
As an alternative, you could adjust it
for minimum audio output when receiving an AM station in FM mode, which
is actually a bit more precise.
The last trimmers to be adjusted are
the two trimmer capacitors in the SSB
product detector. Tune in to a known
ter, but if you select AM you may be
able to tell via the loudspeaker if there
is any interference in the DRM signal.
After setting up the soundcard on a PC
(this is described in the March 2004 issue of Elektor Electronics) the receiver can be tuned to a DRM transmitter. Many DRM stations don’t transmit
continuously, but according to a certain schedule. You can check via the
loudspeaker of the receiver whether
you’ve tuned into a DRM transmission.
A harsh noise should come out of the
loudspeaker. You can then turn down
the volume on the receiver. Finally, you
use the recording control on the PC to
set the required level.
(030417-I)
33
technology
communications
Spy Number St
Curious short wave transmissions...
Jochen Schäfer
Their programme content is not what you
would describe as entertaining but these
strange broadcasts have not only attracted
the attention of shortwave enthusiasts but
also songwriters and filmmakers… not
forgetting government counter-intelligence
departments.
Those of you who choose to do a spot of short wave
surfing during their idle moments will no doubt have
hit upon one or two stations where the programme
content is made up entirely of someone (usually female)
monotonously reading out a series of numbers. Your first
assumption may be that you have gate crashed a pirate
radio version of Bingo until you notice that some of the
numbers are repeated and grouped either in fours or
fives.
The enigmatic nature of these broadcasts is further
enhanced by the station identification which can take the
form of a repeated short extract of music played on what
sounds like a Stylophone. Once you add in ionospheric
distortion and fading the overall effect is quite spooky.
Recording artists have exploited this aspect of the
transmissions and used sampling and mixing to produce
interesting effects. The film ‘vanilla sky’ also features them
in the soundtrack. Directional antennas have been used
to pin-point the sources of these transmissions but they do
not appear on any official broadcast schedule and when
the authorities have been asked what the transmissions
actually represent their explanation seems to favour
‘meteorological data’.
Emerging Patterns
Transmissions from the number stations conform to a
regular pattern: They begin (usually on the hour or half
hour) with an interval signal (tone sequence, melody or
call sign), followed by a three number ID (usually), a
count value corresponding to the number of code groups
in the transmission followed by transmission of the code
groups. The former West German BND transmissions
were always terminated by a combination of two
characters from the NATO alphabet (over 80 different
34
combinations were used) followed by a tone burst.
The ‘Lincolnshire Poacher’ is an active station purported
to emanate from the RAF base at Akrotiri in Cyprus. Each
transmission begins on the hour and lasts for 45 minutes.
Transmission starts with the first 15 notes of the English
folk tune repeated 12 times followed by a 5 figure ID
repeated six times and two notes from a glockenspiel.
The message follows and is always composed of 200
groups of five numbers (the last number in each group
is pronounced with a rising terminal). The message ends
with six repeats of the poacher tune.
Number Stations do not exclusively use voice
transmission; Morse is also used as well as tone signalling
where each tone represents a different number. These
‘polytone’ stations are thought to be operated by the
Russian authorities.
Some irregularities
Not all the stations retain their original identity; take for
example a German speaking station which had been
broadcasting for over 30 years from a site in Poland.
Its station identification was an extract of the ‘Swedish
Rhapsody’ by Mantovani played on a music-box chime
or ‘ice cream van’ as it became more affectionately
known. The station closed after the catastrophic floods
of 1997 but today it is again in operation, at the same
site and frequency but this time thought to be run by the
British SIS. The mechanised voice which popped up on
this station was instantly recognised by Number Station
listeners as belonging to ‘Cynthia’ who they had heard
before on a station thought to be used by the American
security department.
The American security department ceased all Number
Stations activity in October 2003 but up until then they
elektor electronics - 12/2006
ations
Tools of the trade
One of the advantages of communicating via Number Stations is that no specialised equipment is needed to pick up the message, just a regular travel or ‘world
band’ radio which can be purchased on the high street (some stations use upper
side band). The operative just needs patience and an undisturbed period of time
to take down the numbers. Communication is only one-way of course but small
portable two-way HF radios have also been developed for use in the field.
An example of this type of radio equipment is the SP-15 short wave set developed in the late 50s by Wandel & Goltermann together with Pfitzner. The receiver
was fully transistorised while the transmitter used valves. The entire kit including
aerial, headphones, Morse key, fast code generator, crystals, NiCd batteries and
mains charger are all contained in a neat briefcase which enables the set to be
ready for use in a very short time. A notable design feature is the simple superhet receiver covering 2.5 to 24 MHz in two ranges with a BFO. For its time it had
an extremely good specification with a sensitivity of 22 µV at 10 dB S/N. Current
consumption is just 8 mA. It was used by the German secret service ‘Gehlen’ (later the BND) and also by the German armed forces for military reconnaissance.
The excellent receiver design also saw service in other NATO roles.
An example of a similar piece of kit ’from the other side’ is the Soviet built R-353
set, the transmitter and receiver are housed in a light metal case together with
the power supply. In contrast to the SP-15 the R-353 receiver is a valve double
superhet. An interesting feature of this unit is the programmable magnetic band
cassette shown attached to the front of the radio. Number sequences can be recorded onto the tape using a 10-way dial and then sent very quickly in a short
transmission burst. This dramatically reduces the risk of the transmitter being discovered by direction finding.
(EK)
were thought to be operating (like the Russian security
services) in Europe, transmitting in many languages
including English, German and Spanish. Up until a few
years ago there was also an Arabic speaking station.
It has never been established which government was
responsible for the ‘Swedish Rhapsody’ transmissions and
some stations are not always what they first appear; a
recent example was heard broadcasting from the Indian
subcontinent in July 2005 and given the designation
‘E22’, the transmissions bore the hallmarks of a Number
Station but it was finally identified in December 2005
as test transmissions by ‘All India Radio’ for a new
transmitter site. The station has since become fully
operational and reclassified as a broadcast station. More
details of the transmissions can be found at [1] under the
heading ‘E22 private room and study page’.
The West German SP-15 (Photo: Max O. Altmann)
Current activity
Number Stations first appeared at a time when there
were no alternative paths available to relay messages
worldwide. Today we have a number of options which
offer global coverage including satellite communication
and the Internet which you might reasonably assume
would eventually supplant radio broadcasts but in recent
months shortwave activity has actually grown more
intense. Among the organisations thought to be making
use of these stations are the British Secret Intelligence
Service (formerly MI6) The Israeli intelligence agency
(a.k.a. Mossad) and the Russian Federal Security Service
(FSB) together with other organisations broadcasting
from Chechnya, Cuba, Korea and two English speaking
stations in Egypt.
Their continued use by many countries highlights the
advantages of this form of broadcast which includes low
12/2006 - elektor electronics
Its Eastern Block rival: The R-353 (Photo: Max O. Altmann)
35
technology
communications
Table
Frequency
(kHz)
6959/ 9251/
11545
3150/4270,
4461,
6840/9130/
11565/
13533
Enigma
Identity
Details
E03
Thought to be used by the British Secret Intelligence
Service. Transmissions begin on the hour. Mode: USB
(J3E). Station ident: A repeated keyboard rendition of
a phrase from the ‘The Lincolnshire Poacher’.
E10
Thought to be used by the Israeli Mossad. Letter
groups are used instead of number groups.
Transmissions begin on the hour and half hour.
Mode: Mostly AM (A3E). Station ident: A three letter
sequence.
cost, low tech equipment (the signals can be picked up
on a relatively inexpensive ‘world band’ type of radio),
broad coverage (assuming the signals are not subject
to Electronic Counter Measures) and excellent message
security.
The stations usually operate at frequencies between the
recognised shortwave bands, using either AM (Amplitude
Modulation) or USB (Upper Side Band). They can be
located by sweeping these frequencies on the hour and
half hour. The table gives details of two stations which
are currently active. The ‘E03’ refers to the station identity
assigned by ENIGMA.
Those of you interested in exploring this subject in more
depth are recommended to pay a visit to the website of
Simon Mason [2] it contains many fascinating curiosities.
A popular American site is also included [3] — it has
a link to the ‘FRS Commander Bunny’ transmissions
complete with an attempt at decoding.
proved to be perfectly secure is the Vernam cipher or
‘one time pad’. The encryption key is a series of random
numbers each of which are added to each character
in the plain text message. The resulting cipher text is
transmitted and the receiver subtracts the key on their
‘one time pad’ to recover the plain text message. Even
with knowledge of some part of the message it does not
help to decipher the rest of the message provided truly
random keys are used. The weakness of this method is
that a ‘one time pad’ containing the keys must be carried
securely by the operator and then destroyed (by both
sender and receiver) after use, if it is stolen or copied
security will be jeopardised.
Over the years many agents working in foreign lands
(both east and west) have been found in possession of
tiny ‘one time’ pads sometimes concealed in a hollowed
out soap bar, sometimes inside the shell of a walnut (the
most literal form of code cracking?). Governments are
obliged to remain silent on matters of national security
and have never acknowledged the existence of Number
Stations or their possible role in espionage, the identical
nature of sequences transmitted by Number Stations and
those found on agent’s one time pads may of course
prove to be purely coincidental.
Without any hard evidence we are forced to accept
the official account that the transmissions are in fact
reporting snow fall levels in the Mediterranean but if you
find that explanation too difficult to swallow then maybe
that thought about pirate radio Bingo (at the tax payer’s
expense) really is the only plausible alternative.
(060229-I)
[1] http://mysite.wanadoo-members.co.uk/thesecretsiteofmike
[2] www.simonmason.karoo.net/page30.html
The Usergroup
The Number Station broadcasts have fascinated listeners
for decades. Since the end of the 1990s there has been a
European mailing list covering this subject from a Yahoo
group with the name ENIGMA 2000 (European Number
Information Gathering and Monitoring Association)
which evolved form the earlier ENIGMA group. This
organisation has catalogued and classified all the known
Number Stations and given them the designations E =
English speaking, G = German speaking, M = Morse
transmissions, S = Slavic speaking (Russian, Polish etc.),
V = various for transmissions in other languages and X
= transmissions using special types of modulation (e.g.
polytones as already mentioned). The mailing list for
ENIGMA 2000 has currently grown to around 700
subscribers.
Securing the information
It’s a sobering thought that while most radio programmes
are designed to capture the greatest number of listeners
this must be one of the few examples where the intended
audience for an international broadcast may be just a
single listener who alone possesses the key to unlock
the message. When transmitting cipher text to what is
potentially an entire global audience it is vital that the
encryption method employed is absolutely bullet proof
otherwise government counter intelligence departments
with their vast computational resources will surely
resolve (eventually) any encryption scheme based upon
a mathematical algorithm. One such cipher which has
36
[3] www.spynumbers.com
Further interesting links:
http://home.freeuk.com/spook007
http://www.irdial.com/conet.htm
http://en.wikipedia.org/wiki/Numbers_station
The Author
Jochen Schäfer is 34 years old and has been blind from
birth. He works as a documentation assistant at the German Blind Studies Institute (Blista) in Marburg. He has
been tuning into Number Stations in his free time since
1977 when he first started listening to short wave. His
archive contains more than 1000 cassettes documenting Number Station broadcasts; he is a subscriber to the
ENIGMA 2000 mailing list and one of four site moderators. In his capacity as Germany’s leading Number Stations specialist he has made contributions to many publications on the subject.
His Email address is (the_kopf@yahoo.com). He is keen to
get in touch with anyone who has any recordings of earlier
broadcasts (from the 60’s to the 80’s) to add to his collection, he has particular interest in broadcasts made by West
German BND stations with a two-letter call sign.
elektor electronics - 12/2006
LED/Light Kits
6 LED V.U. Meter
10 LED V.U. Meter
Wireless V.U. Meter
2 LED Flasher
3 LED Running Light
5 LED Flasher
10 LED Running Light
1CH 700W AC Flasher
2CH 1400W AC Flasher
10 LED Running Light, S
10 LED Running Light, C
Two Way Chasing Light
Circular Casing Light
12V Xenon Tube Flasher
Super Electronic Roulette
Mini Roulette wheel
Heads or Tails Game
Deluxe Electronic Dice
Dancing Light 3CH
LED Flasher 2 colour
Priority Machine
Xenon Tube Flasher
Super Bright Flasher
Robot LED Flasher
Fire Fly Light:
6 Prog chasing light kit
11 LED Star Flasher
25 LED Star Burst
4 Way Traffic Light
21 LED Chasing Arrow
Atom Chasing light
Two Way Chasing Light
Triangle Flasher 6 LED
Electronic Windmill
Electronic Sand Glass
£3.99
£4.99
£12.99
£3.99
£4.49
£4.99
£4.99
£5.99
£7.99
£8.99
£8.99
£8.99
£10.99
£8.99
£24.99
£9.99
£4.99
£10.99
£19.99
£3.99
£5.99
£15.99
£5.99
£5.99
£3.99
£15.99
£6.99
£16.99
£9.99
£9.99
£4.99
£7.99
£4.99
£9.99
£14.99
Radio Kits
FM Wireless Mic 3V
FM Wireless Mic 9V
Easy FM Radio:
Easy AM Radio:
Walky Talky Kit:
Advanced AM Radio Kit:
Advanced AM/FM Radio
IC Shortwave Receiver
MK484/ZN141 AM Radio
£6.99
£6.99
£14.99
£7.99
£34.99
£13.99
£16.99
£9.99
£9.99
12/2006 - elektor electronics
Noisy Kits
Steady Hand Game
Ambulance Siren
Fire Siren Kit
Police Siren Kit
Two Tone Door Bell
Christmas Tune Generator
13 Tone Mini Organ
25 Tone Mini Organ
Car Reversing Alarm
We would like to wish all
the readers of Elektor a
very merry Christmas
and a happy new year!
Telephone Kits
Recorder Starter:
£9.99
Telephone Amplifier
£9.99
10W Speaker ringer
£9.99
Phone Eavesdropper
£8.99
Dual Station Intercom
£10.99
Telephone In-Use display £2.99
Privacy Protector
£7.99
FM Telephone Transmitter £8.99
Ring Controlled Relay:
£12.99
Do-It-Yourself Phone:
£28.99
Free postage on any
kits ordered above £10
to the UK from now
until Jan 1st 2007.
Orders below £10 add £1 P&P
Outside UK: £1.50 per item
To order simply visit
www.lichfieldelectronics.co.uk
By Phone: (01543) 256684
Switching Kits
Light Activated Switch
£7.99
Light Controlled Switch
£8.99
Touch Switch
£8.99
Timer Switch 0-3hr
£9.99
Timer Switch 0-10hr
£16.99
Voice Activated Delay off £9.99
Voice Control Switch
£9.99
25ft Infrared Remote Relay £18.99
2CH 50ft IR Remote Relay £29.99
Light Dimmer 500W
£5.99
Electronic Code Switch
£14.99
Water Level Control
£9.99
By e-mail: sales@lichfieldelectronics.co.uk
Or by post to:
Lichfield Electronics, Conduit Street
Lichfield. Staffordshire. WS13 6JU
Other Kits
TV transmitter
£27.99
Sound Reversing Car
£14.99
DIY Digital Multimeter
£16.99
Electric Shock Machine
£5.99
300mA 6-9-12V PSU kit
£8.99
0-30V 3A DC Regulator
£15.99
0-30V 1A DC Regulator
£7.99
Emergency Light circuit 6V£5.99
Rat/Cockroach Banisher £8.99
£7.99
£5.99
£4.99
£5.99
£5.99
£3.99
£4.99
£8.99
£4.99
Other Kits
Muscle Massage equipment £12.99
PIR Sensor with 7A relay £27.99
Human to Robot Voice
12.99
Raining Alarm & speaker £6.99
Stereo Simulator Kit
£14.99
Low Battery Alarm 12V
£6.99
0-12V 500mA Regulator
£4.99
Digital Counter 2 digit.
£12.99
Transistor Checker
£6.99
Other Kits
Intruder Alarm
£9.99
Visitors Chime
£7.99
Magnetic switch Alarm
£16.99
Infra Red Alarm
£24.99
50V 30W Mono Amp OTL £14.99
12V 15W Stereo Amp BTL £24.99
8W 12V Sterep Amp
£13.99
2W 9-12V Stereo Amp
£9.99
Stereo Tone Control kit 15V£22.99
37
TECHNOLOGY READER CIRCUITS
Wireless Key
Compact and secure
Gert Baars
Secure wireless switching can be useful in a variety of applications, such as
activating or deactivating an alarm system, operating a garage door opener,
or operating an anti-start device in your car. The circuit described here is
compact and quite secure, and it operates in the 433-MHz ISM band.
The circuit described here consists of
a transmitter and a receiver operating
on a frequency in the 433-MHz ISM
band. Frequencies in this band can be
used without a licence if the transmitter power does not exceed 10 mW (10
dBm). The transmitter is small enough
to be fitted in a key fob, and it is powered by a 3-V lithium cell. The transmitter only emits RF signals in short
bursts, so the current consumption is
no more than 8 mA. As the battery
has a capacity of more than 200 mAh,
it will last for several years.
The receiver operates from a supply voltage of 5 V and can be powered from a
simple AC adapter. If it is used with a
car battery, a 78L05 must be connected
between the battery and the circuit. In
the case of a 6-V battery (such as is used
with motorcycles and scooters), a lowdrop 5-V regulator must be used.
A 24-bit code is used to ensure that the
receiver only responds to the proper
key transmitter. More than 16 million
combinations are possible with such a
code. To give you an idea of how big
this number is, with a burst length of
300 ms for each transmitted code the
transmitter would have to transmit an
uninterrupted series of sequential
codes for two months in order to work
through all possible combinations.
Microcontrollers are used in both units
(transmitter and receiver) to encode
and decode the key.
Transmitter
The actual transmitter consists of a
Maxim MAX1472 (Figure 1). This tiny
8-pin IC contains a crystal-controlled
38
ASK transmitter for frequencies in the
300–450 MHz range. It includes a
fixed-ratio PLL and a crystal reference
oscillator. This type of design is more
precise and more stable than an ISM
transmitter based on a SAW filter.
The frequency selected for this project
is 433.920 MHz. It results from the fact
that the PLL multiplies the crystal frequency by 32. Due to the narrow tolerances of crystals, the match between
the transmitter and the receiver is
always adequate without any need for
tweaking, in part because the receiver
has a relatively large bandwidth.
The crystal has a frequency of 13.560
MHz. It comes from Hong Kong X’tals
[1]. The coil in series with the crystal
ensures that the circuit oscillates at
the series frequency of the crystal.
The transmitter IC has a 10-mW output
stage, which is good for a range of at
least 10 metres indoors (the author
only tested the unit in his house). A
loop antenna etched on the circuit
board is often used in such applications, but a dipole antenna (too short
for the actual frequency) proves to give
better results. Two radiating elements,
each approximately 4 cm long and
0.04” thick are fitted to the transmitter
board for this purpose.
The transmitter IC has a data input for
ASK modulation. Although it can be
modulated at up to 100 Kb/s, we have
to point out that the reliability of the
system is better at lower rates. As only
a relatively small amount of data must
be transmitted, a bit rate of 100 b/s
can be used. It takes only 240 ms to
send 24 bits at this rate.
The total burst length is 300 ms,
including the preamble and start bit.
This burst is transmitted when the
transmit button is pressed, with a total
current consumption of around 8 mA.
The microcontroller switches the transmitter off immediately after the burst
has been sent and then enters a
power-down mode, with the result that
the total current consumption drops to
less than 2 mA as long as the button
is held pressed. This reduces the battery consumption to a minimum. In
practice, it means the transmit button
must be held down for approximately
half a second to transmit all the data.
A microcontroller is used to generate
the code key and transmit it by modulating the transmitter IC. The selected
microcontroller is the Atmel ATtiny15L,
which is available in an 8-pin DIP package. This microcontroller has an internal RC clock generator that runs at 1.6
MHz, and it can be set within a tolerance of 1% using a calibration byte.
After the flash code memory has been
programmed, the calibration byte must
read from the internal ‘signature space’
using the programming software and
then written to location 1023 ($3FF) of
the flash memory. This only has to be
done once when the microcontroller is
programmed. Each time the microcontroller starts up, it reads the calibration
byte from the flash memory and then
writes it to the oscillator calibration
register to optimise the accuracy of the
internal processor clock.
Receiver
The receiver is also a Maxim IC (Figure
2). In this case it is a MAX1473, which
elektor electronics - 12/2006
contains a complete superheterodyne
receiver. Here again the reference frequency, in this case the local oscillator
frequency, is provided by a PLL oscillator. As the PLL multiplication factor is 32
and the IF is 10.7 MHz, the crystal frequency must be 13.2256 MHz (see [1]).
A remarkable feature of this IC is the
integrated image rejection mixer,
which effectively means that the mixer
stage also provides adequate image
frequency suppression, so there is no
need for a preselection filter at the
input. As a result, the required number
of external components can be
reduced to around 20.
The receiver does not have to be tuned,
and it boasts a sensitivity of better than
approximately 1 µV. The receiver IC also
has an audio filter that is intended to
reduce noise and interference. A ‘data
slicer’ after the filter provides automatic
operating point adjustment to provide
data that is as reliable as possible even
under weak signal conditions.
The receiver unit is also fitted with an
ATtiny15L microcontroller. The data
slicer output of the receiver IC is connected to an input of the microcontroller, which decodes the burst signals
from the transmitter. If the decoded
data contains the correct key, it activates two outputs. Here OUT1 provides a 1-second pulse if the correct
key is received. It is intended to be
used to drive a mechanical latch
opener, which is usually fitted with a
solenoid. The second output (OUT2) is
an alternate-action switch. OUT2
S1
BT1
C4
C1
CR2032
10n
10n
L1
8
3V
7
*
VDD
IC1
5
2
6
PB0
RESET
PB4
PB2
PB1
PB3
5
1
XTAL1
EN
7
1
IC2
3
6
X1
DATA
ATTiny15L
MAX1472
4
4
C2
PAOUT
XTAL2
8
13.560MHz
C3
GND PAGND
100n
3p3
2
3
* see text
4cm
4cm
*
*
050364 - 11
Figure 1. The transmitter consists of only two ICs and can be built in a very compact form.
switches to +5 V when the key is
received once and then toggles to 0 V
the next time it is received. It can thus
be used to switch something on or off,
such the control unit of an alarm system or an anti-start device in a car.
the microcontroller to see whether a
valid key has already been generated.
If it hasn’t, the software immediately
generates a 24-bit code and stores it in
the EEPROM memory along with the
signature so it will not have to generate a new key the next time.
A random number generator that provides an arbitrary 24-bit value is
needed to generate the key. This is
implemented using a capacitor connected to pin 2 of the microcontroller.
The algorithm used for this purpose is
based on timing. Pin 2 is first configured as an output and held low for a
certain length of time to discharge the
capacitor. After that the pin is config-
Software
The software consists of two separate
programs – one for the transmitter and
one for the receiver. At somewhat more
than 400 lines each, these programs
are fairly modest in size.
At the transmitter end, after starting
up the software looks for a particular
value (a ‘signature’) in the EEPROM of
C1
C2
2
10n
+5V
14
7
10n
AVDD AVDD DVDD
11
1
X1
IRSEL
AGCDIS
XTAL1
XTALSEL
LNAOUT
C11
28
3
100n
L1
MIXIN1
1n
PDOUT
12
FL1
6
XTAL2
LNAIN
MIXOUT
DATAOUT
C12
16
100n
C4
Ø 15mm
C5
1n
2p7
8
26
R1
OUT
DSP
18
17
9
IFIN2
DFFB
IFIN1
DSN
OPP
LNASRC
C8
10n
1n
5
10
13
OUT1
PB0
PB2
PB4
PB3
PB1
5
2
6
ATTiny15L
OUT2
4
R2
22
20
C6
R4
220p
19
21
R3
AGND AGND DGND
L2
IC2
RESET
23
MIXIN2
DFO
4
C7
3
1M
10,7MHz
7mm
7
MAX1473
GND
5mm
1
DATA
25
IN
L2:
8
820 Ω
C3
IN
15
IC1
L1:
4mm
24
10k
8mm
VDD5
PWRDN
220k
C9
C10
100p
47n
JP1
LEARN
D1
R5
820 Ω
27
D2
LOW
CUR
22nH
050364 - 12
Figure 2. The receiver is somewhat larger, primarily due to the extensive circuitry around the MAX1473.
12/2006 - elektor electronics
39
TECHNOLOGY READER CIRCUITS
ured as an input, which also means that
an internal pull-up resistor is connected
to the pin. The capacitor is charged via
this resistor, but the charging time constant is much longer than the processor
speed. During the interval when the
input is still low, the value of the key is
changed in rapid succession by a certain algorithm. This continues until the
capacitor is fully charged. The input
read as a ‘1’ at this point, at which time
the key value stops changing. This
results in a key with a random value.
The randomness arises from the fact
that the charging process is slow relative to the processor speed. The charging time corresponds to approximately
10,000 processor cycles. Tests have
shown that the charging time is not
entirely constant. A 0.01-percent
change in the charging time is sufficient to yield a different result. This
also shows that two different transmitters will never generate the same key,
especially because most capacitors
have a tolerance of around 5%.
With a normal start-up process
(switching on power), the transmitter
first reads the key present in memory,
but it cannot be transmitted just like
that. A preamble with a duty cycle of
exactly 50% is transmitted first to allow
the operating point of the data slicer to
be adjusted. The transitions in the preamble also serve to synchronise the
timing at the receiver end with the timing at the transmitter end. A start bit
is transmitted after the preamble to
indicate that the key will be transmitted next. The ‘real’ data is only sent
after the start bit, after which the
transmitter is switched off immediately and the microcontroller enters the
power-down mode to minimise battery
consumption.
A different program is naturally necessary at the receiver end. In contrast to
the transmitter, the receiver is always
on. In this case the microcontroller listens to see whether anything is being
received. If it is, it checks whether the
received data stream has the correct
format, which includes checking
whether the start bit synchronisation
is valid. After this it reads the key and
compares it with the value of the key
stored in its EEPROM memory.
The microcontroller has a jumper that
enables it to read the key. If the jumper
is fitted, the receiver is in ‘learn mode’.
If a key is received in this mode, it is
stored in the EEPROM. This means
that the first time the receiver is
switched on, the jumper must be fitted
40
and the code transmitted using the key
transmitter so the key can be stored in
the receiver’s memory. After the
jumper is removed, the receiver should
switch its outputs when the key transmitter is activated. It may be necessary to disconnect the receiver
antenna for this procedure and activate
the transmitter close to the receiver.
This will reduce the effect of any interference signals present at 433.92 MHz.
The watchdog timer is also enabled in
the microcontroller on the receiver end.
If the program counter is affected by a
noise pulse on the supply line, for
instance, which can lead to a ‘hung’
system, the microcontroller will be
reset automatically after approximately
1 second and the receiver will execute
a normal start-up.
Data slicer and Manchester coding
The data slicer in the receiver is built
around an opamp in the receiver IC
that is intended to be used for this purpose. A data slicer is simply a comparator with a reference level that is automatically set when a data signal with
a duty cycle of 50% is received. It is
thus important that the duty cycle of
the data signal is in fact 50%. Manchester coding is used to achieve this result.
The circuit of the data slicer is shown
in Figure 3. The time constant of R
and C is large relative to the data bit
rate that is used. As a result, the DC
level at the inverting input is equal to
the average value of the received signal. With a 50% duty cycle, it will thus
be exactly halfway between the minimum and maximum values. However,
a certain amount of hysteresis is desirable because the receiver also generates noise when no signal is present or
the signal is relatively weak. This is
provided by Rl and R2.
It is also necessary to have the transmitter send a preamble before the
actual data. The preamble consists of
a series of clock pulses with a duty
cycle of 50% and a period that is much
shorter than the RC time constant of
the data slicer. This ensures that the
operating point of the data slicer is set
properly. Manchester coding must be
used to achieve the required 50% duty
cycle with the signal used here. This
essentially amounts to EXORing the
data with a clock signal having the
same period as the bit interval.
The original bit stream is then recovered in the same manner by EXORing
R1
R2
RSSI
DATA OUT
R
C
050364 - 13
Figure 3. Design of the data slicer in the
MAX1473.
the received data with a clock signal.
A drawback of Manchester coding is
that it increases the bandwidth of the
transmitted signal by a factor of 2, but
this is not a problem with the low bit
rate used here.
In theory, the bytes output by the
decoding logic when the clock is high
are the same as the bytes output by
the decoding logic when the clock is
low. Both bytes are in fact decoded by
the receiver software. The second byte
should essentially be a copy of the first
one. The correctness of the timing is
monitored closely by comparing the
two bytes after each set of eight
decoded bits.
Timing
The most important aspect with regard
to the algorithms used in the software
at both ends (transmitter and receiver)
is proper timing. The transmitter must
generate a bit stream with a certain
frequency, and the receiver must be
able to recover it with sufficient accuracy. For this reason, the key transmitter sends the receiver a preamble consisting of clock pulses at the same frequency as the bit rate before it sends
the coded 24-bit key. The preamble is
followed by a start bit that enables the
receiver to deduce that the preamble is
finished and the actual data will follow.
The ATtiny15L microcontrollers at each
end run on their internal RC clocks. A
calibration byte for the clock generator is
programmed into each microcontroller
by the manufacturer. However, this calibration value is intended to be used
with a supply voltage of 5 V. As the
transmitter operates at 3 V, its clock frequency is lower than it would be at 5 V.
For this reason, an extra provision has
been made in the software of the
receiver. With the receiver in learn mode
(as described below), the receiver
elektor electronics - 12/2006
Advertisement
copies the timing of the transmitter as well as its 24-bit
key. Both values are stored in the EEPROM of the microcontroller and used when the receiver is operating in the
normal mode.
Programming settings
The code for the ATtiny microcontrollers in the transmitter and receiver is available free of charge on the
Elektor Electronics website under item number
050364-11 (see month of publication). The microcontrollers must be programmed using a suitable programmer before they are fitted to the circuit boards.
When you program the microcontroller, don’t forget to
read the oscillator calibration byte from the signature
memory of the microcontroller and reprogram it in flash
memory location 1023 ($3FF). It is also important to tick
BOD as enabled and set the BOD level to 4.0 V for the
receiver or 2.7 V for the transmitter. If BOD is not programmed, the content of the EEPROM memory (which
contains the key code) can be affected when power is
switched on or off, with the result that things may no
longer work properly.
The ‘very quickly rising power’ fuse settings (CKSEL =
11) must also be programmed for the transmitter and
the receiver. This is partly related to proper operation of
the watchdog timer.
34 Channels sampled at 500 MHz
Initial use
After you have fully assembled the transmitter board,
you must first activate it so it can generate and store a
code. It’s a good idea to do this a couple of times to be
sure that a 24-bit code has been generated and stored.
Next, fit J1 on the receiver board. Then switch on the
receiver, but do not connect any antenna yet (to avoid
picking up interference). Hold the transmitter close to
the receiver board and activate it a couple of times.
Now remove J1 while the receiver is still on and reset
the receiver (by switching it off and on) without J1. The
supply voltage must drop to nearly zero during this
process, so watch out for the effects of any large capacitors in the power supply. If everything goes well, the
LED connected to OUT1 of the receiver will now light
up for around 1 second each time you press the transmit button. To be on the safe side, hold the transmit
button of the receiver pressed for at least 1 second
each time you activate the transmitter.
If everything is working the way it should, you can connect the receiver outputs to the device or circuit to be
operated. Bear in mind that it may be necessary to connect a buffer stage or semiconductor relay between the
receiver and the load.
(050364-1)
[1] Crystals used in this project:
www.hongkongcrystal.com
Tx: 9SMI356000E03FAFZOOO
Rx: 9SMI322560E03FAFZOOO
Coil data
L1 (transmitter): 3 turns 0.3 mm dia. silver-plated copper wire, coil diameter 2.5 mm, length 5 mm
L1 (receiver): see Figure 2.
L2 (receiver): see Figure 2
12/2006 - elektor electronics
Professional Features – Exceptional Price
Sophisticated Multi-level Triggering
Transitional Sampling / Timing and State
Connect this indispensable tool to your PC’s
USB 1.1 or 2.0 port and watch it pay for itself within hours!
Economical
2-day
Shipping!
500 MHz Max.Timing Mode
Sample Rate (internal clock)
Interpreters for I2C, SPI and
RS232 included
200 MHz Max. Sample Rate
State Mode (external clock)
Integrated 300 MHz Frequency
Counter
Multi-level Triggering on Edge,
Pattern, Event Count, Group
Magnitude/Range, Duration etc.
+6V to -6V Adjustable Logic
Threshold supports virtually all logic
families
Real-Time Hardware Sample
Compression
Full version of software free to
download
Qualified (Gated) State Mode
Sampling
Mictor adapter available
www.pcTestInstruments.com
Visit our website for screenshots, specifications and
to download the easy-to-use software.
Intronix Test Instruments, Inc.
16421 N.Tatum Blvd. Unit 130
Phoenix,Arizona 85032 USA
Tel: (001) 602 493-0674
Fax: (001) 602 493-2258
www.pcTestInstruments.com
* The amount of £210 was calculated at the time this ad
was created.The actual price will depend upon the
exchange rate at the time of purchase.
Price does not include the UK VAT of 17.5% plus a small
import fee of typically £5-10.These fees will be collected
at the time of delivery (companies can claim a refund for
the 17.5% VAT from HMRC). E&OE
HANDS-ON HOME & GARDEN
Go with the Fl
F
Jürgen Maiß
In this day and age enjoying
constant hot water should
not be an expensive luxury.
All the same, the way some
households squander energy
to provide this on tap makes
you wonder. Our novel circuit
saves both your bank
balance and the environment!
The circulation pump in a heating system is the device that makes possible
the accustomed luxury of having hot
water at the turn of the tap. Unavoidable heat losses in the plumbing system and the constant operation of the
circulation pump lead to profligate
waste of energy, however. In fact,
according to Oxford University’s Environmental Change Institute, the pump
consumes as much energy as lighting
or the fridge in most households, and
a pan-European drive is now underway to find ways of saving costs for
what is one of the most ignored electrical devices in the home. To cut
power consumption we need to lag all
pipes well and reduce the amount of
time that the pump operates to only
when hot water is actually required.
There are already a variety of control
systems on the market for managing
circulation pumps. One of these relies
on flow sensors, which react adequately fast but are hardly simple to
install (they require plumbing into the
pipe system). Other devices observe
42
temperature variations. These determine the temperature difference
between the hot water and circulation
pipes, using any variation from the
desired value as the trigger. The systems are both reliable and very simple
to install, since no ‘invasive action’ is
necessary to the plumbing. Unfortunately the sluggish action of the temperature sensors means they do not
operate very rapidly. One feature that
both systems share is their relatively
high purchase cost. Our approach here
takes a rather different route. A simple,
readily available piezo sounder used as
an acoustic emission sensor detects
the sound of flowing water in the
plumbing system. This is what we use
as the switching trigger.
Piezoelectric effect
Piezoelectric sounders, also known as
electronic buzzers, are used in many
electronic circuits as audible transducers or ‘bleepers’. Easy to use, they also
take up little space and are extremely
affordable. A change in the voltage
low
Piezoelectric transducer
controls domestic water system
+5V
P
D6
R8
100k
R3
R6
100k
C1
C3
C5
6
IC2
10n
2
4
2
1
SWIN
IC4
C
3
R15
GND
100n
33n
20k
GP0
D4
7
N
DETECT
8
PIC12C675P
SCAN
R14
R11
1k
10k
R10
1µ
060099 - 11
+5V
78L05
1N4007
D2
6
D1
IC5
+9V...+12V
M
5
4
5k1
C8
100k
C7
M1
PUMP ON
GP1
R17
R12
C4
D3
3
GP3/MC
7
VSS
8
P2
100n
R
LM2907
RC
P1
100k
C2
4
5
GP2
2
LTC1049
5k6
1M
10k
R5
FIN
230V
BC550
GP5/CIN
1µ
4
LTC1049
R7
R2
1
10k
GP4/COUT
VCC
C6
6
IC3
10n
4
LTC1049
R4
7
3
2
6
30k
2
7
3
6
IC1
T1
R16
IC6
7
3
47n
1k
R9
5k1
R13
1M
10k
SENSOR
100k
RE1
51k
R1
C11
C10
C9
10µ
25V
100n
100n
Figure 1. The pump control comprises five capacitor-coupled stages.
applied to the piezo crystal in the
sounder deforms it slightly, an effect
that is used to generate sound. The
principle is reversible too, enabling the
transducer to work as a highly sensitive ‘microphone’ as well. The variations in voltage charge resulting from
altered mechanical stress on the crystal
structure can be registered electronically and processed further. Piezoelectric sensors are widely used today in
workshop tools, for automotive applications and in technical setups as
pressure, power and dynamic
accelerometer devices.
The pump control system shown in
Figure 1 comprises several modules,
each with their own function.
• The
first is a preamplifier stage
with an amplification factor of from 10
to 1,000. The signal coming from the
sensor has a value in the region of 100
to 300 µV and is delivered via coupling capacitor C1 to the non-inverting input of IC1. The amplification of
this stage is set by trimpot P1. The
12/2006 - elektor electronics
extremely low-noise precision amplifier LTC1049 (from Linear Technology)
functions not only as an amplifier but
also as an impedance adapter with a
high-Z input resistor to avoid loading
the sensor signal. Coupling capacitor
C3 takes the now boosted sensor signal to a second amplifier (once more
type LTC1049 with an amplification
factor of 10). Total amplification is
defined as A = [(R3 / (R4+P1)) +1] x
[(R6 / R5) +1].
• The signal, now amplified to around
1.0 V is fed via C5 to a comparator
stage. The switching threshold for IC3
(yet again a LTC1049) is set in the
region of around 0.5 to 1.55 V by R8,
R15 and trimpot P2. If the sensor
detects a flow signal, a squarewave
output voltage of approximately 1 kHz
appears at the output of IC3.
C7R12) delivers a High output level,
which —
• provides the trigger signal for the
PIC16F675 microcontroller that follows.
As each of the switching elements is
connected by coupling capacitors, they
can all be modified easily for other
applications too. The amplifier circuitry
would, for instance, make a magnificent microphone amplifier, ultrasonic
amplifier or ground motion detector
(geophone). Linear Technology provides for its own modules a free and
easy-to-use simulation tool called
SwitcherCadIII. A simulation program
for this project is contained in our Project Software archive file 060099-11.zip
which can be downloaded free of
charge from www.elektor.com.
• This squarewave signal is taken via
Soft- and hardware
C6 to the frequency-to-voltage converter LM2907 (IC4). This module
operates in ‘speed switch-mode’ and
above the frequency set as f = 1 / (2
The software controlling the pump is
relatively straightforward; the program
can be modified according to requirements and the particular circuit appli-
43
HANDS-ON HOME & GARDEN
cation. After applying supply voltage
the system is initialised, with all three
LEDs activated sequentially. The flashing of SCAN indicator LED on Pin 7 of
the Controller indicates that monitoring is in progress. A High output produced by the analogue signal processing section is recognised by the microcontroller at pin 4 and activates the
FLOW DETECT indicator LED on pin 6.
False triggering is excluded to a large
extent automatically as the Controller
waits about three seconds for the High
signal to be repeated. If this is the
case, then the FLOW DETECT and
PUMP ON indicator LEDs are activated
for 50 s, together with circuit’s output
pin 3 and consequently the relay as
well. Turning on the hot water tap
briefly (3 s) is sufficient to put the circulation pump into operation. Switching times should be adjusted in the
software according to the length and
other characteristics of the plumbing.
To facilitate this, the Project Software
includes the source code in BASIC as
well as the HEX data for programming
the Controller.
Construction and installation
As the header photo shows, the author
built the hardware using SMD technology, to enable everything to fit inside a
small case. However, this is not
absolutely necessary, after all the microcontroller comes in a normal DIL enclosure for easy reprogramming. Transparent silicon rubber (as used around sinks
and showers) is ideal for fixing the sensor to the back of the enclosure, three or
four small fixing pads on the edge of the
sensor being entirely adequate. Figure 2 indicates how the little device can
be clamped to the hot water pipe with
a couple of cable ties.
The Controller switches the circulation
pump on and off using a transistor
driver stage and a suitable relay. The
relay is connected (looped) into the
Live (L) wire of the pump as shown.
Since we are dealing with mains voltage here all connections to the relay
must be protected from accidental
touch contact and installed according
to the IEE Wiring Regulations, which
are the national standard to which all
domestic and industrial wiring must
conform in Britain.
Both potentiometers should be
adjusted to their middle setting the
first time the circuit is installed, so as
to give an amplification factor A of
around 500 (trimpot P1) and a signal
switching threshold of around 1.0 V
44
Figure 2. Two cable ties clamp the enclosure and sensor (glued onto it) to the hot water pipe.
(trimpot P2) for the signal amplitude.
The final task is to turn on the hot
water tap and adjust the overall signal
amplification factor of the system at P1
so that the circuitry recognises the
noise of water flowing reliably (maximum amplification with the trimpot
turned fully to the left). If required, the
amplitude switching threshold can
also be increased or reduced.
(060099-I)
Links:
Data sheet LTC 1049:
www.linear.com/pc/downloadDocume
nt.do?navId=H0,C1,C1154,C1009,C1
027,P1189,D3403
Simulation program:
http://ltspice.linear.com/software/
swcadiii.exe
Data sheet LM2907:
www.national.com/pf/LM/LM2907.ht
ml#Datasheet
University of Oxford/European
Union SAVE II project:
www.eci.ox.ac.uk/lowercf/
eusave_circulation.html
IEE Wiring Regulations:
http://en.wikipedia.org/wiki/Electrical
_wiring_(UK) and
www.iee.org/Publish/WireRegs
elektor electronics - 12/2006
This Issue:
24-PAGE ‘E-TRIXX’ SUPPLEMENT
escaped from the Elektor kitchen
Next Issue:
FREE CD-ROM FROM MICROCHIP/LABCENTER/ELEKTOR
MPLAB — C30 Compiler — EXPLORER-16 — Proteus VSM for PICs
Want to know everything each month?
SELECT YOUR OWN ANNUAL SUBSCRIPTION
AND RECEIVE A FREE 1W LUXEON LED TORCHLIGHT*
Fill out the Order Form with this issue today!
Subscription rates and conditions may be found at the back of this issue.
* Offer available to Subscribers who have not held a subscription to Elektor
Electronics during the last 12 months. Offer subject to availability.
CONTENTS
Did someone ring?
4
A touch more bling, perhaps?
5
Empty battery?
6
Lord and master!
7
Battery saver circuit
7
Battery indicator for the caravan
8
When the siren sounds…
9
Temperature-controlled switch
10
Check your contacts
11
Check Out Your LEDs
12
It’s Wet!
13
Dicing with LEDs
14
Surf simulator
15
Save Your Ears
17
Electronic poltergeist
18
Pump it up: MP3 booster
19
collection
Circuits Allsorts
This December 2006 issue of Elektor Electronics comes
with a free collection of simple yet useful and sometimes
playful circuits for home construction. The circuits, we hope
and expect, are easily understood and reproducible and
should appeal to the less advanced electronics enthusiast,
although more experienced readers are also bound to
find interesting bits and bobs to help them through the
dark winter evenings in a pleasant and educational way.
A deafening siren that will make burglars run off
— electronic dice adds fun to games — a lighting house
number helps your friends and acquaintances find your
home more easily in the dark. Do you want to stand out
from the crowd in the local disco, or encourage annoying
visitors to leave early, with a bit of help from an electronic
poltergeist? All these, and more, circuits may be found
in this 24-page i-TRIXX supplement, which is offered to
you free of charge.
All i-TRIXX circuits in this supplement originate from
the Elektor labs and are spin-offs from larger projects,
scribblings on the back of envelopes, dead-bug fiddling,
‘quick and dirty’ solutions, or even design ideas that
eventually became so intriguing the designer just had
to develop it out for enthusiasm, curiosity and technical
satisfaction — all at the same time in not a few cases.
In this respect, i-TRIXX are truly ‘tricks of the trade’ that
got pencilled down and eventually — with the help of
our editorial team — made it to this publication in print
instead of just ending up in a drawer or the more
contemporary Windows Recycle Bin.
From over 30 years experience in publishing for electronics
enthusiasts all over the world we know that the Christmas
holidays are a great time for fun circuits that cost next to
nothing to build, often from parts found in the junkbox
(now’s a good time to clean it out!). Especially newcomers
to the hobby will find i-TRIXX useful to learn about the
process that begins with the ability to ‘read’ a circuit
diagram and culminates in powering up a fully working
prototype.
Have fun with the i-TRIXX collection!
Musical saw
20
Luminous house number
22
Applause generator
23
Jan Buiting, Editor
Did someone ring?
If you’re expecting an important visitor but you just have to step out for a
moment, an electronic doorbell memory can come in handy so you can see
whether someone rang while you were out. Of course, you can’t tell whether
it was the visitor you were expecting who dropped by then, but a call to the
mobile phone of the person concerned can quickly answer that question. A
doorbell memory can also save you the trouble of going to the front door (if you
live upstairs) when you think you heard the bell but aren’t sure. And if you can’t
buy one, then of course you can build one yourself! Read on to find out how.
Doorbell memory
It takes only a handful of electronic components to
build a handy tale-tale with an LED that indicates
whether someone pressed the button of your doorbell.
How many times have you thought you heard your
doorbell while watching television in the evening? The
sound of the well-known ‘ding–dong’ chimes occurs all
too often, especially during the many commercials that
nowadays remind us at the most inconvenient times
that the gripping film we’re watching is only a fantasy.
A glance at the LED of the doorbell memory will tell
you whether you have to go to the door or can try to
escape the ads by zapping to a different channel. Or
if you’re expecting someone but have to make a quick
trip to the neighbours to borrow a few beers for the
occasion, it can be handy to be able to see whether
your visitor already arrived while you were out. If so,
you can always call him or her on the mobile to confess
that you hadn’t properly prepared for the expected
visit.
BC547B
BC557B
+3V
R2
C
E
B
D1
T2
S1
100k
BC557B
T1
R1
1mA
470k
1N4002
BC547B
2x 1V5
R3
doorbell
BT1
D2
47k
C1
R4
1k2
The circuit is as simple as it
is effective. It is connected
in parallel with the bell and
powered by a 3-V supply
formed by two 1.5-V penlight
batteries connected in series.
The doorbell memory draws
so little current that a set of
batteries will last several years
in normal use.
The circuit works as follows.
When the supply voltage
is switched on with switch
S1, capacitor C1 (initially
uncharged) prevents
transistors T1 and T2 from
conducting. LED D2 is off,
and the memory is armed. When the doorbell button
is pressed, the memory circuit receives an AC or
DC voltage via diode D1, depending on the type of
doorbell. It can handle either type. Transistor T1
thus receives a base current, so it starts conducting
and drives T2 into conduction. The LED lights up as
an indication that the doorbell has rung (i.e. was
energised). The combination of transistor T2 and
resistor R3 keeps T1 conducting after the bell voltage
goes away (when the button is no longer pressed).
The memory remains in this state until switch S1 is
opened. This switch thus acts as a reset switch as well
as a power switch.
The circuit can be assembled compactly on a small
piece of perforated prototyping board, so it can be
fitted into just about any model of doorbell. The
transistors can be replaced by other, equivalent types
as long as you use a combination of NPN and PNP
types.
10n
051001 - 11
Electronics inside out !
4
i-TRIXX collection - 12/2006
C2
100n
P1
74HC4060
1
Q12
2
Q11
Q12
Q9
15 Q9
Q13
3
Q13
Q7
14 Q7
Q5
4
Q5
Q8
13 Q8
Q4
5
Q4
R
Q6
Q3
6
Q6
7
Q3 PO PO
8
PI
GND
100k
BT1
12 RESET
11 PI
10
C1
3V
9
11
47n
RX
4
5
6
RCX
CT
PO
7
8
9
1M
11
10 PO
9
3
+
R5
12
12
13
CT=0
8
D1
BAT85
!G
CX
R1
8k2
16
CTR14
10k
Q11
R6
V
DD 16
IC1
7
5
4
6
14
13
15
1
2
R2
D2
3k9
BAT85
R3
D3
Figure 1. A binary
counter (IC1) raises the
current through LED
D5 from zero to the
maximum value in 16
steps.
2k2
BAT85
R4
D4
1k
BAT85
3
D5
74HC4060
051002 - 11
Want to attract attention at bling-bling parties? That’s not so easy, since all the party animals there
are doing their best to grab attention. The more bling, the more looks you get. An electronic disco
brooch with a cool skin as an eyecatcher can certainly help. If you make it yourself, your chances of
success lie entirely in your own hands. And it’s certainly not all that difficult!
A touch more bling, perhaps?
Disco brooch
The disco brooch described here is a small but
attention-getting bit of electronic circuitry. It lights
up a LED in steps until it reaches maximum brightness
and then goes out, after which the entire process
repeats itself, so you can attract attention at discos
and parties. If you solder all the components on a small
piece of perforated prototyping board, possibly along
with a small battery, you can keep the whole thing nice
and compact. We’ll come back to that later.
First let’s have a brief look at how the circuit works.
It is built around IC1 (see Figure 1), which is a binary
counter chip. A digital pattern of ones and zeros
appears at its outputs. Only the outputs on pins 4 to
7 are used in this particular case. They thus generate
a count from 0 (binary 0000) to 15 (binary 1111). All
outputs are zero at the start, and the LED is dark.
Resistors R1–R4 are dimensioned such that the current
flowing through LED D5 increases in steps to its
maximum value during the following 15 states of the
counter. This all happens at a rate that is controlled by
the RC network connected to pins 9–11 of the IC. At
the end of the cycle, the counter goes back to zero and
the LED goes dark. You can use trimpot P1 to adjust
the duration of each of the steps. The maximum length
of the full step cycle is 1.5 seconds, and the minimum
Figure 2. Fit the
components as close
together as possible on
a piece of perforated
circuit board to keep the
finished circuit compact.
i-TRIXX collection - 12/2006
length is 0.14 seconds. At the minimum time setting, it
appears that the LED just blinks.
Diodes D1–D4 isolate the individual outputs of IC1 so
they do not interfere with each other. They are Schottky
diodes, which have a lower forward voltage drop
than normal diodes. As a result, there will be enough
voltage to provide adequate current through resistors
R1–R4 (and thus through the LED) even when the
battery voltage is low. Of course, the brightness of the
LED still depends on the battery voltage. In practice,
a battery voltage of 3 V gives very nice results. This
voltage can be provided by a button cell, since the
maximum current at this supply voltage is only around
2 mA. If you use a suitable battery holder designed for
PCB mounting, you can tuck the button cell away nicely
on the circuit board (see Figure 2).
Of course, you also have to be able to switch the
brooch on and off. You can use a safety pin for this
purpose, and at the same time it can do what it is
designed for, which is to attach the brooch to your
clothing. To prepare the safety pin for this purpose, cut
a piece out of the fixed leg. That represents the opened
switch. The easiest way to do this is to first solder
the pin in the proper position on the circuit board
(see Figure 3). Use a bit of sandpaper to roughen the
surface of the pin first so the solder will stick better.
After the pin is soldered securely in place, cut out a
piece in the middle and your DIY switch is ready to use.
Figure 3. You can make
an on/off switch from a
safety pin, which also
serves to pin the brooch
to your clothing.
5
Empty battery?
Is the battery empty, or is there something wrong with the device? That’s always a difficult
question when your walkman or some other battery-powered device appears to be dead when
you switch it on. Before you take it to the shop for servicing, the first thing you should do is to
test the battery or batteries. Of course, this means you need a reliable battery tester, but it also
means you can limit the damage to the cost of a battery or two and a one-time investment of
time and money in building a suitable tester.
Photo: www.energizer.com
Battery tester
Many commercial battery testers consist of nothing
more than a resistor, a simple little meter and a
pushbutton. Some manufacturers include an even
simpler tester with a set of batteries, consisting of a
strip of plastic with a layer of some sort of electrically
conductive material that changes colour when a
current flows through it. If you press this strip over the
battery between the positive and negative terminals, a
fully charged battery will cause a more intense change
in colour than a partially discharged battery. Naturally,
tests of this sort do not provide especially reliable or
accurate results.
The idea behind the circuit described here is to load a
single battery, a set of batteries connected in series,
a rechargeable battery, or even a small button cell
with a reasonably constant current and use a separate
multimeter or voltmeter module (M1) to check the
voltage. A quickly decreasing voltage indicates that
the battery or batteries will have to be replaced soon.
BD139
E
B
BF245
BC639
BC640
G
E
C
D
B
C
S
S1
C1
R1 T2
D1
22 Ω
100n
T3
T1
T4
2x
BC639
T5
4k7
R2
BD139
The active constant-current element is transistor T1.
The current through it is held constant by comparing
the voltage across resistor R1 in its collector path with
a relatively constant reference voltage across diode D1.
This comparison is provided by differential amplifier
T3/T4. The voltage across diode D1 (a Schottky type) is
reasonably constant by nature, but it is also stabilised
by using FET T5 as a simple constant-current sink.
T5 also limits the current at relatively high voltages
(with several batteries in series). The constant voltage
across D1 is transferred to resistor R12 by differential
amplifier T1/T2, so a constant current flows through
R1 from the battery or batteries being tested. R1 has
a relatively low resistance, so this current is larger
than the current drawn by the rest of the circuit. The
quiescent current, which incidentally is also reasonably
constant, is thus negligible. The test current thus
remains reasonably constant while the battery or
batteries is/are being tested.
The maximum battery voltage that the tester can
handle is set by T5, and here it is 30 V. To ensure that
T1 does not get too warm at high battery voltages,
keep the test as short as possible. Use a pushbutton
switch as a test switch so the battery being tested
cannot be left under load by accident.
BAT85
BC640
M1
BT1
If a constant-current circuit is used for the load, the
current can never too be large and there is no need
to make an adjustment for the number of cells. The
constant-current circuit is specially designed to work
with a voltage as low as 0.9 V. It’s quite difficult to
make a circuit work at even lower voltages with normal
transistors.
051003 - 11
BF245B
In these days of environmental awareness
it has become desirable to use batteries
as efficiently as possible. A lot of battery
powered electronic devices do not switch
themselves off automatically, such as
multimeters. Forgetting to turn these off isn’t
just a nuisance (since you’ll find that you don’t
have a spare battery at the crucial moment),
but you’re also unnecessarily reducing the
lifespan of the batteries and causing extra
pollution. And it isn’t really necessary for
those electronic thermometers to display the
temperature day and night. As a skilful and
environmentally aware i-TRIXXer you should
make one or more of these small battery saver
circuits.
Electronics inside out !
6
i-TRIXX collection - 12/2006
S1
BC557B
Photo: Tratec Telecom BV – Veenendaal NL
1M
R1
T1
BT1
BC557B
C
R2
E
B
100 Ω
3V
D1
R3
D2
BP
10k
BP104
10
4
051004 - 11
Lord and master!
Men in particular enjoy the convenience of television remote controls – often to the annoyance of
their female partners. Men apparently want to know what they’re missing when the TV is tuned to
a particular programme, so they like to keep zapping to other channels. With the remote control in
their hands, they feel like they are the lord and master of the TV set. They are thus completely at a
loss if the remote control doesn’t work properly. There are many reasons why a remote control unit
can malfunction, such as defective IR receiver in the TV set, a defect in the remote control, or empty
batteries. Here a tester that can determine whether the remote control unit still emits an IR signal can
come in handy. If you want to keep the IR reins firmly in hand, you can build your own IR detector.
IR detector
If you have a few remote control units around the
house, you’ll appreciate this little circuit. The LED
clearly indicates whether the remote control unit
actually emits an IR signal when you press one of the
buttons on the unit.
The circuit uses a photodiode (D1) to sense the infrared
light emitted by the remote control unit (if it is working
properly). The plastic package of this diode acts as an
IR filter that is only transparent to invisible light with
a wavelength of 950 nm. Although there are probably
some remote control units that use IR diodes operating
at a different wavelength, the circuit has enough sensitivity to detect them as well.
If enough light falls on photodiode D1, an electrical
current will flow through the diode. In fact, what happens is that the leakage current increases, since photodiodes are usually operated in reverse-biased mode (as
is the case here). If the current is large enough, transistor T1 conducts and causes LED D2 to light up. If LED
D2 remains dark, this means the remote control unit is
not producing any IR light. This can be due to an empty
battery (or batteries) or a fault in the internal circuitry.
Pay careful attention to the polarisation of the photodiode when wiring it into the circuit. The cathode is
clearly marked by a special pin. For LED D2, use a lowcurrent type that can handle a current of at least 7 mA.
The detector can be powered by a pair of 1.5-V penlight
cells connected in series.
Battery saver circuit
i-TRIXX presents a small electronic switch that
connects a battery to the equipment for a certain
amount of time when a push-button is momentarily
pressed. And we have also taken the ambient light
level into account; when it is dark you won’t be able to
read the display so it is only logical to turn the switch
off, even if the time delay hasn’t passed yet.
The circuit is quite straightforward. For the actual
switch we’re using a well-known mosfet, the BS170.
A mosfet (T2 in the circuit) used in this configuration
doesn’t need a current to make it conduct (just a
i-TRIXX collection - 12/2006
voltage), which makes the circuit very efficient. When
the battery is connected to the battery saver circuit
for the first time, capacitor C2 provides the gate of
the mosfet with a positive voltage, which causes T2
to conduct and hence connect the load (on the 9 V
output) to the battery (BT1). C2 is slowly charged
up via R3 (i.e. the voltage across C2 increases). This
causes the voltage at the gate to drop and eventually
it becomes so low that T2 can no longer conduct,
removing the supply voltage to the load. In this state
the battery saver circuit draws a very small current
of about 1 µA. If you now press S1, C2 will discharge
7
Battery indicator for the caravan
R4
220k
T3
G
R5
330 Ω
BT1
T2
D
S
BF245A
BC547B
12V
T1
47k
R2
2x
BC547B
C
E
B
061005 - 11
The setting of T1 and T2 determines whether LED D2
will light up when the battery voltage drops below
a certain level. Junction FET T3 is used as a current
source in order to try to keep the current through the
LED as constant as possible. In this way the indicator
remains lit even when the battery is in a state of very
deep discharge (< 4 volt). The LED is a good lowcurrent type that is still very bright at a very small
current (1 to 2 mA). Voltage divider R1 and R2 has
been calculated such that T1 will start to conduct when
the voltage of the battery is greater than 12 V. If you
think this threshold is too high (or: if you think that you
can still start your car with a lower battery voltage),
then you can reduce the value of R2 or replace it
with a 50-k preset (connected as an adjustable
resistor). When T1 conducts, the base current to T2 is
interrupted and the collector of T2 will become high
through R4. In this state T3 does not conduct and
the LED is off. When the battery voltage drops below
12 V, T1 will block and T2 will start to conduct. R5 is
now connected to ground via T2 which turns T3 into
a current source of about 2 mA that drives the LED.
There is, of course, a transition region during which
and the circuit returns to its initial state, with a new
turn-off delay. Resistor R5 is used to limit the discharge
current through the switch to an acceptable level.
You only need to hold down the switch for a few
hundredths of a second to fully discharge C2.
In our prototype, connected between a 9 V battery
and a load that drew about 5 mA, the output voltage
started to drop after about 26 minutes. After 30
minutes the voltage had dropped to 2.4 V. You should
use a good quality capacitor for C2 (one that has a
very low leakage current), otherwise you could have to
wait a very long time before the switch turns off!
The ambient light level is detected using an LDR
(R1). An LDR is a type of light sensor that reduces
in resistance when the light level increases. We
recommend that you use an FW150, obtainable from
the current through the LED slowly increases; after
all, T1 and T2 do not switch with infinite gain! In our
prototype the LED changed from fully off to fully on
at a voltage variation from 12 to 11 V. As a bonus, a
partially illuminated LED gives a rough indication as to
how much the voltage actually is. Diode D1 prevents
the circuit from inadvertently giving up the ghost if the
circuit is connected incorrectly to the battery (reverse
polarity).
In practice, because of variations in the specifications
of the transistors, the threshold and the current level
through the LED can be different. Test the circuit
thoroughly before using it. If you want a brighter
indicator, you can increase the current through the LED
by replacing T3 with a BF245B or BF245C.
When the LED is off, the current through the circuit
is barely 30 µA at a battery voltage of 14.4 V. With
the LED is on and at a battery voltage of about 10 V,
the current consumption is about 2 mA. Even with an
illuminated LED, the circuit is not likely to be the cause
of a flat battery. Even a good quality battery will have
a selfdischarge rate which is many times greater than
the maximum current consumption of this circuit!
BC547B
S1
R2
C
E
B
C2
100µ
25V
R5
100 Ω
R3
1M
1M
R1
This i-TRIXX circuit can prevent a whole lot of trouble
for those of you who go on holiday in a caravan.
It would be a significant damper on your holiday
spirit when you are ready to leave the camping and
discover that you have used your battery too much
and that you are now unable to start the car. This
annoyance can be avoided if you were warned early
enough by an illuminated LED when the charge in the
battery threatens to become too low. A quiet, out of the way, in the countryside
camping is what modern people look for to be able to unwind. However, we do
not want to be completely deprived of all our creature comforts. We don’t cope
very long without electric light or a TV! And in the absence of a mains power
outlet the car battery has to function as energy source, with the risk that later
on there will be too little left to start the engine. The little circuit presented
here gives you an early warning when the battery voltage (and therefore its
stored energy) threatens to become too low.
BT1
R4
9V
9V
T1
BS170
R1
C1
BC
547B
R3
10M
BAT85
BF245
100 Ω
D2
10M
D1
100n
G
T2
LDR
D
S
S
G
061003 - 11
BS170
D
Electronics inside out !
8
i-TRIXX collection - 12/2006
National Archaeological
Museum of Athens
When the siren sounds…
Siren
There are lots of different ways to make an electronic
siren. Here we use a binary counter (IC1) and an
analogue multiplexer (IC2, which is a digitally
configurable switch). The counter is a type 4060 IC,
which has an integrated oscillator. The oscillator
generates the tone of the siren. The frequency of this
tone depends on the resistance between pin 10 of IC1
and the junction of C1 and R9.
The trick here is that the analogue multiplexer adjusts
the clock rate of the counter depending on the state of
the counter. The frequency of the oscillator decreases
as the resistance between pin 10 of IC1 and the
junction of C1 and R9 increases, and the lower the
frequency of the oscillator, the longer the counter
remains in its current state. This means that high
frequencies present on pin 9 are generated for shorter
times than low frequencies.
The values of resistors R1–R8 increase in uniform steps
of approximately 10 kΩ, with the result that the output
on pin 9 is a series of eight decreasing frequencies, and
this series is constantly repeated in cyclical fashion.
Transistor T1 (BD139) and resistor R10 are included
to boost the signal from pin 9 to a level suitable for
driving a loudspeaker. The sound
produced by this circuit may be
familiar to some of our readers
(especially if their memories extend
back to older types of pinball and
arcade game machines).
You can also adjust the characteristics of the sound,
since this circuit is primarily an invitation to experiment
with the component values – in particular R1–R8
(10 kΩ minimum), but also C1. The values of R1–R8 do
not have to follow a strictly increasing series; they can
also be selected randomly.
The current consumption is primarily determined by
resistor R10 and the loudspeaker (in our case an 8-Ω
type). The siren circuit draws approximately
33 mA at a supply voltage of 9 V. The supply
current is 11 mA at the minimum supply
voltage of 4 V, and it increases to 60 mA at
the maximum supply voltage of 15 V.
+9V
C2
T1
100n
IC1
C1
The battery saver circuit can be added to devices
that use 6 or 9 volt batteries and which don’t draw
more than 100 mA. The circuit can be built on a piece
of experimenter’s board and should be made as
compact as possible so that it can be built into the
battery powered device.
i-TRIXX collection - 12/2006
BD139
3
CTR14
9
33n
10
11
R9
CX
4
!G
5
+
RX
RCX
6
7
CT
1M
8
9
11
12
12
CT=0
13
8
7
5
4
R10
BT1
9V
6
14
LS1
13
15
1
2
3
4060
BD139
+9V
C3
R1
e.g. Conrad as part number 183547-89. When
there is too little light its resistance increases and
potential divider R1/R2 causes transistor T1 to
conduct. T1 then charges up C2 very quickly through
R4, which limits the current to a safe level. This
stops T2 from conducting and the load is turned off.
The choice of value for R2 determines how dark it
has to be before T1 starts to conduct.
16
100 Ω
In Greek mythology, a siren was a demonic being
(half bird, half woman). Later on this idea was
transformed in art into a mermaid: a combination
of a fish and a woman. Mechanical and
electromechanical versions were invented even
later, and electronic models were developed in
the last century. Sirens are characterised by their
ability to produce sounds that attract attention.
With the exception of the flesh-and-blood models,
they are thus used to warn people in a particular
area of impending danger. The electronic versions
are the most suitable for DIY construction.
10k0
100n
R2
16
20k0
R3
30k1
R4
40k2
R5
49k9
R6
60k4
R7
69k8
R8
13
14
15
12
1
5
2
4
VDD
0
MDX
0
1
2
8x
3
IC2
0
7
1
2
4
5
6
COM
4051
7
G8
VEE
VSS
7
8
11
E
10
B
C
9
3
4051
4060
6
80k6
051007 - 11
9
Temperature-controlled switch
It sounds rather mysterious: a switch that is controlled by its ambient temperature.
All without the touch of a human hand, except for when you’re building this sort of
electronic thermostat.
The LM35 is available in several different versions. All
versions have a rated temperature range of at least
0–100 °C. One thing you may have to take into account
is that the sensor has a relatively long response time.
According to the datasheet, the sensor takes 3 minutes
to reach nearly 100% of its final value in still air.
The opamp has very low drift relative to its input
voltages, and in the low-power mode used here it
draws very little current. The sensor also draws very
little current, so the total current consumption is
less than 80 µA when LED D1 is off. The advantage
of low current consumption is that the circuit can be
powered by a battery if necessary (6 V, 9 V or 12 V).
The sensor has a rated operating voltage range of
4–30 V, and the TLC271 is rated for a supply voltage of
3–16 V. The circuit can thus work very well with a 12-V
supply voltage, which means you can also use it for
car applications (at 14.4 V). In that case, you must give
additional attention to filtering out interference on the
supply voltage.
The circuit uses a TLC271 opamp as a comparator. It
compares the voltage from the temperature sensor,
which is connected to its non-inverting input (pin 3),
with the voltage on its inverting input (pin 2). The
latter voltage can be set with potentiometer P1. If the
voltage from the sensor rises above the reference value
set by P1 (which represents the desired temperature),
the output of the comparator toggles to the full
supply voltage level. The output is fed to
transistor T1, which acts as a switch so
+5V
the output can handle more current. This
makes it possible to energise a relay in
order to switch a heavy load or a higher
C2
voltage. The transistor also supplies current
100n
to LED D1, which indicates whether the
temperature is above the reference value.
The reference value can be adjusted by
P1 over the range of 18–30 °C with the
IC2
indicated component values. Of course,
you can adjust the range to suit your
needs by modifying the value of R1 and/or
R2. To prevent instability in the vicinity
LM35
D1
R2
5V
R7 20mA max.
3k3
There are lots of ways to measure the temperature
of an object. One very simple way is to use a
semiconductor sensor, such as the National
Semiconductor LM35 IC.
This sensor is accurate to within 0.5 °C at 25 ºC, and
few other sensors can do better or even come close to
this level of accuracy. In the circuit described here, the
sensor (IC2) generates an output voltage of 10 mV/°C,
so the minimum temperature that can be measured
is 0 °C. At 25 °C, the output voltage of the sensor is
(25 °C × 10 mV/°C) = 0.25 V.
of the reference value, a small amount of hysteresis
is provided by resistor R4 so the temperature will
have to continue rising or falling by a small amount
(approximately 0.5 °C) before the output state
changes.
390k
There are a lot of handy uses for a thermally controlled
switch. If the temperature inside your PC gets too
high sometimes, the circuit can switch on an extra
fan. You can also use to switch on an electric heater
automatically if the room temperature is too low.
There are innumerable potential applications for the
thermostat described here.
0V30
P1
10k
R3
2
8
2
1
3
1k
6
T1
R5
6k8
5
1
BC547B
4
TLC271
0V18
3
7
IC1
R4
C1
820k
R6
10k
15k
R1
220n
LM35
BC547B
051006 - 11
TLC271
C
E
B
Electronics inside out !
10
i-TRIXX collection - 12/2006
Photo: Hirschmann Electronics BV
Check your contacts
Having good contacts is important – not only in your daily life, but also in electronics.
In contrast to social contacts, the reliability of electrical contacts can be checked
quickly and easily. Various types of continuity testers are commercially available
for this purpose. Most multimeters also have a continuity test function for electrical
connections. A simple beep helps you tell good contacts from bad ones. However,
in some cases the tester doesn’t produce a beep because it won’t accept contact
resistances that are somewhat higher than usual. Also, poorly conducting (and thus bad)
connections are sometimes indicated to be good. Here e-trix comes to your aid with a design
for a DIY continuity tester that helps you separate the wheat from the chaff.
Continuity tester
Many multimeters have a built-in continuity test
function. However, in many cases the resistance
necessary to activate the beeper when you are looking
for bad connections is just a bit too high. It can also
happen that the beeper sounds even though the
resistance of the connection is unacceptably high. This
circuit lets you adjust the threshold between bad and
good contacts to suit your needs.
The circuit is built around an operational amplifier
(IC1) wired as a comparator. The opamp compares the
voltage on its inverting input (pin 2) with the voltage
on its non-inverting input (pin 3). The voltage on pin 3
can be set using potentiometer P1, so you can set the
threshold between good and bad connections. When
test probes TP1 and TP2 are placed on either side
of a connection or contact to be tested, a voltage is
generated across the probes by the current flowing
though resistors R1 and R3, and it appears on pin 2
of the opamp. This voltage depends on the resistance
between the probe tips. If the voltage on pin 2 is lower
than the reference voltage on pin 3, the difference
R3
100k
C2
180k
R2
22 Ω
BC547
TLC271
100n
C
E
B
TP1 R1
1k
2
3
D1
TP2
5
1
R4
D2
BAT
85
P1
6
IC1
4
2M2
R5
10k
4k7
330n
7
TLC271
D3
C1
8
is amplified so strongly by the opamp that its output
(pin 6) is practically the same as the supply voltage.
This causes transistor T1 to conduct, which in turn
causes DC buzzer BZ1 to sound. This means that
the resistance of the connection being tested is less
than the threshold value set by P1, and thus that the
connection is OK.
By contrast, a bad connection will cause the
relationship between the voltages on the inputs of
the opamp to be the opposite, with the result that its
output will be at ground level. The transistor will not
conduct, and the buzzer will remain still.
To ensure that the opamp ‘toggles’ properly (which
means that its output goes to ground level or the
supply voltage level) when the difference voltage is
sufficiently large and does not oscillate during the
transition interval due to small fluctuations in the
difference voltage produced by interference, its output
is coupled back to its non-inverting input (pin 3) by
resistor R4. This causes any change on the output to be
passed back to this input in amplified form, with the
result that the detected difference voltage is amplified
(and thus boosted).
Diodes D1, D2 and D3 protect
the circuit against excessive
positive and negative input
R7
voltages that may come from
the connections or contacts
being tested. They also
ensure that the continuity
C3
BZ1
tester does not inject
12V
100µ
excessively high voltages
25V
into the item under test.
T1
BT1
Capacitor C1 suppresses
high-frequency interference.
BC547
9V
The circuit draws only a small
supply current, so it can
easily be powered by a 9-V
R6
battery.
25k
2x
1N4148
i-TRIXX collection - 12/2006
051005 - 11
11
Check Out Your LEDs
You have to admit that these tiny electronic lamps are
handy, and they last almost forever. Around 40 years after
Nick Holanyak developed the first LED, they have become
just about indispensable. Any self-respecting electronics
hobbyist always has a few in his junk box. But before you
use LEDs, it’s a good idea to check them out. With a LED
tester, you can even do it in the dark!
LED tester
S1
D1
2V7
0W4
D2
BT1
9V
This circuit can be used to test up to three LEDs at
once, connected in series. You can easily increase that
number by using a higher supply voltage. If you do
so, you should allow 2.7 V for each additional LED.
The Zener diodes are included in the circuit so it can
also be used to test one or two LEDs. Another benefit
of the Zeners is that even if one or more of the LEDs
are defective or connected with reverse polarity, the
remaining ones will light up normally. That makes it
easy to spot suspect LEDs. If you extend the tester to
handle more LEDs, you must add another Zener diode
for each LED position.
The test current that flows through the LEDs is held
reasonably constant by FET T1, independent of the
number of LEDs being tested. The FET is used as
a constant-current source to keep the circuit as
simple as possible. The drawback of this approach
is that the tolerance range of FET characteristics is
especially large. The type used here even has three
versions: A, B and C. We used the B version here so
the current through the LEDs can be adjusted using
D5
2V7
0W4
D3
D6
2V7
0W4
T1
LEDs are available nowadays in all shapes and colours.
There are types with clear, colourless packages, while
others have coloured plastic packages. Many modern
types of LEDs need less current than older types. Some
of them provide quite a puddle of light if you give them
a decent amount of current.
When you’re working with used LEDs from the junk
box, there’s a good chance that you can’t tell which
lead is which any more. (If the leads haven’t been
trimmed, the short lead is always the cathode lead
and the long lead is the anode lead.) If you use several
LEDs in a display where they all have the same current,
you naturally want all the LEDs to have the same
brightness. But that’s not always the case, even with
LEDs of the same type. To save yourself unnecessary
soldering work, it’s a good idea to check the LEDs out
first. That’s the job of the LED tester described here.
D4
BF245B
BF245B
P1
2k5
G
D
S
051009 - 11
potentiometer P1 over the range of 1–7 mA. If you
need more current, you can use a BF254C instead, but
then you will also need a higher supply voltage. For
example, you can connect two 9-V batteries in series or
power the circuit from a mains adapter.
However, some LEDs have a maximum rated current of
only 5 mA. You should thus always start testing at the
lowest current by setting P1 to maximum resistance.
You can easily see from the brightness whether you
need more current. If an LED does not light up, it
may be defective or connected the wrong way round.
Reduce the current to the minimum level before
reversing or replacing any LEDs.
If you label the polarity of the terminals on the LED
tester, you can easily mark the cathode and anode
leads of the tested LEDs. To make it easy to swap the
LEDs, you can use an IC socket as a test socket.
The selected Zener diodes were chosen to make the
tester suitable for red, yellow and green LEDs. Red
LEDs have a forward voltage of 1.6 V to 1.8 V. The value
for yellow LEDs is around 1.9 V, and with green LEDs
the forward voltage can be as high as 2 V. If you also
want to test modern blue or white LEDs, you will have
to replace the Zener diodes with types having a voltage
of 4.7 V or 5.1 V. The supply voltage will also have to
be increased accordingly – for example, by connecting
two 9-V batteries in series.
Electronics inside out !
12
i-TRIXX collection - 12/2006
It’s Wet!
Have you ever seen the stairs to one of the upper stories in your house turn into
a waterfall? Or maybe you’ve come home to find your aquarium fish trying to
swim across the carpet? For your sake, we hope not, because the consequences
are usually fairly dramatic. With a handful of electronic components, you can at
least ensure that you will be warned before you have to put on your waders.
Water alarm
probes, the input of IC1a is held low by R1 and the
output of IC1b is also low. The oscillator is not active
in this state. If moisture is sensed, the supply voltage
pulls input 1 of gate IC1a high via the conductive
water, causing the gate to start oscillating. Whenever
the output of IC1b is high, the tone generator built
around IC1c is enabled, and in turn it energises buzzer
BZ1. The net result is a periodic, intermittent beeping
tone.
It’s better to prevent water problems than to have to
correct them. But no how many precautions you take,
an occasional leak can still happen. A burst water
supply hose for the washing machine, a bath tap that
someone forgot to turn off, a broken aquarium wall,
or a leaking boiler or central heating tank – anything
is possible. In such cases, it’s nice to be warned as
quickly as possible, for example by an acoustic water
alarm. Then you can at least limit the damage.
If you’re handy with a soldering iron and know the
difference between an IC and a PC, you’ll no doubt
enjoy building the electronic water alarm described
here.
You can adjust the intermittent effect of the sound
produced by the water alarm to suit your taste by
simply adjusting the value of R2 or C2. You can also
set the pitch of the sound with P1. The closer the pitch
is to the resonant frequency of buzzer BZ1, the louder
the tone will be. You should set the sound to the most
irritating level possible.
Gate IC1d is used to boost the amount of power than
can be pumped into the buzzer. It inverts the output
signal from IC1c to double the voltage applied to the
buzzer.
The circuit takes advantage of the fact that ‘normal’
water is always slightly contaminated, even if only
slightly, and thus conducts electricity to a certain
extent. It is built around an popular IC from the
somewhat antiquated 4000-series logic family: the
4093. This IC contains four inverted-output AND gates
(NAND gates) with Schmitt-trigger inputs. If water is
detected between the probes, it emits an intermittent
and rather irritating beeping tone.
The conductivity of the water is used to active the
circuit built around IC1a. The two electrodes (probes)
are fitted at the lowest point where water will come
to stand. They can be two tinned copper wires, but
you can also use two pieces of circuit board with the
copper surface coated with solder. The combination
of IC1a, resistor R2 and capacitor C2 forms a simple
oscillator that produces the intermittent (on/off)
effect of the alarm. If no water is present between the
14
IC1 = 4093
14
1
Naturally, the circuit of the alarm must be fitted
somewhere that will remain high and dry. Use a pair of
thin twisted wires to connect the electrodes (probes) to
the board. Naturally, you should use insulated, flexible
wire for this purpose. Twisting the wires together
makes the relatively long connection between the
probes and the circuit less sensitive to false alarms due
to external electromagnetic interference.
The current consumption is very low (less than 0.1 µA)
when everything is dry. When the buzzer is energised,
the current consumption can rise to around 2 mA. We
measured 3 mA with the frequency set to the maximum
value. The battery will
thus last for several
years as long as no
water is detected. Of
IC1.D
course, you should
13
11
&
12
bear in mind that the
BZ1
battery might start
leaking after a while…
C1
IC1
100n
7
IC1
4093
1
IC1.A
&
2
7
8
3
5
6
IC1.B
&
4 8
IC1.C
&
9
R2
1M
1M
R1
10
BT1
P1
R3
1M
9V
100k
C2
C3
330n
2n7
051010 - 11
i-TRIXX collection - 12/2006
13
Dicing with LEDs
Every self-respecting DIYer makes his own
electronic dice with LEDs as spots. Then you don’t
have to throw the dice anymore – just push the
button. The electronics also ensures that nobody
can try to improve his luck by fiddling with the
dice. Too bad for sore losers!
This circuit proves that an electronic die built using
standard components can be made quite compact. The
key component of here is a type 4060 digital counter
(IC1). This IC has an integrated oscillator stage, so
only two resistors (R7 and R8) and a capacitor (C7) are
necessary to generate the clock signal. The clock signal
is divided by various factors by the internal digital
circuitry of the IC. The division factors are designated
by ‘CT’ in the IC drawing symbol. For instance, the
signal on the CT3 output (pin 7) is a square wave with
a frequency equal to the clock frequency divided by
23 (8). The clock signal is divided by 24 (16) on the CT4
output, by 25 (32) on the CT5 output, and so on. This
means the output signals form a binary number that
+9V
T1
R6
BC557B
CTR14
CT=0
12
IC1
R7
10
C1
R8
9
220p
11
RX
CX
3
4
5
!G
220k
470k
16
6
+
CT
7
8
9
RCX
11
470k
4060
12
13
S1
D1
D10
D4
R5
R1
2k7
7
5
D2
4
D7
D5
6
14
13
15
1
D3
2
D6
3
2k7
8
R2
R3
2k7
100n
3k3
10k
C2
R4
counts upwards, which is naturally what a counter
does.
Of course, a die has only six possible values marked
on the six sides of a cube. This means that at least
three bits (the first three outputs) of the counter are
necessary to drive a display. Eight different counter
states (23) can be represented with three bits, but in
this case the counter must be restricted to six states.
To make sure this happens, D11, D12 and R6 are used to
reset the counter to its initial state when it reaches the
seventh state, which means when it reaches a binary
count of 110. When this happens, pins 4 and 5 of the IC
are both logic ‘1’ (high level), which causes a logic ‘1’
to be applied to pin 12 via resistor R6. This causes the
counter to be reset, which is what we want.
The display consists of seven LEDs arranged in the
same pattern as the usual markings on a normal die.
This arrangement is shown in the schematic diagram.
Before you begin thinking about the proper logical
connections between the LEDs and the counter
outputs, you can start by noting that except for the ‘1’
state there will always be two LEDs lit up at the same
time. This means that only four distinct indications
are necessary, instead of seven (with a total of seven
LEDs). Another advantage of this is that the current
consumption can be reduced by connecting pairs of
LEDs in series.
Resistors R1–R4 limit the current through the LEDs to
approximately 2 mA. This means you have to use lowcurrent LEDs. They are nice and bright at a current of
2 mA. Resistor R3 has a higher value because only one
LED is driven via it.
For convenience, the circuit is dimensioned based
on using a 9-V battery. The current consumption of
the circuit depends on the number of LEDs that are
illuminated, and with our prototype it varied over a
range of approximately 2.5 mA to 6.5 mA. The LEDs
still produce enough light even when the supply
voltage is as low as 6 V, but this depends strongly on
the characteristics of the low-current LEDs used in the
circuit.
Diodes D8–D10 and transistor T1 are necessary to
enable all the states of a normal die to be shown. By
that, we primarily mean the states with two or three
spots, which must be located diagonally.
D8
D11
D12
D9
D8...D12 = 1N4148
061001 - 11
state
1
2
3
4
5
6
binary
000
001
010
011
100
101
LEDs ‘on’
spots
1, 3, 4, 6, 7
5
1, 6, 7
3
1, 2, 3, 4, 5, 6
6
7
1
1, 3, 4, 6
4
1, 6
2
Electronics inside out !
14
i-TRIXX collection - 12/2006
Surf simulator
Do you long for a beach holiday on a tropical
island, but you don’t have the necessary
wherewithal? We’ve got just the answer: build the
i-TRIXX surf simulator, put on your headphones,
and dream yourself away from this dreary realm.
Let the rhythmic rush of the waves transport you
to a sun-drenched beach with gently swaying
palm trees, and relax for a while before returning
to a chilly confrontation with reality. That’s the
ultimate in low-budget travel. Book now!
For readers who want to delve more deeply into the
design, the following table shows the six different
binary states, which LEDs are lit up for each state, and
the number of spots shown by the die.
The die is operated by switch S1. In the quiescent state,
the break contact of S1 is closed and the oscillator
is stopped because the input of the oscillator stage
is connected to ground via the switch. When S1 is
pressed, the oscillator starts running and causes the
states of the LEDs to change at a rate of 1 kHz, which
is too fast to follow with the naked eye. This high
frequency ensures that the state of the die is purely
random when S1 is released, so there is no regularity
or pattern in the results.
Isn’t it great to relax on the snow-white sand of a
tropical beach with a cool drink in your hand? To enjoy
the magnificent of earthly creation while letting your
thoughts drift on the hypnotic mantra of the breaking
surf? No relaxant brewed by human hands can possibly
compete with it! But when you start thinking about
how much it all costs, you’ll reach for the headache
pills instead. Fortunately, there’s a less expensive way
to relax — with a bit of electronics that imitates the
soothing sound of the sea. You’ll have to imagine the
corresponding surroundings on your own. A sunlamp
and a few scoops of sand may help…
This circuit uses more components that most of the
i-TRIXX do it yourself projects, but this doesn’t make
it harder to understand how it works. We have also
designed a PCB layout for the circuit, which makes DIY
construction that much easier.
Noise is usually the last thing you want in any sort of
audio circuitry. Noise is generated in semiconductor
The circuit can be assembled on a small piece of
perforated prototyping board. Fit the LEDs in exactly
the same pattern as shown in the schematic diagram,
since otherwise the spot patterns will not correspond
to a real die. When you have assembled the circuit
board, fit it in a plastic enclosure along with a 9-V
battery to provide power.
+4V5
R15
33k
100k
D2
R14
2M5
3
R17
E
220k
C
2
B
C8
R18
100k
1
IC1.A
BT1
2µ2
1N4148
R13
6
2M2
+4V5
5
9V
7
IC1.B
120k
- 4V5
10k
1M
R4
47p
47µ 25V
10k
100n
13
14
R5
C3
220n
R3
100n
14
1
47p
IC1
10
3n3
D1
11
TL084
C4
22k
4k7
1M
R2
IC1.D
220µ
25V
100k
- 4V5
12
- 4V5
C5
9
R6
1M
C1
C11
IC1 = TL084
R10
R8
C9
IC1.C
8
R9
C6
1k
220n
7
8
R7
10k
R1
C7
100k
C2
4
IC1
green
R12
+4V5
100n
R19
D3
R11
C10
3k9
P1
R16
2k2
BC547B
1N4148
T1
- 4V5
065109 - 11
BC547B
i-TRIXX collection - 12/2006
15
T1 generates a constant noise signal, which doesn’t resemble the
sound of breaking surf. If you listen carefully to the sound of real
surf, you’ll notice that it resembles a noise signal that increases
rapidly in volume (as the wave rolls up the beach) and then slowly
dies down. This means the noise must rise and decay in a sawtooth
waveform. To achieve this effect, we make use of the AC impedance
of a normal diode (D1 in the schematic diagram), which depends on
the amount of DC current flowing through the diode. The higher the
current through the diode, the lower its AC impedance (and thus its
impedance to the noise signal). The voltage across R10 determines
how much current flows through diode D1. The noise signal is
amplified by IC1d and applied to the diode, and the voltage across
the diode is further amplified by IC1c to the output level. As already
mentioned, the amplitude of the noise signal depends on the DC
current through diode D1. What we have to do now is make the
current through D1 (or in other words, the voltage across R10) vary
in a sawtooth pattern. This job is handled by amplifiers IC1a and
IC1b. You can use P1 to adjust the form of the sawtooth (and thus
how the noise grows and decays) according to your taste.
The circuit works best with a clean 9-V supply voltage, so an AC
mains adapter with a stabilised 9-V output is the preferred choice
as a power source. A balanced supply voltage is required for proper
operation of the circuit, so a virtual ground is necessary. This is
created in a simple manner by a voltage divider (R18 and R19).
To reduce the current consumption (in case you want to power the
circuit from a battery), it also provides a ‘power on’ indication.
The current consumption of the circuit is approximately 9 mA,
which means the battery would have to be replaced already after
two days of continuous use if it is powered by a battery, so using
an AC mains adapter is certainly advisable.
COMPONENTS LIST
devices (transistors and diodes) as an undesirable by-product.
However, in our surf simulator we just can’t get enough of it! Noise
forms the basis for imitating the sound of breaking surf. We take
advantage of the fact that a reverse-biased base–emitter junction
of a transistor generates noise like the devil if the voltage is high
enough. The noise source in the schematic diagram is transistor
T1. The base–emitter junction of this transistor breaks down at
approximately 7 V (depending on the specific transistor). R1 limits
the current to a level that avoids destroying the transistor.
Resistors
R1,R2,R6 = 1 MΩ
R3 = 4kΩ7
R4,R8,R10,R14,R16 = 100 kΩ
R5 = 22 kΩ
R7,R12 = 10 kΩ
R9 = 1 kΩ
R11 = 120 kΩ
R13 = 2MΩ2
R15 = 33 kΩ
R17 = 220 kΩ
R18 = 3kΩ9
R19 = 2kΩ2
P1 = 2MΩ5 preset
Capacitors
C1,C10,C11 = 100 nF
C2,C5 = 47 pF
C3,C6 = 220 nF
C4 = 3nF3
C7 = 47 µF/25V radial
C8 = 2µ2 MKT lead pitch 5 or 7.5mm
C9 = 220 µF/25V radial
Semiconductors
D1,D2 = 1N4148
D3 = LED, 3mm, green, low current
T1 = BC547B
IC1 = TL084
Miscellaneous
6 PCB solder pins
BT1 = 9V battery with clip-on leads
(however 9 V battery eliminator
preferred)
We designed a PCB layout for this project to make it easier to
assemble, since it is a bit more complex than most of the i-TRIXX
circuits. If you follow the illustrated component layout, you
shouldn’t have any trouble at all. The idea here is that you print
the copper layout at actual size (52 × 52 mm) on transparent film.
The easiest way to do this is to use the supplied pdf file which you
can open with Adobe Reader. You can then use the film to expose
a circuit board, develop it, and then etch it. Alternatively, you can
take the film to your local electronics shop and have them make a
board for you.
If you need a real holiday before tackling this, visit your local travel
agent!
Electronics inside out !
16
i-TRIXX collection - 12/2006
Save Your Ears
‘Hello… HELLO! Are you deaf? Do you have disco ears?’ If people ask you this and
you’re still well below 8 , you may be suffering from hearing loss, which can come from
(prolonged) listening to very loud music. You won’t notice how bad it is until it’s too
late, and after that you won’t be able to hear your favourite music the way it really is
– so an expensive sound system is no longer a sound investment. To avoid all this, use
the i-trixx sound meter to save your ears (and your neighbours’ ears!).
Noise meter
With just a handful of components, you can build a
simple but effective sound level meter for your sound
system. This sort of circuit is also called a VU meter.
The abbreviation ‘VU’ stands for ‘volume unit’, which
is used to express the average value of a music signal
over a short time. The VU meter described here is what
is called a ‘passive’ type. This means it does not need a
separate power supply, since the power is provided by
the input signal. This makes it easy to use: just connect
it to the loudspeaker terminals (the polarity doesn’t
matter) and you’re all set. The more LEDs that light
up while the music is playing, the more you should be
asking yourself how well you are treating your ears
(and your neighbours’ ears).
Of course, this isn’t an accurately calibrated meter. The
circuit design is too simple (and too inexpensive) for
that. However, you can have a non-disco type (or your
neighbours) tell you when the music is really too loud,
and the maximum number of LED lit up at that time
can serve you as a good reference for the maximum
tolerable sound level.
Although this is a passive VU meter, it contains active
components in the form of two transistors and six
FETs. Seven LEDs light up in steps to show how much
power is being pumped into the loudspeaker. The steps
correspond to the power levels shown in the schematic
for a sine-wave signal into an 8-ohm load.
LED D1 lights up first at low loudspeaker voltages.
As the music power increases, the following LEDs (D2,
D3, and so on) light up as well. The LEDs thus dance to
the rhythm of the music (especially the bass notes).
D14
T8
39 Ω
BC337
1N4004
R8
BC337 BF245
30W
12W
D7
D6
D5
D4
D3
D2
D13
D12
D11
D10
D9
D8
0W5
D1
R7
4V7
2V0
T3
T2
T1
R6
R5
R4
R3
R2
R1
1k
BC337
6V8
T4
1k
22µ
63V
10V
T5
T1...T6 = BF245A
i-TRIXX collection - 12/2006
1W5
18V
1k
D
S
390 Ω
E G
3W
T6
1k
T7
C1
B
6W
10k
27V
C
To ensure that each LED only lights up starting at a
defined voltage, a Zener diode (D8–D13) is connected
in series with each LED starting with D2. The Zener
voltage must be approximately 3 V less than the
voltage necessary for the indicated power level.
The 3-V offset is a consequence of the voltage losses
resulting from the LED, the FET, the rectifier, and the
overvoltage protection. The overvoltage protection
1k
63V max
If you want to know more about the technical details of
this VU meter, keep on reading.
Each LED is driven by its own current source so it
will not be overloaded with too much current when
the input voltage increases. The current sources also
ensure that the final amplifier is not loaded any more
than necessary. The current sources for LEDs D1–D6
are formed by FET circuits. A FET can be made to
supply a fixed current by simply connecting a resistor
to the source lead (resistors R1–R6 in this case). With
a resistance of 1 kΩ, the current is theoretically limited
to 1 mA. However, in practice FETs have a especially
broad tolerance range. The actual current level with
our prototype ranged from 0.65 mA to 0.98 mA.
60W
1k
R9
This circuit can easily be assembled on a small piece
of prototyping board. Use low-current types for the
LEDs. They have a low forward voltage and are fairly
bright at current levels as low as 1 mA. Connect the VU
meter to the loudspeaker you want to monitor. If LED
D2 never lights up (it remains dark even when LED D3
lights up), reverse the polarity of diode D8 (we have
more to say about this later on). In addition, bear in
mind that the sound from the speaker will have to be
fairly loud before the LEDs will start lighting up.
065107 - 11
17
is combined with the current source for LED D7. One
problem with using FETs as current sources is that the
maximum rated drain–source voltage of the types used
here is only 30 V. If you want to use the circuit with an
especially powerful final amplifier, a maximum input
level of slightly more than 30 V is much too low. We
thus decided to double the limit. This job is handled by
T7 and T8. If the amplitude of the applied signal is less
than 30 V, T8 buffers the rectified voltage on C1. This
means that when only the first LED is lit, the additional
voltage drop of the overvoltage protection circuit is
primarily determined by the base–emitter voltage of
T8. The maximum worst-case voltage drop across R8 is
0.7 V when all the LEDs are on, but it has increasingly
less effect as the input voltage rises. R8 is necessary so
the base voltage can be regulated. R7 is fitted in series
with LED D7 and Zener diode D13, and the voltage drop
across R7 is used to cause transistor T7 to conduct.
This voltage may be around 0.3 V at very low current
levels, but with a current of a few milliampères it can
be assumed to be 0.6 V. Transistor T7 starts conducting
if the input voltage rises above the threshold voltage
of D7 and D13, and this reduces the voltage on the
base of T8. This negative feedback stabilises the supply
voltage for the LEDs at a level of around 30 V. With
a value of 390 Ω for R7, the current through LED D7
will be slightly more than 1 mA. This has been done
intentionally so D7 will be a bit brighter than the other
LEDs when the signal level is above 30 V. When the
voltage is higher than 30 V, the circuit draws additional
current due to the voltage drop across R8.
The AC voltage on the loudspeaker terminals is halfwave rectified by diode D14. This standard diode can
handle 1 A at 400 V. The peak current level can be
considerably higher, but don’t forget that the current
still has to be provided by the final amplifier. Resistor
R9 is included in series with the input to keep the
additional load on the final amplifier within safe
bounds and limit the interference or distortion that
may result from this load. The peak current can never
exceed 1.5 A (the charging current of C1), even when
the circuit is connected directly to an AC voltage with
an amplitude of 60 V. C1 also determines how long
the LEDs stay lit. This brings us to an important aspect
of the circuit, which you may wish to experiment with
in combination with the current through the LEDs. An
important consideration in the circuit design is to keep
the load on the final amplifier to a minimum. However,
the combination of R9 and C1 causes an averaging
of the complex music signal. The peak signal levels in
the music are higher (or even much higher) than the
average value. Tests made under actual conditions
show that the applied peak power can easily be a
factor of 2 to 4 greater than what is indicated by this
Electronic
How about an amusing (although your victims may
not agree) circuit that you can use to play a trick on
your friends or family, or to get rid of your mother-inlaw? A handful of electronic components can create
a cricket-like chirping sound every few minutes or so.
You should hide this electronic poltergeist such that
it’s difficult to find, but can still be heard clearly. It is
guaranteed to drive people mad! i-TRIXX shows you
how to build this irritating circuit. Read on...
- Did you hear that? - What? - That chirping noise.
- Chirping noise? - Yes, I think there’s a cricket
somewhere in the room. - I didn’t hear anything!
- Well, it’s stopped now!
A bit later: - There it is again! Did you hear it? - I heard
nothing, go back to sleep! - I’m not going mad am I?
I’m sure there’s a cricket around somewhere! We have
to get rid of it, otherwise I won’t be able to sleep!
This could be a possible conversation in the bedroom
where you’ve hidden the electronic poltergeist. But
before you can create all this mischief you’ll have to
use your soldering iron. You’ll be pleased to hear that
the construction of this circuit won’t give you sleepless
nights, as long as you work carefully.
At the heart of the circuit is an old favourite IC, the
4093. This chip contains four NAND gates, each of
which has two inputs. These NAND gates have Schmitt
triggered inputs (more on this later) and are ideal for
use in this particular application. Three of the gates
are used as oscillators. The oscillator built around IC1A
produces a high-pitched sound similar to that made by
crickets when they rub their wings together. A second
VU meter. This amounts to 240 W or more with an
8-Ω loudspeaker. You can reduce the value of C1 to
make the circuit respond more quickly (and thus more
accurately) to peak signal levels.
Now a few comments on D8. You may receive a
stabistor (for example, from the Philips BZV86 series
or the like) for D8. Unlike a Zener diode, a stabistor
must be connected in the forward-biased direction.
A stabistor actually consists of a set of PN junctions in
series (or ordinary forward-biased diodes). Check this
carefully: if D2 does not light up when D8 is fitted as a
normal Zener diode, then D8 quite likely a stabistor, so
you should fit it the other way round.
Electronics inside out !
18
i-TRIXX collection - 12/2006
R3
D1
47k
10k
R2
poltergeist
oscillator (IC1B) is used to interrupt this noise at
regular intervals. The third oscillator (IC1C) is used to
turn on the chirping noises for a short while, with a
few minutes silence in between.
So how do these oscillators work? We’ll take IC1A
as an example. The output of this binary NAND gate
will only be a logic low (virtually equal to 0 Volts)
when both inputs are at a logic high level (typically
just over half the supply voltage). Each input has a
Schmitt trigger circuit so that slowly changing input
signals can be also be dealt with. The Schmitt trigger
makes the gate switch state suddenly when the input
is slowly increasing and reaches the point where it
could be considered at a logic high state. Inputs that
are hesitating somewhere between low and high are
effectively given a helping hand upwards. When the
supply voltage (a 9 V battery) is first connected to the
circuit, the input at pin 2 is at a logic low level (since
capacitor C2 has not yet charged up). This means
that the output of the NAND gate at pin 3 will be at
a logic high level. Capacitor C2 is now charged up via
the output at pin 3 and resistor R3 until the Schmitt
trigger decides that the voltage level at pin 2 has
increased enough for it to be at a logic high level. The
output of the gate now switches over to a logic low
level, assuming that the input at pin 1 is at a logic
high level. Capacitor C2 will now be discharged by the
output at pin 3 via R3, which causes the input at pin 2
to become low again, and the whole cycle repeats
itself. The speed (frequency) at which this charging and
discharging takes place depends on the values of R3
and C2.
1N4148
9
3
C2
IC1.C
&
&
2
3M3
8
1
IC1.A
10
C1
13
10n
12
IC1.D
&
R5
R4
47k
The oscillator built around
220µ
IC1.B
IC1C works in a slightly
16V
6
differently way. In this
&
5
IC1 = 4093
case a diode (D1) causes
C3
capacitor C1 to charge
330n
via both R1 and R2, which
1
is much quicker than
when C1 discharges via
IC1
BT1
R2 only. This causes the
4093
poltergeist to be quiet for
9V
three minutes (the slow
discharge) and only make
a noise for a second or so
(the fast charge).
The fourth gate (IC1D)
is used to combine the two oscillators (IC1A, which
makes the high-pitched sound of the wings rubbing
together and IC1B, which imitates the periodic
movement of the wings) and drive the sounder.
It is possible to change the frequencies at which gates
IC1A to IC1C operate by varying the values of resistors
R1 to R4. A lower value results in a higher frequency,
and a higher value lowers it. You could also consider
reducing the volume of the sounder by increasing the
value of R5; this will make it even more difficult to find
the circuit.
The current consumption of our prototype was under
300 µA during the time it was silent, which rose to
about 1.3 mA when the circuit was producing sounds
(this only lasts 1 second). With an ordinary 9 V battery
this circuit can operate for several (irritating) months,
and there won't be many people who could put up
with that!
11
1k
R1
BZ1
4
+9V
C4
14
IC1
100n
7
065126 - 11
Pump it up: MP3 booster
MP3 players are all the rage these days.
The smaller ones in memory-stick format are
particularly easy to take with you; your very
own ‘personal sound system’ on the move!
It’s when you want others to share your taste
in music that you find these players to have a
lack of power. You can get round this problem
with the help of the i-TRIXX MP3 booster, a
small amplifier that can be used to connect
your MP3 player directly to your Hi-Fi. When
you next invite your friends to a party you can
ask them to bring their ‘personal music’ as
well as the usual drinks! But first we have to
build this booster!
The small battery-powered players have an output
signal that is more than sufficient to drive a set
i-TRIXX collection - 12/2006
of 32 Ohm headphones. You’ll often find that with
an output of 1 mW the sound pressure level (SPL)
produced can reach up to 90 dB. This would be
sufficient to cause permanent damage to your hearing
after only one hour! The maximum output voltage will
then be around 200 mV. This, however, is insufficient
to fully drive a power amplifier. For this you’ll need
an extra circuit that boosts the output voltage. Power
amps usually require 1 V for maximum output, hence
the signal has to be amplified by a factor of five. We
will also have to bear in mind that quieter recordings
may need to be amplified even more. We’ve used a
simple method here to select the gain, which avoids
the use of potentiometers. After all, the MP3 player
already has its own volume control. We decided to
have two gain settings on the booster, one of three
times and the other ten times.
19
Musical saw
Resistor R4 (R8) takes the amplified MP3 signal to the
output socket K2 (K3). A cable then connects these
phono sockets to the input of your power amplifier.
The resistors connected in series with the output (R4
and R8) are there to keep the booster stable when
a long cable is connected to its output. Cables have
an unwelcome, parasitic capacitance. This capacitive
effect could (due to phase shifts of the signal) affect
the negative feedback of the booster in such a way
Electronics inside out !
20
H1
P1
C7
C1
H2
061002-1
C5
P4
C8
R19
C6
R4
T2
OUT
D1
R13
H4
P3
T1
C4
C3
BT2
+
R20
R12
R11
-
R14
POT2
POT3
POT1
POT
by a saw. Oscillator IC1B creates a type of vibrato
effect (this is a quick variation of the frequency, and
is comparable to the effect produced when you use
your hand to rapidly change the amount of bending of
the saw). Oscillators IC1C and IC1D, which oscillate at
lower frequencies, create an arbitrary variation of the
frequency (melody) of the electronic saw (corresponding
to small changes in the level of bending of the saw).
The oscillators operate as astable multivibrators with
a Schmitt trigger. We’ll take the oscillator built round
IC1B to explain how it works. When power is first
applied, capacitor C1 will be completely discharged
and a voltage of -9 Volts will appear at the inverting
input (pin 6) of IC1B. Since the non-inverting input (pin
5) will be at a higher potential (via R1 and R2), the
output (pin 7) will be at a high level (virtually equal to
the supply voltage of +9 Volts). This output will now
charge capacitor C1 via P1 and R3, until such time
that the voltage at pin 6 rises above that of pin 5. At
that moment the output of the opamp switches over
to -9 Volts (the negative supply voltage). This causes
capacitor C1 to discharge via P1 and R3, until the
voltage at the inverting input falls below that of the
non-inverting input, when the whole process repeats
itself. The output of the oscillator is therefore a square
wave. Resistor R4 and capacitor C4 smooth the sharp
that a positive feed back occurs, with the result that
the booster oscillates and possibly damages the power
amplifier! The resistors (R4 and R8) effectively isolate the
output of the booster from the parasitic capacitance of
the output cable. They also protect the booster outputs
from short circuits.
We’ve used a TS922IN opamp in this booster because it
can operate at very low supply voltages (the maximum
is only 12 V!), but can still output a reasonable current
(80 mA max.).
For the supply we’ve used rechargeable batteries (e.g.
NiCd or NiMH cells) so that we don’t need a mains
supply. To keep the number of cells required as small as
possible, we’ve chosen a supply voltage of 5 volt; this
can be supplied by four rechargeable batteries. It is also
possible to use four ordinary, non-rechargeable batteries;
it’s true that the supply voltage then becomes a bit higher
(6 Volts), but that won’t cause any harm.
Since we’ve used a symmetrical supply for the booster
(2 x 2 batteries), it will be easiest if you use two separate
i-TRIXX collection - 12/2006
K1
H3
IC1
R8
R3
R2
R1
R15
R16
R17
R18
BT1
T
Amplifiers IC1A and IC1B (for the right and left
channels) are housed in a single package, a TS922IN.
The output signal of the MP3 player is fed via a stereo
cable and socket K1 to the inputs of the amplifiers. The
gain depends on the relationship between resistors R2
and R1 (R6 and R5 for the other channel) and is equal
to ten times. When you add jumper JP1 (JP2), resistor
R3 (R7) will be connected in parallel with the negative
feedback resistor R1 (R6), which causes the gain to
be reduced to about three. When you start using the
booster you can decide which gain setting works best
for you.
R7
R6
R5
R9
R10
ROTKELE )C(
The circuit consists mainly of three, almost identical,
adjustable oscillators built round IC1B, IC1C and IC1D.
The frequencies can be adjusted via potentiometers
P1 to P3, with a slight overlap in the frequency range
between the oscillators. A fourth (similar) oscillator,
which produces the final output, is modulated by the
first three oscillators, which creates a varying frequency
that quite closely resembles the sounds produced
C2
+
Usually it’s the person who sings when the sawing goes well.
But the saw itself can also be made to produce musical sounds
by drawing a bow across its edge. By bending the saw in
different ways an experienced saw player is able to extract a
musical melody from the saw. If you don’t have a spare saw and
bow to hand you could always build this electronic version. At
least with this instrument you’ll never cut yourself! So warm up
your soldering iron and get cracking!
Everybody who has sawn some wood will be aware
that the saw can sometimes produce musical sounds,
especially when the sawing isn’t going very smoothly.
There are even professional musicians who use specially
manufactured saws as musical instruments. You’ll only
appreciate their skills if you happen to like this style
of music though. As a dedicated electronics hobbyist
you’d rather build an electronic version, of course, and
we’ve designed a simple circuit for you. The circuit
diagram may appear a bit complex at first, but don’t let
it worry you. We have also designed a printed circuit
board for this circuit, which makes the construction a
lot easier than if you had to use experimenter’s board.
The complete circuit is built around a single IC, a TL084,
which contains four opamps. These opamps have been
configured as oscillators in this circuit, and together
they produce sounds similar to that of a musical saw.
1-200160
P2
R2
R16
IC1 = TL084
100k
C1
–9V
1µ
63V
R15
5
7
IC1.B
6
R3
330k
C5
100k
2µ2
63V
9
IC1
D1
11
BT2
8
R8
P2
9V
100n
T1
G
1M2
R14
R7
C8
D
P4
1M
S
BF256C
4M7
–9V
22k
1N4148
10
IC1.C
061002-1
100n
4
2M7
R6
C2
C7
9V
R17
R18
R13
1M
100k
2
BT1
1
IC1.A
220n
P1
47k
R5
3
4k7
R4
(C) ELEKTOR
100k
1k
R1
+9V
100k
–9V
+9V
47k
1M
R10
–9V
4µ7
63V
14
IC1.D
13
R12
P3
47k
1M2
10µ
63V
1M
100n
R19
G
D
S
C
K1
C6
C4
R11
1
BC517
BC517
TL084
R20
E
B
–9V
edges of this square wave, which produces a more
pleasing sound when it is fed into the final oscillator
(IC1A). The outputs of the other two oscillators (IC1C
and IC1D) are also added to this modulation signal.
P4 is used to adjust the level of modulation that is
presented to the input of FET T1. This FET is connected
in parallel with resistor R18 and determines the
frequency (pitch) of oscillator IC1A. This oscillator also
produces a square wave output (on pin 1). However,
the output is taken from pin 2, which is a more rounded
signal and therefore won’t sound so harsh. This signal
is first buffered by Darlington transistor T2 so that the
frequency isn’t affected when you connect a load to the
output.
The electronic musical saw can also be played by hand.
In that case you should replace presets P1 to P3 with
slider potentiometers, which can be easily and quickly
adjusted by hand.
–9V
061002 - 11
The current consumption of the circuit is on average
+/- 8 mA. When you use two 9 V batteries they can
provide the circuit with many hours of (musical?) power.
To make the construction of this circuit a lot easier
we have designed a PCB layout. Since we have also
included the component layout there’s not much that
could go wrong with the construction. You should copy
the track layout at its true size (70 x 50 mm) to a sheet
of transparent plastic. That is easiest if you download
the PDF file supplied for this project and use Adobe
Acrobat Reader to print it out. You can use this to
expose, develop and etch your own PCB. Alternatively,
you could take it to a local electronics shop who may be
able to produce the PCB for you.
R3
battery holders, each with two AA cells. The two
holders are connected in series. Make sure that
the batteries are connected the right way round;
the positive of one always has to be connected to
the negative of the next. This also applies to the
connection between the two battery holders. S1A/B is
a double pole switch, which is used to turn both halves
of the battery supply on or off simultaneously.
If you can’t find the (dual) opamp we’ve used (or an
equivalent), you could always use standard opamps
such as the NE5532, TL082 or TL072. These do need
a higher supply voltage to operate properly. In these
cases you should use two 9 V batteries and replace
resistor R9 with a 15 kΩ one. Do take care when you
connect the circuit to your power amplifier because
the output signal can be a lot larger and you could
overload the power amplifier. (Although you’re more
likely to damage the loudspeakers, rather than the
amplifier!)
(Please note that these two 9 V batteries can’t be used
as a supply for the TS922IN!)
+5V
4k7
IC1
S1.A
JP1
1
R2
10k
TS922IN
i-TRIXX collection - 12/2006
IC1
R1
1k
BT1
2
3
K1
IC1.A
1
R4
K2
R
100 Ω
R9
C1
*
BT2
3k3
C3
BF256C
1M
100k
12
10k
R9
T2
–9V
100k
100n
8
R
IC1 = TS922IN
IC1
L
4
5
R5
1k
6
IC1.B
7
BT3
R8
L
100 Ω
K3
10k
R6
4k7
R7
C2
*
BT4
D1
100n
JP2
* see text
S1.B
065107 - 11
In our circuit we’ve used a stereo jack socket for the
input and phono sockets for the output because these
are the most compatible with MP3 players and power
amplifiers respectively. If you wanted to, you could solder shielded cables directly to the circuit instead, with
the correct plugs on the ends. You’ll never find yourself
without the correct connection leads in that case!
21
Luminous house number
During the dark autumn and winter months in
particular you’ll find that ordinary house numbers
are more difficult to read, especially if your home
is further back from the road. If you want to avoid
that your family, friends and deliverymen drive
past your home, you should build this electronic
version. At the same time you can show the world
that you’re an avid electronics hobbyist! It can
easily be put together during a rainy weekend.
For a change this isn’t a hi-tech circuit, but instead is
quite straightforward and it comes in useful too. House
numbers can sometimes be difficult to read. In this
article we show you how to build a luminous version
that uses large 7-segment displays made by Kingbright,
which have digits with a height of 57 mm. There are
of course other manufactures that produce similar
displays and it is not essential that you use the same
displays that we’ve used here. You do have to keep
an eye on the number of LEDs used to make up each
segment (7 in total) and on the forward voltage drop
when the segments are lit, so that you don’t exceed the
maximum current through the LEDS, but more on this
later.
The required segments of the display are connected
to the supply via resistors. When you connect all
segments you obtain the digit 8, for a 0, 6 or 9 one of
the segments is disconnected and a 1 for example only
needs two segments lit. In this way you can display all
of the digits from 0 to 9. Each segment in the display
consists of a number of LEDs connected in series (in
our module this is 4). In the display used by us the
anodes of the first LED in every segment are connected
together, hence the term ‘common anode display’.
The common anode obviously has to be connected to
the positive supply. The ‘free’ ends of the segments
that have to be lit are connected via a resistor to the
negative supply (ground). The current through the
segments and hence their brightness is dependent on
the value of the resistors.
For the power supply we can use a doorbell
transformer, which is often already present and which
generally supplies 8 Volts AC. The alternating voltage
of the doorbell transformer is rectified so that the
peak current through the LEDs doesn’t exceed their
limits. The rectifier circuit consists of a standard bridge
rectifier followed by a smoothing capacitor. These
can supply the right voltage for a one or two-digit
house number. For house numbers of three digits or
Electronics inside out !
22
more you will have to experiment with the resistors
to find what values give the correct current through
the segments. But for house numbers like 88, 80 or
90 (many segments will be lit) this also applies. For
every extra segment that is lit the current consumption
increases, which causes the supply voltage to drop by
an amount dependent on the quality of the doorbell
transformer. It is therefore difficult to give an exact
figure for the value of the current limiting resistors
in the circuit; the 47 Ω mentioned here should be
taken as a guideline to give a current of about 22 mA
(which is a safe value for the type of display we used)
when an 8 Volt doorbell transformer is used. If you
want to be sure that you’re not exceeding the current
limit you should place an ammeter in series with the
supply and divide the measured current by the number
of segments that are lit. If the result is greater than
22 mA you should increase the values of the resistors,
and reduce their value if the result is less than 22 mA.
To determine what current flows through a segment
you could of course measure the voltage across the
resistor connected to it and use Ohm’s law to calculate
the current.
For the installation it’s best to split the circuit into
two parts: the bridge rectifier (B1) and the smoothing
capacitor (C1) near the transformer and the display
and resistors in a weather proof box next to the front
door. When your house number consists of two digits
you should build two display modules (shown in the
circuit diagram inside the dotted lines). The display
section should be connected to the power supply
section using a twin-cored cable such as loudspeaker
cable. Try to use a cable with reasonably thick cores,
especially if a lot of segments are lit. For ‘88’ the
current requirement is 0.3 A!
The house number also has to be legible in strong
sunlight. To achieve a better contrast you can place
a piece of transparent plastic (with the same colour
as the LEDs) in front of the display, or you could stick
transparent tape over the segments (using the right
colour). In the latter case you should only cover the
active segments so that you can still see the number
without an applied voltage.
The table below shows you which resistors need to be
soldered to produce each digit:
0
1
2
3
4
5
6
7
8
9
R1,R2,R3,R4,R5,R6
R2,R3
R1,R2,R4,R5,R7
R1,R2,R3,R4,R7
R2,R3,R6,R7
R1,R3,R4,R6,R7
R1,R3,R4,R5,R6,R7
R1,R2,R3
R1,R2,R3,R4,R5,R6,R7
R1,R2,R3,R4,R6,R7
8V
B1
C1
C1
B80C1500
i-TRIXX collection - 12/2006
1000µ
1000µ
16V
16V
Applause generator
No, the performance of our national team during the 2006 World Cup Soccer Tournament did not
deserve any applause. Better occasions come to mind where a resounding applause is much more
appropriate. And to prevent injury to your hands when this occasion does present itself, we developed
an electronic applause generator. In this way you can avoid being the only one applauding in a busy hall
and prevent both yourself as well as the receiver from feeling uncomfortable.
The emitter voltage is set by R7 and R8 to half of the
power supply voltage. T2 comes out of conduction for
part of the time because of the size of the noise signal.
There is therefore a switching effect with the noise that
gives a strong impression of applause. The signal is
available at the output via C7.
If you object to the lower frequencies then you can
pick a smaller value for C7. D1 ensures that C6 is
discharged quickly when the circuit is switched off
so that when it is switched back on again the same
delay and growth of applause results. If you like
experimenting, you can change the values of C2, C3
and R4 or choose a different opamp and try to make
the sound even more realistic. If you would also like
the applause to decay slowly, you can connect R5
alternately to the positive and negative of the power
supply, for example with the aid of a change-over
switch (D1 is omitted in this case). A smaller value of
C6 accelerates the effect.
The current consumption of the circuit is only 4 mA;
a 9-V battery is therefore eminently suitable as the
power supply. If you plan to put the circuit in a box
then do not forget the on/off switch.
Once you’ve got the circuit to work, the first applause
is naturally for yourself!
C4
47µ
25V
100n
R5
D1
R7
R9
2k2
C2
2k2
1M
R2
2M2
R1
100k
1N4148
R4
C1
BT1
9V
3
100n
2
1
7
5
IC1
C7
R6
1M
6
100k
Applause from a large crowd sounds very similar to
noise. Noise is therefore also the source signal in this
circuit. A proven method for the generation of noise is
to ‘zener’ the base-emitter junction of a transistor.
That is what T1 and R1 are for. The base-emitter junction is wired as a diode in the reverse biased direction
and connected via R1 to the power supply voltage of
9 volts. Because of this relatively high voltage, the
base-emitter diode will break down, but the high resistance value of R1 prevents the junction from failing
because of excessive current. The voltage across the
base-emitter diode is relatively constant; this diode
acts like a zener diode. The noise that is produced is
very small however and is considerably amplified
by opamp IC1 first, before any further processing. The
opamp is set to half the supply voltage by R2 and R3.
The noise signal is applied to the input of the opamp
via C1. This capacitor ensures the separation of the
high DC voltage (the ‘zener voltage’) at the emitter of
T1 and the half-power-supply-voltage at pin 3 of the
opamp. In our prototype this zener voltage amounted
to 8.3 V. It can happen that this voltage is greater than
9 V. In that case you will have to either increase the
power supply voltage or pick another transistor for T1
that has a lower zener voltage. This is just a case of
swapping the transistor and repeating the measurement.
Because of the presence of C2 and C3, the opamp amplifies mostly the low-frequency part of the noise. This
results in the best approximation of real applause.
The transistor stage that follows provides the clapping
sound.
The noise is presented to the base of T2 via C5. C6
is charged slowly via R5 after the power supply is
switched on and this causes T2, via R6, to progressively
conduct more and more. As a result the noise signal at
the collector of T2 will grow to a maximum.
C5
T2
10µ
63V
470n
4
BC547B
TL081
T1
R2
R2
*see text
BC547B
*
LD1
LD1
*
77
R4
R4
*
44
R5
R5
*
11
R3
R3
R6
R6
*
R7
R7
*
66
22
99
10
10
55
aa
88
CA
CA
33
CA
CA
47µ
25V
10µ
63V
R8
R10
TL081
BC547B
bb
C6
100k
*
C3
2k2
R1
1M
R3
061006 - 11
cc
dd
ee
C
ff
E
B
gg
dp
dp
SA23-12EWA
SA23-12EWA
R1
R1 ...
... R7
R7 == 47Ω
47Ω
061004
061004 -- 11
11
i-TRIXX collection - 12/2006
23
TWO-CHANNEL USB
PC OSCILLOSCOPE
£ 399,99
4HE0#35DIGITALSTORAGEOSCILLOSCOPEUSESTHEPOWEROF
YOUR0#TOVISUALIZEELECTRICALSIGNALS
(IGHSENSITIVERESOLUTIONDOWNTOM6
s
1Gs/ Hz
M
2x60
THREE-IN-ONE LAB UNIT
#ONTENTSMULTIMETERPOWERSUPPLYSOLDERINGIRON
4HISUNITCOVERSOFYOURHOMELABNEEDS
0ERFECTSOLUTIONFORSTARTERSORSCHOOLS
PCSU1000
10MS/s PERSONAL
POCKET SCOPE
£ 154,99
7ORLDSMOSTPORTABLEOSCILLOSCOPE
TION
£ 89,99
EC
ONN
PC C RS232
LAB1
PPS10
D
MBLE
ASSEDUCTS
PRO
MODULES
SUBWOOFER KIT
#OMPLETEKITJUSTADDWOOD+ITINCLUDES
AMPLIlERANDlLTERCIRCUITTWOCM
WOOFERSALUREARPANELBASSREmEXTUBE
SMALLHARDWAREANDASSEMBLY
INSTRUCTIONSFORBOTHAMPCIRCUITAND
ENCLOSURE
£ 79,99
8-CHANNEL RF REMOTE
CONTROL SET
CHANNEL2&REMOTECONTROLSET
,ATCHINGORMOMENTARYMODEFOREACHOUTPUT
200W
£ 69,99
K8077
VM118
EXTENDED USB
INTERFACE BOARD
%XCELLENTVALUEFORMONEY&ULLYDISCRETEDESIGN/VERLOADANDSHORTCIRCUITPROTECTION
UNIVERSAL)/CARDWITH53"©INTERFACE
3AFEOPTICALLYISOLATEDCOMMUNICATION
BETWEENALLTHE)/SANDYOUR0#
K8061
£ 39,99
VM100
2 MODULAR DIGITS WITH
SERIAL INTERFACE
"RIGHTREDJUMBOSIZEDSEGMENTDISPLAYSMM
#OMPLETEWITHSNAPINENCLOSURE-ULTIPLEUNITSCAN
BELINKEDINANEASYWAYTOCREATELARGERREADOUTS
FOREGSCOREKEEPINGTIMINGCOUNTING
%ASYPLATFORMINDEPENDENTCONTROLVIA23
THREEWIREINTERFACE3OLDPERPAIR
ULTRASONIC RADAR
MODULE
"UZZEROUTPUTFASTSLOWCONTINUOUS$ISTANCEINDICATION
WITH,%$BAR!DJUSTABLEALARMDISTANCE,%$ALARM
INDICATOR$RYCONTACT./.#RELAY!PPLICATIONSROBOT
PROJECTSOBJECTDETECTIONDISTANCEMEASURINGPARKINGAID
SECURITYSYSTEMS
£ 19,99
£ 29,99
VM125
K8063
FLASHING YELLOW
LED STAR
ELECTRONIC
CHRISTMAS
TREE
DE LUXE XMAS
TREE
60-LED MULTIEFFECT STAR
FLASHING
RED LED
STAR
SOUND STAR
VU METER
MK170
RIDING SANTA
MK169Y
MK100
MK169R
!,,4(%3%+)43(!6%,%$!.)-!4)/.3 MK117
For more information and dealers, see www.vellemanprojects.com
MK172
Suggested retail prices VAT included
www.velleman.eu
MK116
Suggested prices quoted in sterling including VAT. Misprints reserved. All pictures and texts are subject to change without prior notice.
£ 65,99
200W POWER AMPLIFIER MODULE
12/2006 - elektor electronics
69
know-how
rf
WLAN Antenna Desi
Increased range the easy way
Stefan Tauschek and Thomas Scherer
The domestic use of WLANs has grown rapidly as DSL routers with built-in wireless Ethernet have
become available, and now it is easy to use a notebook PC to surf the Internet wirelessly from the
comfort of one’s sofa. However, things get trickier if a reinforced concrete wall stands in the way, or
if a neighbour happens to be using the same frequency...
The enormous popularity of WLANs
(wireless local area networks) is easy
to understand: not everyone has their
desk situated next to a telephone socket. Even in the case of desktop PCs it
is now easier to provide a fast Internet connection via the ether rather
than by installing fixed cables. Unfortunately, things do not always go perfectly smoothly in practice: sometimes
it can be difficult to set up a reliable
connection between two devices even
just a short distance apart in the same
building.
The problems and their causes
The frequency used for WLAN communications according to IEEE 802.11b or
802.11g is around 2.4 GHz. At this frequency radio waves propagate quasi-optically and are considerably attenuated by moisture in walls. Reinforced concrete and limestone block
the waves to an even greater extent
because of their high metal content.
A further limitation is that in Europe
transmit power in this band is limited
to 100 mW.
Often also corners are cut in the interests of cost reductions. A WLAN
router with a price tag of a few tens of
pounds will make a few compromises
in performance: a typical device will
have inside a mini-PCI WLAN card,
70
just as a laptop might. Such cards often only output around 50 mW rather
than 100 mW, and poor matching to
the antenna often accounts for a few
more dB of loss. The overall effective
transmit power might only be around
10 mW or 20 mW.
A straightforward way to recoup some
of this loss is to use a special antenna
that offers gain. And that is what this
article is about: how to build a DIY directional WLAN antenna which focuses the available transmit power in the
desired direction, providing a gain of
several dB over a conventional omnidirectional antenna. Furthermore, a directional antenna does not just provide
gain in the transmit path; signals received from sources within the antenna’s beam are amplified by the same
factor. Since the WLAN connection is
bidirectional, this means that a directional antenna can give us a considerable increase in range without the need
for complex RF electronics.
The solutions
When faced with a poorly-performing WLAN, it is wise to pause before
reaching for the soldering iron. A couple of aspects should be considered
before deciding to make or buy a directional antenna.
First it is worth noting that the best
Ethernet connection is a wired
one. A wired connection is both
faster and more reliable. If this is
not an option for any reason, or if
(because connection is to be made to
a notebook) it is not convenient, then
there are a few things one can do to
improve the performance of a wireless
network. The first step is to try moving the router a few feet closer to the
computer. Another option is to splash
out a few tens of pounds on a repeater, which can add five to ten metres
of range within a building. Even better is to deactivate the WLAN part of
the router and, for a similar sum, purchase an access point (see Figure 1).
This is a box of electronics which takes
an Ethernet connection on one side
and provides a WLAN connection on
the other. The device can be connected to the router using an Ethernet cable. Because the device is dedicated
to the one function, we might reasonably hope that it would provide better
RF performance. A more significant
advantage is that multifunction devices that combine a DSL modem with
WLAN router and switch functions do
not normally have a usable RF connector, and so it is hard to connect an external antenna. The small stub antenna usually provided is connected by a
fixed wire to a sub-miniature connector on the internal WLAN card. This
connector is not designed for repeated
elektor electronics - 12/2006
ign
plugging and unplugging and is unmanageably
tiny. Access points, however, are available with common-or-garden SMA connectors (see Figure 2), making it easy
to connect an external purchased or
home-made directional antenna.
One further piece of advice: it is preferable to use a longer Ethernet cable
rather than a longer antenna cable. It
is easy to achieve data transfer rates of
100 Mbit/s over 50 m or more of CAT5
cable; but the losses in 50 m of antenna
cable could easily cancel out the benefits of a directional antenna.
Antenna types
In the following discussion we shall
not consider WLAN routers that employ several antennas and MIMO (multiple input multiple output) technology.
Commercially-available access points
(such as the one shown in Figure 1) are
usually fitted with a so-called quarterwavelength stub antenna, or monopole (see Figure 3). Sometimes the stub
can be entirely inside the enclosure (as
long as it is made of plastic). The antenna consists of a piece of wire with
12/2006 - elektor electronics
length
λ/4.
At
2.44 GHz, this is
a quarter of 300×106 / 2.44×109 metres,
or slightly more than 3 cm.
At the other end of the spectrum from
this simple antenna is the parabolic reflector, which can have a diameter of
several metres. This can offer a gain
of up to 60 dB over the simple quarter-wavelength monopole. European
regulations only allow such antennas
to be fed at a very low power. Experiments in the USA with specially-constructed (and expensive) antennas of
this type have achieved ranges of up
to 200 km.
A wide range of directional antenna
designs between these two extremes
have been tried for WLAN applications. Two designs have proved most
successful, offering good gain and simple construction. The first type takes
the form of a waveguide and goes by
the catchy name of ‘cantenna’ (see
left-hand half of Figure 4). The second
type consists of specially-arranged
diamond-shaped sections in front of a
reflector, and is called a ‘biquad’ (see
right-hand half of Figure 4). The lat-
Figure 1. SMA connector on the rear panel of an access point.
Figure 2. A typical access point: sometimes these are used to
help reduce the length of antenna cable needed.
71
know-how
In the Elektor E
rf
ter type can in theory offer a gain of
around 12 dB (although as we shall see
later, practice can deviate from theory!), somewhat more than the 10 dB
that the cantenna can provide. Both
types give considerable improvements
over typical integrated antennas, and
we shall now go on to look at them
both in more detail.
Figure 3. Quarter-wavelength stub antenna suitable for a
WLAN router or access point.
The cantenna
As mentioned above, the cantenna
operates as a kind of waveguide. The
theory of such antennas is far from trivial; those interested can find a good introduction at [1]. As can be seen from
Figure 4 and the drawing in Figure 5,
the antenna consists of a can of certain
specified dimensions and a carefullyarranged feed.
Figure 4. Prototype cantenna and biquad antenna constructed
in the Elektor Electronics laboratory.
100 mm
123 mm
Radiator
Radiator
Offset
060056 - 11
Figure 5. Construction drawing for the cantenna.
There is a wide range of guides available to constructing cantennas of various dimensions [2]. The following suggestions have the advantage that they
have been tested by simulation, carried
out by Stefan Tauschek using a software package called 4NEC2, available
for free download from [3]. The program
is based on the so-called boundary element method [4]: the idea is to convert
Maxwell’s equations into a system of
linear algebraic equations, which are
then stepwise integrated to calculate
the current distribution in the antenna.
The ‘NEC’ in the program name stands
for ‘numerical electromagnetic code’.
Although their derivation is complicated, the results themselves are simple:
Figure 6 shows the three-dimensional
model of the Cantenna in 4NEC2 and
Figure 7 the calculated radiation pattern. The directional nature of the antenna is clear.
To make an accurate cantenna the can
must be exactly one wavelength long.
At 2.44 GHz this is very nearly 123 mm.
The internal diameter is approximately
100 mm, slightly more than 4/5 λ.
The feed stub, shown as a radiator,
should be approximately λ/5, or 25 mm,
long. Ideally this is a wedge-shaped or
tapered piece of metal, with the thicker
end pointing to the middle of the can.
The distance from the base of the can,
the ‘radiator offset’, is the rather odd
multiple of 7/32 times the wavelength,
or 27 mm.
Figure 6. Three-dimensional model of the cantenna plotted
by 4NEC2.
72
It is difficult to find ready-made cans
with these dimensions. A reasonably
accurate version can be made by hand
from copper sheet as shown in Fig-
When we had constructed prototypes in
our laboratory we naturally wanted to
test them immediately. The test equipment comprised an ordinary laptop and
a PC as the fixed station to which the
various antennas were connected. The
walls of the laboratory building are built
using a type of brick that absorbs RF of
this frequency very well. The layout of the
building is a chain of rooms in a slightly
staggered arrangement. There are many
PCs and other electronic devices in the
laboratory and in the editorial offices,
producing a high level of electromagnetic
interference.
We tested the ranges
of four antennas: an
ordinary λ/4 stub, the
biquad, the cantenna,
and a commercial
model (the HAO14SD from Hawking Technology: see
Figure 12) costing
around fifty pounds,
with a quoted gain
of at least 14 dB. In
each case the antenna was connected to
the WLAN card in the
ure 4; deviations of up to 10 % from the
nominal dimensions should be tolerable for ordinary use. Of course, there is
plenty of scope for experimentation.
The trickiest part of construction is
connecting to the radiator. We can
start from a commercially-available Ntype RF connector. Figure 8 shows an
example of this type of connector with
a radiator (sometimes called ‘exciter’)
soldered to it. A suitable hole must be
made in the can to fit the connector.
Washers should be used when fitting
the connector to avoid damage to the
can. The antenna is now ready for use.
If it is to be used outside it is worth
considering waterproofing the connection to the radiator.
Adaptor cables from N-type connectors to SMA connectors or other types
are available ready made; alternatively,
it is easy, as well as cheaper, to make
up a suitable cable oneself. As noted
above, the antenna cable should be no
Web links
[1] Waveguide theory:
http://en.wikipedia.org/wiki/Waveguide_
%28electromagnetism%29
[2] Various antenna designs:
http://qdg.sorbs.net/qdgant.htm
elektor electronics - 12/2006
Electronics lab
PC using an RF cable three metres long.
We achieved the following results:
10 m
4) Cantenna: 26 m
Here again the cantenna comes out on top.
The performance of the commercial antenna teaches us two things: first, one should
not always believe in a manufacturer’s
sometimes rather optimistic gain figures
(the antenna is delivering an estimated gain
of around 6 dB rather
than 14 dB); and second, the DIY approach
often pays off!
Figure 12.
The commercial directional
WLAN antenna used for
comparison tests.
longer than necessary in order to get
the most benefit from the gain of the
antenna.
Biquad
12/2006 - elektor electronics
m
[5] Download page for NetStumbler:
http://www.netstumbler.com/downloads
m
[4] Boundary element method:
http://en.wikipedia.
org/wiki/Boundary_element_method
Figure 8. N-type connector with a tapered radiator made from
copper sheet soldered to it.
,5
[3] 4NEC2 software:
http://home.ict.nl/~arivoors
As shown in Figure 4, a suitable piece
of copper pipe can be used to fix the
biquad figure-of-eight. The pipe is soldered to the reflector and the RF coaxial cable passed through the pipe as
the feed. The central conductor of the
cable is then directly soldered to the
middle of the figure-of-eight. Alternatively, an N-type connector can be used
as with the cantenna, the correct distance to the reflector being achieved
using two pieces of copper wire of suitable length.
The figure at the beginning of the article shows the radiation pattern of a biquad antenna whose reflector is fitted
with two plates, 30 mm high, on opposite sides to attenuate the rearwardspointing lobes. Using this construction
a gain of between 10 dB and 12 dB can
be achieved. There are reports of laptops equipped with biquad antennas
connecting to an access point (also
with a specially-constructed antenna)
10 km away.
Figure 7. Radiation pattern of the cantenna calculated using
4NEC2.
30
An alternative design of antenna,
which is also easy to construct, is the
biquad. This takes the form of an angular figure-of-eight pattern of wire
in front of and parallel to a reflector
surface. The design has proved very
popular on the Internet, where there
are countless construction guides, no
doubt because of its good theoretical
performance and ‘high-tech’ appearance. The design described here has
the advantage, shared with the cantenna above, that it has been simulated and optimised by computer.
In essence the biquad is a folded multiple λ/4 dipole. As Figure 9 shows, the
resulting shape resembles a figure-ofeight. Each edge of the two squares
is λ/4 = 30.5 mm long. A suitable material is 1 mm copper wire. The feed
connection is made between the point
where the two squares meet and the
open ends, which are connected to
ground (the feed cable screen). Figure 10 clearly shows the current distribution due to the individual antenna elements. Current nodes and antinodes can be seen at the corners of the
square: the antenna is in resonance.
The figure-of-eight should be mounted approximately 15 mm to 17 mm in
front of the reflector. Practical experiments have shown that it is possible
to achieve an excellent SWR (standingwave ratio) of 1:1.15.
It is recommended that the side of the
reflector should be equal to one wavelength. In other words, the ideal reflector is a conducting square of metal
measuring 123 mm on each side. Various materials are suitable: in the prototype we found copper-clad printed
circuit board satisfactory. A reflector
could also be made from a CD (the
metallised part has diameter approximately 118 mm). The dimensions of the
multi-dipole are unfortunately rather
critical.
m
21 m
m
3) Biquad:
,5
2) HAO14SD: 20 m
30
1) Stub:
060056 - 12
Figure 9. Construction drawing for the figure-of-eight biquad
antenna.
Miscellanea
Tall tales of spectacular results obtained by avid WLAN hunters abound,
but it is true to say that there can be
enormous differences in practicallyachievable range depending on the
local townscape or countryside, on
Figure 10. Current distribution for the biquad antenna
calculated using 4NEC2.
73
know-how
rf
Antennas in practice
The most detailed calculations and highest technical specifications
count for nothing if good results are not achieved in practice. We
therefore decided to take the antennas we built in the Elektor Electronics laboratory according to the designs calculated by Stefan
Tauschek for a practical test. The most stringent test involved installing
the various antennas at the home of Thomas Scherer in Frankfurt city
centre and then using an ordinary Centrino laptop running NetStumbler to measure the signal strength in the street outside and determine over what range a connection could be achieved.
only 4 km away. The WLAN router is situated in a hallway on the third
floor, surrounded by walls. Even just 5 m away, on the floor above,
signal quality has dropped from ‘excellent’ to merely ‘good’. Inside
the building only four of the 21 WLANs shown in Figure 11 can be
received. The building thus makes an excellent test location..
Table 1 shows how far the radio waves propagate along the street
outside the building, after attenuation by one wall. The directional antennas were, of course, correctly aligned for the test. The first surprise
is that the biquad is clearly outperformed by the cantenna. The reason for this disagreement with the theoretical results was not found:
cable connections and the like were thoroughly checked. In the city
(and with one wall interposed between transmitter and receiver) the
cantenna offers approximately double the range of the ordinary stub
antenna. The performance of the biquad antenna sits between the
two: we eagerly await the comments of our expert readership in the
Elektor Electronics online forum!
To see what a directional antenna is capable of, we need to get away
from the electromagnetic smog of the city. To this end we moved the
test setup to a house on the outskirts of a small village. The antenna,
connected to an access point, was arranged to transmit from the
(open) front door of the house over the fields beyond. The measurements therefore give the line-of-sight performance of the antennas.
Besides the test WLAN, NetStumbler found two other WLANs in range, but both were at least six channels away from the test frequency.
Table 2. Ranges achieved in open countryside.
Figure 11. List of WLANs found by NetStumbler.
Antenna type
The screendump in Figure 11 was taken immediately outside the
building. It shows that in this are there are many WLANs competing
for the airwaves. The strongest signal, with SSID ‘IfPP Test Kanal 1’,
is being produced by the access point shown in Figure 2 set up for
this test.
Table 1. Typical antenna ranges in the city centre.
Antenna type
Distance
Stub
Biquad
Cantenna
20 m
–84 dB
–80 dB
–72 dB
30 m
–
–85 dB
–80 dB
40 m
–
–
–86 dB
The building is a five-storey reinforced concrete structure built in the
1980s. The walls screen radio signals so effectively that radio and
digital television reception is difficult, even though the transmitter is
house layout and construction material, and even on the neighbours! For
example, in Frankfurt city centre where
Thomas Scherer carried out his antenna field tests there is practically no
point where a laptop cannot pick up
signals from at least 15 WLANs, and
the same would go for any other major European city. To this we can add
interference from microwave ovens,
mobile telephones and other transmitters, all in or near the frequency band
we are interested in. Things are quieter (as yet) in the 5 GHz band used by
74
Distance
Stub
Biquad
Cantenna
40m
22 Mbit
48 Mbit
54 Mbit
60m
–
11 Mbit
54 Mbit
120m
–
–
5,5 Mbit
Table 2 shows that line-of-sight communication is possible over considerably greater distances than in the city. We have shown the communication rates achieved, as this is the most practically useful figure.
Communication over 120 m using a tin can is not a bad achievement, we think! The tripling of range achieved using the cantenna,
compared to the stub antenna, is in line with the theoretical gain
figure of 10 dB.
If both access point and laptop are equipped with directional antennas, ranges under these conditions of over 200 m can easily be
obtained. In this case the laptop must be used in conjunction with an
external WLAN adapter (either a PCMCIA card or connected via USB)
which has an RF connector, although this arrangement does make the
laptop rather unwieldy!
IEEE 802.11a WLANs. It is also worth
noting that the channels available in
Europe, numbered from 1 to 13 in the
IEEE 802.11b and 802.11g standards,
provide for only three non-overlapping
channels. A powerful WLAN run by a
neighbour can interfere with between
three and six channels on either side.
If problems with signal quality are encountered, the first thing to check is
what transmitters are active in the
neighbourhood. The NetStumbler program [5], a favourite of ‘wardrivers’
(people who drive around looking for
WLANs using a laptop) is helpful here.
It scans the radio frequencies in a configurable fashion and shows information about the various networks available, including their SSIDs and signal
strengths. Depending on the WLAN
hardware, the results might not always
be perfectly accurate, but the relative
values do usually give a good overview
of the situation.
(060056-I)
elektor electronics - 12/2006
ELECTRICAL SAFETY INFO & MARKET
In all mains-operated equipment certain
important safety requirements must be
met. The relevant standard for most
sound equipment is Safety of Information Technology Equipment, including
Electrical Business Equipment (European Harmonized British Standard BS
EN 60950:1992). Electrical safety under
this standard relates to protection from
• a hazardous voltage, that is, a voltage greater than 42.4 V peak or
60 V d.c.;
• a hazardous energy level, which is
defined as a stored energy level of
20 Joules or more or an available
continuous power level of 240 VA
or more at a potential of 2 V or
more;
• a single insulation fault which would
cause a conductive part to become
hazardous;
• the source of a hazardous voltage
or energy level from primary power;
• secondary power (derived from
internal circuitry which is supplied
and isolated from any power
source, including d.c.)
Protection against electric shock is
achieved by two classes of equipment.
Class I equipment uses basic insulation ; its conductive parts, which may
become hazardous if this insulation
fails, must be connected to the supply
protective earth.
Class II equipment uses double or
reinforced insulation for use where
there is no provision for supply protective earth (rare in electronics – mainly
applicable to power tools).
The use of a a Class II insulated
transformer is preferred, but note that
when this is fitted in a Class I equipment, this does not, by itself, confer
Class II status on the equipment.
Electrically conductive enclosures
that are used to isolate and protect a
hazardous supply voltage or energy
level from user access must be protectively earthed regardless of whether the
mains transformer is Class I or Class II.
Always keep the distance between
mains-carrying parts and other parts as
large as possible, but never less than
required.
If at all possible, use an approved
mains entry with integrated fuse holder
and on/off switch. If this is not available, use a strain relief (Figure, note 2)
on the mains cable at the point of entry.
In this case, the mains fuse should be
placed after the double-pole on/off
switch unless it is a Touchproof® type
or similar. Close to each and every fuse
must be affixed a label stating the fuse
rating and type.
The separate on/off switch (Figure,
note 4), which is really a ‘disconnect
device’, should be an approved doublepole type (to switch the phase and neutral conductors of a single-phase mains
supply). In case of a three-phase supply, all phases and neutral (where used)
must be switched simultaneously. A
pluggable mains cable may be considered as a disconnect device. In an
12/2006 - elektor electronics
approved switch, the contact gap in the
off position is not smaller than 3 mm.
The on/off switch must be fitted by
as shor t a cable as possible to the
mains entry point. All components in
the primary transformer circuit, including a separate mains fuse and separate
mains filtering components, must be
placed in the switched section of the
primary circuit. Placing them before the
on/off switch will leave them at a hazardous voltage level when the equipment is switched off.
If the equipment uses an open-construction power supply which is not
separately protected by an ear thed
metal screen or insulated enclosure or
otherwise guarded, all the conductive
parts of the enclosure must be protectively earthed using green/yellow wire
(green with a narrow yellow stripe – do
not use yellow wire with a green stripe).
The ear th wire must not be daisychained from one part of the enclosure
to another. Each conductive part must
be protectively ear thed by direct and
separate wiring to the primary ear th
point which should be as close as possible to the mains connector or mains
cable entry. This ensures that removal
of the protective earth from a conductive part does not also remove the protective ear th from other conductive
parts.
Pay particular attention to the metal
spindles of switches and potentiometers: if touchable, these must be protectively earthed. Note, however, that such
components fitted with metal spindles
and/or levers constructed to the relevant British Standard fully meet all insulation requirements.
The temperature of touchable parts
must not be so high as to cause injury
or to create a fire risk.
Most risks can be eliminated by the
use of correct fuses, a sufficiently firm
construction, correct choice and use of
insulating materials and adequate cooling through heat sinks and by extractor
fans.
The equipment must be sturdy:
repeatedly dropping it on to a hard surface from a height of 50 mm must not
cause damage. Greater impacts must
not loosen the mains transformer, electrolytic capacitors and other important
components.
Do not use dubious or flammable
materials that emit poisonous gases.
Shor ten screws that come too
close to other components.
Keep mains-carrying par ts and
wires well away from ventilation holes,
so that an intruding screwdriver or
inward falling metal object cannot touch
such parts.
As soon as you open an equipment,
there are many potential dangers. Most
of these can be eliminated by disconnecting the equipment from the mains
before the unit is opened. But, since
testing requires that it is plugged in
again, it is good practice (and safe) to
fit a residual current device (RCD)*,
rated at not more than 30 mA to the
1.
2.
3.
4.
5.
6.
7.
8.
9.
10.
Use a mains cable with moulded-on plug.
Use a strain relief on the mains cable.
Affix a label at the outside of the enclosure near the mains entry stating the
equipment type, the mains voltage or voltage range, the frequency or frequency range, and the current drain or curent drain range.
Use an approved double-pole on/off switch, which is effectively the ‘disconnect device’.
Push wires through eyelets before soldering them in place.
Use insulating sleeves for extra protection.
The distance between transformer terminals and core and other parts must
be ≥ 6 mm.
Use the correct type, size and current-carrying capacity of cables and wires
– see shaded table below.
A printed-circuit board like all other parts should be well secured. All joints
and connections should be well made and soldered neatly so that they are
mechanically and electrically sound. Never solder mains-carrying wires
directly to the board: use solder tags. The use of crimp-on tags is also good
practice.
Even when a Class II transformer is used, it remains the on/off switch whose
function it is to isolate a hazardous voltage (i.e., mains input) from the primary circuit in the equipment. The primary-to-secondary isolation of the
transformer does not and can not perform this function.
mains system (sometimes it is possible
to fit this inside the mains outlet box or
multiple socket).
* Sometimes called residual current
breaker – RCB – or residual circuit current breaker –RCCB.
These guidelines have been drawn up
with great care by the editorial staff of
this magazine. However, the publishers
do not assume, and hereby disclaim,
any liability for any loss or damage,
direct or consequential, caused by
errors or omissions in these guidelines,
whether such errors or omissions result
from negligence, accident or any other
cause.
3-core mains cable to BS6500 1990 with three stranded
conductors in thick PVC sheath
Max current
conductor size
Nom cond area
overall cable dia.
3A
16/0.2 mm
0.5 mm2
5.6 mm
6A
24/0.2 mm
0.75 mm2
6.9 mm
13 A
40/0.2 mm
1.25 mm2
7.5 mm
Insulated hook-up wire to DEF61-12
Max current
1.4 A
Max working voltage
1000 V rms
PVC sheath thickness
0.3 mm
conductor size
7/0.2 mm
Nom cond area
0.22 mm2
overall wire dia
1.2 mm
3A
1000 V rms
0.3 mm
16/0.2 mm
0.5 mm2
1.6 mm
6A
1000 V rms
0.45 mm
24/0.2 mm
0.95 mm2
2.05 mm
3-flat-pin mains plug to BS 1363A
75
know-how
wlan
Where am I – and W
Position determination using WLAN
Dennis Vredeveld
Wireless LAN (WLAN) has become incredibly popular in the last few years. Wireless
internet has now become very common in a corporate environment as well as in public
spaces and in our homes. But what not many people know, is
that WLAN can also be put to excellent use for other
applications, such as, for example, determining the
position of devices on the wireless network. How
this works you can read in this article.
Position determination technology is the basis for the now
universally known GPS (Global Positioning System). GPS
operates relatively accurately in the city and motorways,
but is not suitable for use inside buildings. The reason is
that the reception of the GPS signals from the satellites are
too weak inside buildings so that there is very little left of
the original accuracy.
An indoor equivalent of the GPS system is currently not
available and there is not even the slightest sign of a
possible standard. As a result we have to resort to other
76
systems, which, even though they were not developed
for this purpose, still provide surprisingly good results in
practice. Such as, for example, position determination
using WLAN!
WLAN – how does it work?
Devices with WLAN functionality generally have an integrated WLAN card. This card maintains the connection with the base station, the so-called Access Point (AP).
elektor electronics - 12/2006
here are the Others?
Wireless transfer of data can take place once a successful
connection has been established. Because the quality of
the connection can diminish – or even disappear completely – when the user moves, the card scans at regular
intervals for other Access Points that may be within range.
Based on different algorithms that are implemented by
the manufacturer in the driver for the card, the card can
decide to establish a connection with another AP and
terminate the old connection. This is called a ‘handover’. Which access point becomes the new connection
depends on the signal strengths of the various APs. The
signal strength is determined by the card based on data
packets that are sent by the AP specifically for this purpose. The packets are called beacons and are transmitted by the AP at regular intervals (about 10 times per
second). These beacons contain the unique MAC-address
of the AP, as well as the name of the wireless network, the
so-called SSID. Should the signal strength of the AP that
the card is currently connected to become too weak then
the card starts to search for better alternatives. This is only
valid of course if we have the use of a network with multiple APs (refer to the example in Figure 1).
Relationship signal strength-position
The method described here implies that there is a relationship between the position of the user (or more accurately:
of the WLAN card in the device that the user is carrying)
and the measured signal strength of the beacons that are
received from different APs. This relationship can also be
used to locate the user within the WLAN network. In principle two different methods can be thought of to achieve
this:
Figure 1.
This signal strength
information is normally
hidden from us by
Windows.
dows-API for communicating with network cards. In this
way it is possible to obtain the necessary information
from the WLAN-driver in a pre-defined manner, irrespective of the manufacturer. More information can be
found at [1] and [2].
Performing the calibration
Before our positioning system can be used, we must first
carry out the calibration. This consists of recording the
1. Simulation: We try to calculate the expected signal
strength at each position in advance based on a model.
Afterwards we compare the measured values and try to
find out where the user is based on the results from the
model.
2. Calibration: Instead of calculating the expected signal
strength for each location, the signal strength is measured
at different, predetermined locations. This information is
stored and used as data for comparison for subsequent
measurements when the user is in an unknown location.
Various tests have shown that the first method is not only
significantly more complicated, but is also less accurate
than the second. The reason for this is that the signal
propagation in an indoor environment is so very complex
that even very extensive models cannot take all existing
factors into account.
Reading Signal strengths
The question now arises how we are going to obtain
these seemingly very important signal strengths. Windows, after all, hides this information from us in the first
instance by translating the received signal strengths
into less accurate terms such as ‘very strong’ or ‘weak’.
The answer to this? NDIS! NDIS is a standard Win-
12/2006 - elektor electronics
Figure 2.
Visualisation of the signal
strength of an AP, based on
a simulation. Red means
large signal strength,
blue means weak signal
strength.
77
know-how
wlan
signal strengths of all access points at different locations,
scattered over the space that is covered by the LAN network. More points increase the accuracy, until the points
are less than one meter apart. At every calibration location we measure the received signal strength in four different directions for a few seconds for all APs. This information is stored in a database. The different orientations are
important because the human body has a measurable
influence in the received signal strength. There is therefore
a difference if we are facing the AP or if we are between
the AP and the WLAN device.
The author
Dennis Vredeveld (vredeveld@imst.de) is a software-architect at IMST GmbH in Kamp-Lintfort, Germany, where,
among other things, work is being done on a softwareframework for several indoor position determination technologies, including WLAN.
More information can be found at www.imst.com and
www.centrum21.de under the heading ‘ipos’.
Where am I?
Once this calibration has been completed, the actual position determination can begin. By measuring the signal
strengths of all APs at regular intervals and comparing
these with the calibrated locations, the most likely position
can be calculated. Because we will not usually be at a
calibration location, a weighted average is taken from the
calibration positions that correspond best with the signal
strength profile.
With additional mathematical tricks (such as calculating
a running average of the recent locations) the accuracy
of the system can be further improved. In this way we are
able to achieve a resolution down to a few meters; sufficient to determine in which room or in which corner of a
hall the user is located.
The smarter the algorithm, the better the position can be
determined. The development of a real clever algorithm
for determining the position is certainly no sinecure – we
are curious for your solution. If you have developed a
good method for this, then be sure to let us know!
(060269) Illustrations : IMST GmbH
Web links
[1] www.ndis.com
[2] http://msdn.microsoft.com/library/default.asp?url=/library/
en-us/wceddk40/html/cxconNDISDriverArchitecture.asp
Advertisement
Senior Power
Engineer
redlineplc.com
£35k - £55k
Cambridge, UK
Splashpower Ltd develops wireless power technology for recharging
portable consumer electronics devices. Having completed a major
phase of R&D, they are preparing to license their technology to
global OEM’s. Based in the vibrant and Historic town of Cambridge,
40 miles North of London, it is the centre of new and emerging
technology within the UK.
They are currently seeking a Senior Power Engineer to:
• Lead all Power circuit design and assessment from requirements
analysis through to acceptance and certification, providing
technical leadership to implementation teams
• Develop system level mapping, detail simulation tools
and Design Methodology
• Assist in the capture, and preparation of, design specifications
from primary requirements to the implementation of secondary
requirements
If you have extensive experience in low power electronics (sub 100
Watt), with knowledge of batteries, charging, and/or power systems
(e.g. SMPS), we’d like to talk to you.
If you want to work with some of Europe’s brightest Engineers contact
Lisa Hart on +44 (0)1582 878827 or Lhart@redlineplc.com
Redline
Recruiting Excellence
78
www.redlineplc.com
cv@Redlineplc.com
Luton: +44 (0)1582 450054
Manchester: +44 (0)161 491 7830
Scotland: +44 (0)1698 527120
elektor electronics - 12/2006
design tips
hands-on
Client-server quizmaster
Manuel Schreiner
We have published a number of
designs for such circuits over the
years in Elektor Electronics, but
admittedly none has been particularly straightforward, often
using microcontrollers and even
radio modules!
This circuit demonstrates that we
can do the job very simply, using just a couple of discrete (not
to say antique!) components.
Our terminology, on the other
hand, is bang up-to-date: we
have designed a client-server
quiz system.
The server (i.e., the quizmaster)
+VCC
S2
R1
1k
This circuit is designed for use
in quizzes involving several
players. The situation is a familiar one: whoever presses (or,
in practice, thumps) their button
first lights their lamp and is given the opportunity to answer the
question.
4x 1N4004
D4
D1
D2
D3
LA1
RE1
BT1
T2
S1
T1
BC557
RE2
BZ1
BC547
060218 - 11
is shown on the left in the circuit
diagram. It consists of a relay
(K2), a pull-up resistor (R1) and
a transistor. The relay is used to
switch a buzzer or bell. Switch
S2 is used to reset the system.
The client systems are connected
to the server and to one another using a three-wire bus. Each
client consists of a relay, a transistor and a couple of diodes.
One of the jobs of the relay is to
switch on the lamp on the contestant’s desk.
The three bus wires carry VCC,
ground, and a blocking signal. A
number of client systems, one per
contestant, can be connected to
the three wires. In the quiescent
state the blocking signal carries
the supply voltage VCC. When
one of the contestants’ lamps is
lit this drops to 0.7 V.
When a contestant presses his
button (S1) relay K1 pulls in as
a result of the voltage (the blocking signal minus 0.7 V) applied
to the base of transistor Q1. If
another contestant subsequently
presses his own button nothing
will happen: the voltage appearing at the base of his transistor
will now be 0 V, and so it will
not be turned on.
Relays with a working voltage of
6 V can be used, with a supply
voltage not exceeding 9 V.
(060218-I)
Pencil rubber cleans solder pads
Luc Lemmens
Just in case you didn’t know,
from 1 July 2006 all new electronics equipment has to be
RoHS-compliant. One of the implications of the new rules is that
solder tin containing lead may
no longer be used in newly produced equipment. Interestingly,
exceptions are made for automotive electronics, medical and
military equipment. Currently
there are some doubts regarding the mechanical durability of
the substances used to replace
the lead/tin alloy we’ve been
using for decades to solder our
circuits. The use of lead/tin solder is also still allowed for repair
work on older equipment. Consequently it will take a while before
the ‘old’ type of solder tin has
disappeared from electronics in
general.
Browsing the soldering and soldering tools section of just about
any catalogue from a major
electronics supplier, you’ll soon
be at a loss at what solder tin
to choose, the choice of alloys
and compounds being confusing
to say the least. Still, 90% of all
varieties have one thing in com-
12/2006 - elektor electronics
Now, silver has a nasty habit if
oxidising quickly. Just as table
silver, our boards seem to turn
black and dull after a while, especially if touched by fingers.
It is therefore recommended to
use airtight packaging for circuit
boards when in storage or transit. After all, solder tin will not
flow very well on silver oxide.
mon: the lead component has
been replaced by silver. Only
the tin/silver ratio varies. In most
cases, another metal has been
added to optimise the chemical
and thermal characteristics. Also,
differences exist in the flux type
and the amount of it added to
the solder tin.
Pre-tinned Elektor Electronics
printed circuit boards, too, had
to undergo a change in the pro-
duction process. Since a few
months, our board suppliers
apply chemical silver to cover
copper surfaces and so comply
with RoHs. It took us some getting used to when the first boards
started to arrive in our lab — to
us, it seemed as if the white
component overlay was covering the solder pads! Still, the
boards turned out to be perfectly
solderable.
If a circuit board shows black
smudges, problems in soldering may be prevented by solving the silver oxide traces with a
soft piece of pencil rubber, which
some of you may remember from
the days when a pencil was used
to make notes and drawings. The
silver-oxide spots and areas will
disappear remarkably quickly
and soldering will be a breeze
afterwards. If you solder immediately after applying the pencil
rubber, the problems are solved
for good. So far, we have not
observed significant oxidising in
joints made using the latest silver/tin based solder, so ‘polishing the silver’ is unlikely to become a recurrent subject in this
magazine.
(060230-11)
79
technology
radio remote control
Radio Control using
Dieter Perkuhn
Model aircraft these days fly at lightning
speed and with their backup systems can
cost as much as a small car. If
someone else’s control
system blocks yours, the
results can be dearly fatal for both aircraft!
A new broadband-based system could prove
an ideal solution.
Foto: Weatronic [5]
Until recently it was no exaggeration to describe radio
control (R/C) systems for plane, car and ship models
as utterly ‘stone age’, at least from a communications
technology point of view. Transmission techniques had
not moved forward since amplitude modulation (AM)
was generally ditched in favour of frequency modulation
(FM), and that was several decades ago. The standards
established at that time are largely still in use around the
world. Key points of this standardisation include using
the frequency bands 27, 35 and 40 MHz for control
signal transmission. Across Europe 35 MHz is reserved
exclusively for model aircraft control, whereas a multitude of other users have access to the 27 and 40 MHz
bands. The frequency bands are divided into channels
10 kHz wide, making this a narrowband modulation system. With no guardbands between individual channels,
it is technically simple for signals to bleed over into adjacent channels, requiring signals to be limited to 8 kHz
bandwidth if interference is to be avoided. Most of today’s R/C receivers use IF filters having a 3-dB bandwidth of around 6 kHz.
The simplest way of generating R/C transmissions is to
code the signal using Time Division Multiplex (TDM) technology. To control between 4 (minimum) and 12 (maximum) servo functions, a corresponding number of pulses
of variable width are generated sequentially with a repetition rate of around 20 ms, then used to modify the RF
carrier using frequency modulation. For this kind of coding the term Pulse Position Modulation (PPM) has been
defined. Over the years another system known as Pulse
Code Modulation (PCM) has also been implemented and
there is no single standard in use. Proprietary (manufacturer-specific) data compression systems reduce compatibility between PCM systems.
80
Since the signal structure does no more than distinguish
between two different amplitude levels, the modulation of
the RF carrier boils down to frequency switching between
two fixed values. Figure 1 shows a block diagram of a
current model aircraft R/C receiver using double superhet technology. Its architecture conforms to the classic
frequency conversion process using a first IF of 10.7 MHz
and a second IF of 455 kHz. Signal processing is handled by a microprocessor.
Interference
Unfortunately the interference problem is as old as the
remote control hobby itself. Interference in the airwaves is
both frequent and destructive, arising from many causes
but chiefly through use of the same radio channels by
more than one user simultaneously. In severe cases propagation effects can lead to near total signal blocking for a
moment or two, although total data loss over any extended period is rare.
Various measures can mitigate the problem of two users
occupying the same channel simultaneously. Frequency
band scanners built into the transmitters can prevent operation when it is detected that the selected channel is already occupied. This becomes a total solution only when
every user’s transmitter is equipped in this way, which is
seldom the case. The scanner is of no value of course if
another user cuts into the channel after the first owner’s
plane is already in flight.
A higher level of interference protection is achieved by
using more than one control channel simultaneously, as
in the case of a commercial system that uses one channel each in the 35 and 40 MHz bands at the same time.
The possibility of simultaneous interference on both chan-
elektor electronics - 12/2006
WLAN ICs
nels is more or less excluded, at the cost nevertheless of
increased hardware costs and greater failure risk arising
from the higher component count.
Broadband and DSSS
technology for model aircraft
controlledToshiba
gain
TA31124
RF amp
Toshiba
TA31136
Antenne
A broadband future?
With some model aircraft now using jet propulsion (the
heading illustration is a replica of the Albatros L-39 military jet) costing the same as a small car and representing a significant safety risk with air speeds of well over
300 km/h, the desire for fully interference-resistant R/C
systems is entirely comprehensible. Unfortunately, regulatory requirements and the need for backwards compatibility are hindering the introduction of any fundamentally
new R/C technology. On the other hand practically proven communication techniques have existed for a long time
that would adapt to model control extremely well. Examples taken from mobile radio include the DECT, WLAN,
Bluetooth and ZigBee standards. In all these applications
a multitude of point-to-point or user device-to-user device
radio links are operated bi-directionally and simultaneously in the same frequency domain.
Motorola
SC402082
Demodulator
input bandpass
35 MHz
1st IF filter
10.7 MHz
Servos
1 ... 10
2nd IF filter
455 kHz
plug-in 24.4 MHz xtal
(channel 70)
11.155 MHz
xtal
060009 - 12
3.41 MHz
xtal
Figure 1.
Block diagram of a
conventional remote
control receiver for model
aircraft (Photo: author).
The American Paul Beard and his firm Spektrum have
developed a radio R/C system for models that exploits
modern communication techniques and takes full advantage of cheap, off-the-shelf chipsets [1]. The initial offering, for R/C car models only, was RF modules for three
servo functions. This was a far cry from the latest product,
a fully airworthy system covering six servo functions with
the code number DX6. Figure 2 shows the transmitter
and receiver. The only restriction is that this control system
is intended only for so-called parkflyers and micro helis.
These craft have a range of 100 metres maximum.
The technology
Spektrum’s R/C system operates in the 2.4 GHz ISM
(industrial, scientific, medical) frequency band that is
available for use without a user licence in most countries. Consequently it is used by a multitude of applications including WLANs, Bluetooth and ZigBee. The effect
of these other applications is of minor significance to us,
since in the vast majority of cases the physical distance or
separation between these indoor users and our outdoor
R/C systems will be large enough to cause no difficulty.
The generous breadth of spectrum at our disposal, around
83 MHz (from 2.4 to 2.4835 GHz), enables modern digital modulation techniques to be used to their best advantage. Regulatory conditions lay down a spectral power
density of 10 mW per MHz of bandwidth, capped at a
maximum of 100 mW for the complete band. Depending
on the bandwidth of the signal being radiated, transmit
powers of between 10 and 100 mW are permissible. A
purely theoretical calculation indicates a potential transmission range of over 10 km with 100 mW transmit power, –90 dBm receiver sensitivity and 6 dB antenna gain
at transmitter and receiver—or 4 km using 10 mW. In a
radio controlled aircraft context several conditions would
have to be guaranteed to achieve this kind of range and
experience with WLANs and Bluetooth indicates the dis-
12/2006 - elektor electronics
Figure 2.
DX6 2.4 GHz model aircraft
transmitter with receivers
for six servo functions
(Source: Graupner [6]).
Figure 3.
Transmit RF module and
receiver for remote control
of model cars
(Photo: author).
81
technology
radio remote control
tances achieved in practice are frequently well below the
theoretical values.
To establish what might be realistic results, range tests
were carried out using transmit and receive modules
made by Spektrum for radio-controlling model cars (Figure 3). In these tests the transmitter and receiver were positioned around 1.5 metres off the ground (flat landscape,
ground covering damp, transmit and receive antennas in
direct line of sight with around 800 m separation). Under these conditions the link was rock-solid, without any
interference at all. However, as soon as either transmit or
receive antenna were blocked by human bodies the link
was lost altogether.
The Spektrum transmit module has an output power of
10 mW and thus conforms to specification ETS 300 328
for GSRDs (General Short Range Devices). By way of
comparison, data sheets for commercial WLAN routers
indicate they provide radiated power levels of 15 dBm
(equivalent to 31.6 mW), as they occupy a greater
bandwidth.
DI OVAL
DIO
IRQ
SS
SCK
MISO
MOSI
DSSS
Baseband
A
SERDES
B
DSSS
Baseband
B
RESET
PD
GFSK
Modulator
RFOUT
RFIN
CYWUSB6934Only
060009 - 13
Chips with everything
At the heart of Spektrum’s transmit and receive modules
is the CYWUSB6934 transceiver made by the U.S semiconductor producer Cypress Semiconductor Corporation [2]. Receive sensitivity is –90 dBm (7 µV into 50 Ω)
and transmit output power is 0 dBm (1 mW). Integrated
with this is a 13 MHz reference oscillator for the internal
frequency synthesiser. The oscillator is voltage-controlled
so that it can cover the complete 2.4-GHz ISM band. Its
circuit architecture reveals a single superhet with low IF
and integrated IF filter (Figure 4 gives a simplified block
diagram). According to the manufacturer the module is
intended for cordless applications in PC mouse, keyboard
und joystick applications, for game controllers, remote
controllers, barcode scanners and toys. To achieve the
output power of at least 10 dBm (10 mW) required for
remote control of models, the transmit module uses an
SE2526A power amplifier from SiGe Semiconductor [3]
to boost the signal. This amplifier module is normally used
in WLAN applications built according to IEEE 802.11b
and g specifications that provide RF output levels up to
20 dBm (100 mW). The chip has an integrated lowpass filter and an antenna changeover switch that makes
separate transmit and receive connections possible. The
transmit antenna is well matched and is connected using 50-Ω coax cable. Whether Spektrum actually limits
the output to 10 dBm would need precision measurement
to establish but the power is certainly appreciable. The
82
Construction and layout of transmitter and receiver printed circuit boards are shown in the following pictures. In
Figure 5 the transmit RF board can be seen together with
the subminiature co-ax socket for the antenna connection.
The track from the connector to power amplifier chip has
an impedance close to 50 Ω. Near the edge of the board
is the 13-MHz reference oscillator for the transceiver IC.
Figure 6 shows the transmit signal processing board
with its microprocessor and clock oscillator, Figure 7 illustrates the receiver RF board with the transceiver. The
simple wire antenna, which is not impedance-matched, is
taken direct to the receiver input with filtering. Finally in
Figure 8 we see the signal processing board for the receiver, complete with microprocessor and clock oscillator.
With no adjustable filters used in either the transmit module or the receiver, construction is both straightforward
and affordable. The VLSI chips used have a unit price of
less than five dollars (n quantities of 100 upwards).
Management matters
GFSK
Demodulator
Synthesizer
X13IN
X13
X13OUT
Figure 4.
Simplified block diagram of
the Cypress CYWUSB6934
transceiver
(Source: Cypress).
Digi tal
SERDES
A
author’s tests were confined to making relative measurements. In this process the transmitter was switched on and
off many times at random so as to test occupancy of all
possible 79 RF channels in the 2.4 GHz ISM band. The
total frequency range covered was determined at around
84 MHz, which squares up well with the permitted bandwidth of 83 MHz. The band occupancy of the signal in
use in one of the 79 RF channels was around 830 kHz,
corresponding to the channel spacing of 1 MHz.
Organisation of signal generation and processing is handled in both the transmit RF module and the receiver by
a Cypress CY8C27443-24PVI microprocessor [2]. The
Cypress transceiver is configured so that it operates in
one of the 79 possible channels within the 83 MHz wide
2.4-GHz ISM band. To achieve this, a scanning process
is initiated when the device is powered up. This means
that the Spektrum transmitter operates bi-directionally; the
receiver associated with the transmitter scans the band
and gives the go-ahead to the transmitter only when an
unoccupied channel is found. To avoid mistakes the transmit/receive management system ensures that the devices
do not transmit and receive simultaneously. Test measurement show that the transmit signal is pulsed with an ‘on’
time of just over 5 ms and 13 ms repetition rate.
The signal used to control the servo functions is not in fact
modulated directly onto the RF carrier. Cypress makes
use of a digital modulation system by the name of DSSS
(Direct Sequence Spread Spectrum) [4]. This is one of two
prominent digital modulation techniques, the other being FHSS (Frequency Hopping Spread Spectrum). DSSS
is used in WLANs, ZigBee, GPS and UMTS, with FHSS
employed by Bluetooth. Both techniques have their roots
in the military field. The FHSS technique involves signalhopping among the 79 channels of the ISM band 1,600
times a second, following a fixed sequence determined
individually between each transmitter and receiver.
Military origins
The remote control system that we are using employs
DSSS. In the process the narrowband desired signal is
first processed digitally so as to straddle a significantly
broader bandwidth and is only then modulated onto the
RF carrier. In this way the spectral power density is reduced to a level where the spread-out transmit signal dis-
elektor electronics - 12/2006
appears into the general noise background and can no
longer be detected using conventional methods (the military connection now becomes clear). The receiver, if provided with the same code, can reverse the spread process using what is called ‘processing gain’. The gain here
increases as the straddle code (‘chipping sequence’)
becomes extended. Any transmitters using the ‘wrong’
code will be heard as noise and ignored. It’s not all
gain, however, and there are nevertheless limits that are
set primarily by the limited processing power available.
The bitrate change of the chipping sequence used by
Spektrum for remote control amounts to 64 chips/bits
corresponding to a calculated gain of 10log10(64) =
18 dB. Various losses reduce this in practice to perhaps
16 dB. To achieve an acceptable signal-to-noise ratio of
circa 10 dB and system losses of around 2 dB for good
resistance to potential interference a signal processing
gain of more than 30 dB would be required. In a situation like this the power of the interfering signal might be
20 dB stronger than the wanted signal. This, however,
would imply a bitrate change or chipping code length of
more than 1,000. It’s easy to see how the system parameters for really good interference suppression run rapidly
out of control. It’s worth noting in addition that this process can work only when transmitter and receiver use the
same code. With Spektrum’s remote control the transmit
code is made known to the receiver at the start of operation using the so-called ‘binding process’ (it’s just conceivable that other receivers might be linked in too!).
Finally, here’s an interesting thought to consider: for data
transmission this new system employs one out of 79 channels each 1 MHz wide and prevents another user of the
same system from sharing the same channel. Resilience to
interference in the same channel is nevertheless minimal if
a more powerful user employing another system appears.
The WLAN system on the other hand employs three channels each 22 MHz broad and permits a limited number
of devices using the same system within a single channel.
And on account of the significantly greater channel bandwidth, WLAN is significantly more resistant to same-channel interference. It is nevertheless accepted that increasing
user numbers in a channel reduces the data rate. This
selfsame criterion is not acceptable for radio control,
however; real-time response takes top priority. This is
presumably the reason why Spektrum employs the system
described, even though it is less interference-resistant. A
further pointer in this direction is the fact that the new DX6
remote control for model aircraft actually occupies two
out of the 79 channels simultaneously and the receiver
is also built on a twin-channel basis (Figure 2). It would
be fascinating to learn how remote control systems of this
kind perform when several users are using the same type
of equipment concurrently.
Figure 5.
Transmitter RF board
with subminiature co-ax
connector for antenna
connection. The circuit
track from the connector to
the transceiver IC has an
impedance close to 50 Ω
(Photo: author).
Figure 6.
Transmitter signal
processing board with
microprocessor and clock
oscillator.
Figure 7.
Receiver RF board with
transceiver IC. The simple
wire antenna, which is
not impedance-matched,
is connected direct to the
receiver input without
filtering.
(060009-I)
Web links
[1] www.spektrumrc.com
[2] www.cypress.com
[3] www.sige.com
[4] http://en.wikipedia.
org/wiki/Direct-sequence_spread_spectrum
[5] www.weatronic.com
[6] www.graupner.de
12/2006 - elektor electronics
Figure 8.
Receive-side signal
processing board with
microprocessor and clock
oscillator.
83
hands-on
fpga
Paul Goossens and Andreas Voggeneder
A picture is worth a thousand words, which is probably why alphanumeric displays
have given way to graphic displays in the latest mobile telephones. Although our FPGA
prototyping board is not a small as a mobile phone, it can still generate video.
That’s why it has a VGA port.
As you have probably already guessed, the VGA port
can be used to display video imagery on a VGA monitor. We’ve kept the VGA hardware on the prototyping
board quite simple. Three D/A converters are implement-
84
ed using several resistors. Each converter has a resolution
of 3 bits. They generate the red, green and blue (RGB)
signals. The horizontal and vertical sync signals come
straight from the FPGA.
elektor electronics - 12/2006
VGA signals
Naturally, all five VGA signals are generated by the
FPGA. Before you delve into the details of the implementation, you need to know how theses signals ultimately
create a video image.
Here we limit ourselves to generating an image with
a resolution of 640 × 480 pixels and a frame rate of
60 Hz. Every VGA monitor can easily display this standard resolution.
Each frame consists of a number of lines of video information. A typical video line is shown in Figure 1. A video
line can be decomposed into a certain number of pixel intervals. One image pixel can be generated in each pixel
interval by the combined R, G and B signals.
The R, G and B signals must be low during the first 96
pixel intervals. The HSYNC signal is also low during this
time. The monitor recognises the start of a new line from
this set of signal states.
This is followed by 48 pixel intervals during which
HSYNC is high while the colour signals (R, G and B) remain low.
The actual image line starts next. For each pixel interval,
the R, G and B signals determine the colour and intensity
of the associated pixel. As you might expect, this portion
of the line consists of 640 pixel intervals – one for each
pixel in a horizontal line. At the end of the line, there are
another 16 pixel intervals during which the colour signals
are all low. HSYNC remains high during this time. A single video line consists of 800 pixel intervals in total.
It takes another 479 video lines after this one to send all
the pixels to the monitor. They are followed by ten video
lines that are fully black (RGB = 0). These lines are followed by another two lines (also black) with the VSYNC
signal low. This tells the monitor that the next frame is
coming. Finally, there are another 33 black video lines
with VSYNC high again. The total number of lines in a
frame is thus 525.
Example 1
The pixel frequency (dot rate) in this VGA mode is
25.167 MHz. The length of one pixel interval is thus the
inverse of this frequency, which is approximately 39.7 ns.
For this example, we can round off the frequency to
25 MHz, which is easy to generate from the 50-MHz
clock signal already present on the board. This dif-
12/2006 - elektor electronics
VGA functions
Various low-level routines that make it easier to use the
VGA interface are located in fpga_lib.c. The most important routines are:
Init_screen(): initialises the VGA controller
SetCurrentColor (Color):
sets the new foreground colour
SetCurrentBkColor (Color):
sets the new background colour
Gotoxy(x,y): places the cursor at the specified position
WriteScreen(Text): writes text to the screen starting at
the current cursor position
Clrscr(): erases the screen
Several constants that can come in handy when you’re
programming are located in infpga_lib.h and fpga_reg.h.
They include constants for the various colours.
fers from the desired frequency by approximately 0.7%,
which is fairly small and lies within tolerance.
Example 19 (ex19) generates a VGA signal that produces a colourful image on the monitor.
The code in file ex19.vhd contains a process named vid,
which generates the VSYNC and HSYNC signals. This
process also keeps track of which column and line of the
screen image is ‘at bat’. This information is stored in the
Col and Line registers.
VidEn indicates whether a pixel is being generated or
the R, G and B signals must be set to 0. Note that the Col
and Line signals are only valid when VidEn is set to 1.
The qen signal divides the 50-MHz clock signal by 2.
The rest of this process is very simple. Try to figure out for
yourself exactly how it works!
The fractal process uses the current coordinates (Col and
Line) to determine the intensities of the red, green and
blue signals. The algorithm used here is actually not significant – its only purpose is to generate a particular pattern on the monitor screen.
After you compile the example and load it in the
FPGA, connect a
VGA monitor
85
hands-on
fpga
DIY experiments
Example 19 forms a good basis for some simple DIY experimenting. Be sure to make a copy of the file first so you can
restore things to the original situation.
To get you started, try making the following changes:
- Delete lines 147–149
- Replace the graph process with the following code:
graph : process (clk, clr)
begin
if clr = ‘1’ then
Red <=”00”;
Blue <=”00”;
Green <=”00”;
elsif (clk’event) and (clk = ‘1’) then
if qen = ‘1’ then
if VidEn = ‘1’ then
if Col(3 downto 0)=”1000” then
Green<=”11”;
else
Green<=”00”;
end if;
else
Red
<= “00”;
Green <= “00”;
Blue <= “00”;
end if;
end if;
end if;
end process graph;
The above modification is quite simple. Now try making your
own modification to add a blue horizontal line to the image.
(Tip: the Line register keeps track of the currently active line.)
Once you’ve mastered this, how about generating a dashed
line instead?
Timing
Signal timing is very important in digital circuit design. For
instance, changes to a design can affect its maximum operating frequency.
Registers (flip-flops) are often used in digital designs. They
impose timing requirements on the associated signals to ensure that they work properly. The most commonly used timing
parameters are:
Tsu (setup): specifies how long the data must be valid at the
input of the flip-flop before the clock pulse arrives
Tco (clock to output): indicates how long it takes for the data
on the input to appear at the output after the clock pulse
arrives
Th (hold): specifies how long the data on the input must remain unchanged after the clock pulse arrives
An operation of some sort (such as AND, OR or even addition) usually occurs between the output of one flip-flop and the
input of the next one. These combinational circuits also have
delays between their inputs and the outputs. The interconnects
inside the FPGA also cause delays.
These delays and requirements have certain tolerance ranges.
For example, a flip-flop with a specified Tco of 2 ns may be
able to work even faster in practice. The longest possible time
is called the ‘worst-case time’, and the best possible time is
called the ‘best-case time’.
After Quartus compiles a design, it analyses the timing of the
design. This yields a figure for the maximum operating frequency of the design. To arrive at this result, the program calculates the maximum delay between the clock pulse and when
the signal arrives at the flip-flop data input for each signal
line. This delay is often called the ‘arrival time’. The Tsu of the
flip-flop concerned is included in this delay. Worst-case times
are used for this calculation.
86
The longest delay determines how quickly successive clock
pulses can follow each other. If you set a desired minimal
clock rate in Quartus (refer to Quartus Help), it can determine
whether the design meets your requirement.
Quartus also analyses the Th requirement of the flip-flops.
Here again it calculates the delay between the clock pulse and
when the data signal arrives at each flip-flop. The best-case
times are used for this second analysis. They must always be
greater than the Th requirement of the flip-flop concerned.
You can view the results of the timing analysis in Quartus in
the Compilation Report. This report includes a section named
‘Timing Analyzer’, which provides detailed information about
the timing of your design. The term ‘slack’ appears frequently in this report. Slack designates the difference between the
calculated delay and the desired delay based on your desired
clock frequency. A positive slack for Tsu indicates that the
signal arrives at the data input of a flip-flop earlier than the
required time (by the amount of the slack value). In the case
of Th, the slack indicates how much longer the signal remains
unchanged than the required (minimum) time. Positive slack is
good news, and negative slack is bad news.
If your design is too slow, you have two options. The first
option is to see whether Quartus can help you compile the
design better. To do this, select ‘Timing Optimization Advisor’
under Advisors in the Tools menu. It will help you configure
the Quartus compiler to produce a better result.
The second option is to examine the timing report to see
which signals are slowing things down. Once you know this,
you can try to modify your design to make these signals somewhat faster. You might find that using a different algorithm
helps.
There are also techniques that can be used to adapt a design
to a higher clock frequency, but that’s a different subject!
elektor electronics - 12/2006
to connector K9 on the prototyping board. If everything
went right, you will be treated to a pretty design on the
monitor.
Dynamic
Conjuring up a nice figure like this on your monitor is a
lot of fun, but you’ll get tired of looking at it after while.
It would be better (and more useful) to be able to display
information on the screen. If you combine this with a bit
of intelligence, you might just have something.
This is the purpose of Example 20 (ex20). Here the
trusty T51 microcontroller has been extended to include
the PS/2 interface described in the last instalment and a
VGA interface. The VGA interface is linked to the microcontroller via the XRAM bus. From the perspective of the
microcontroller, this interface is simply a block of memory
starting at address 0x8000 and a set of eight memory locations starting at address 0xAA00.
As you can see from the size of the Graffikkarte-a.vhd file,
this is a fairly sizeable design. For this reason, we don’t
intend to describe it in detail here. Based on what you’ve
already learned from the previous instalments, you should
be able to figure it out on your own.
However, it is good idea to say something about how the
interface is driven by the software.
300
96
48
96
16
HSYNC
R
G
B
060025 - 7 - 11
D
Q
D
Q
D
Q
Driver
Thanks to the C files provided with the course, the driver
routine that runs in the microcontroller is fairly simple. The
first thing ex20.c does is to call Initscreen(). This causes the VGA controller to be configured properly so it is
ready to be used.
Next, SetCurrentColor and SetCurrentBkColor set the image colours. Each character can be displayed in a different colour. Theses functions only change the colour that is
used when the data is written to the screen.
Next, Gotoxy(28,1) positions the (invisible) cursor to coordinates (28, 1).
Writescreen(“text”) places the first text on the screen.
Here the colour is determined by the previous SetCurrentColor and SetCurrentBkColor functions.
Finally, the putchar2() function displays individual characters on the screen. This function also uses the previous
colour information.
Use
A welcome message will appear on the monitor after you
configure the FPGA and connect a PS/2 keyboard. You
can then use the keyboard to type in text, which will appear immediately on the monitor. You can use the function
keys (F1 to F4) to change the colour of the text.
As an exercise to familiarise yourself with the VGA interface, you can experiment with modifying the software.
For instance, try integrating the I2C interface described in
one of the previous instalments so the output of the A/D
converter can be displayed on the monitor. Another idea
would be to display the scan codes received on the PS/2
interface instead of the decoded characters. If you can’t
get that to work, you can always ask other readers of Elektor Electronics for help via the Forum on our website.
Let’s have fun!
After all this hard work, it’s time for a bit of relaxation.
As a special treat, we’ve developed a fun application for
12/2006 - elektor electronics
Figure 1.
Structure of a video line.
CLOCK
060025 - 7 - 12
Figure 2.
Signal flow for a timing
analysis.
you. The hardware is exactly the same as for the previous
example (ex20). The only difference is in the software,
which has been modified ‘slightly’. Example 21 (ex21)
implements the familiar ‘four in a row’ game, but this time
entirely in the FPGA.
This example also shows how you can define your own
characters – that is, if you can tear yourself away from
the game long enough to have a look at the source code.
(060025-7)
Join the FPGA Course
with the
Elektor FPGA Package!
The basis of this course is an FPGA Module powered by an
Altera Cyclone FPGA chip, installed on an FPGA Prototyping Board equipped with a wealth of I/O and two displays (see the March 2006 issue).
Both boards are available ready-populated and tested.
Together they form a solid basis for you to try out the
examples presented as part of the course and so build personal expertise and know-how in the field of FPGAs.
Further information may be found on the shop/kits & modules pages at www.elektor.com
87
hands-on
e-blocks
A New Flowcode
Putting Flowcode 3
through its paces
Jan Buiting
A new version of Flowcode for E-blocks
has just been released — version 3. This
is more than a simple upgrade: Flowcode
has matured into a nice if not impressive
development tool.
Many of our readers are now quite familiar with Flowcode, the software brains behind all E-blocks projects,
with some having actually bought the product. A number
of E-blocks tutorials, reports and application examples
were published earlier this year.
Originally, Flowcode was designed to help College students develop electronic projects based on the popular
PIC microcontroller. As time has passed, more and more
features have been added to Flowcode, and the latest version 3 has actually turned out to be quite a nice development tool not just for the programmers starting out but
also for professional engineers.
In this respect, my own position is that I have been an
electronics enthusiast for 30-odd years and have occasionally worked with microcontrollers, specifically the
CDP1802 (see Retronics in the October 2006 issue). Fortunately, thanks to contributing authors and colleagues in
the Elektor lab I have never been out of touch with microcontrollers and related stuff like (E)PROMs and Flash devices. Although I can read PIC, 8051 and AVR assembly
code and the odd piece of C and Visual Basic, I must admit that I am not au fait with the latest in microcontroller
land when it comes to programming. So, when the new
88
Flowcode landed on my desk I thought I would see for
myself whether it lived up to my expectation: microcontroller programming should be easier than 20 years ago.
On the desk and on the PC
To undertake the review I got hold of the 30-day demo
copy of Flowcode 3, which is available as a free download from our website. At the time of writing I got Beta 5
— the final version will be on our website. I also got out
a set of E-blocks modules comprising a PICmicro Multiprogrammer board, an LED board, a Switch board and
an LCD board. The lot was connected up as shown in
Figure 1.
My colleagues in the lab having estimated my proficiency
(“zero”) and general chances of success (“we’ll see”),
gave me an 18-pin PIC16F88 device (worth £2.75) to
insert in the Multiprogrammer. Somehow I surmised that
this must be a low-end device but having looked up the
datasheet on the Microchip website I discovered the device has 7 kB ROM, 368 bytes of RAM, seven 10-bit A
to D channels, timers, a USART communication port and
a host of other things I had never heard of (BOR? LVD?).
elektor electronics - 12/2006
for E-blocks
Looks like things have moved on a little since I last used a
microprocessor! I was happy with my ‘low-end’ ‘F88 chip
because its simple architecture would prevent me from attempting the impossible (yet).
Having installed Flowcode 3 and the associated PPP driver (which was a breeze) I thought I would see if I was in
control of the hardware and software setup by getting a
single LED to flash. I know the example is corny, but you
have to walk before you run — writing that I2C bus driver
or 8-level DMA demultiplexer will come later!
The sequence
Right. In Flowcode I started a new flowchart, dragged
an Output icon onto the chart and clicked on it to get
the properties up. I set bit 1 of PIC port A to go On and
pressed the ‘PLAY’ button. The software simulated the ‘program’ and pin A0 on the PIC graphic faithfully went red
to indicate a logic 1 had been placed on port A bit 0. After a little trial and error I very quickly had a program that
flashed A0 on and off on the screen. No time wasted on
editing assembly code, re-assembling the code, burning
a new EPROM, stirring the coffee and finally relaunching
the application, fingers crossed and praying for success.
That was 20 years ago.
I learned a few things here: you need to slow Flowcode’s
simulator down so you can see what is going on in your
program. Correspondingly I deduced you need to put
delays in the program otherwise what I had hoped to be
a slowly flashing light will just look like a slightly dim LED.
You can see my first program in Figure 2. Not bad for
10 minutes work.
Flushed with success I then decided it was time to see this
in action on a real PICmicro device. Simulation is great
but I am only convinced by real-life electronics. To download a program to a microcontroller, in my case the lowly
16F88, you simply click on the small chip icon on the
menu and off it goes.
Off it went — but nothing happened on my hardware.
Time to read the Help file and the Getting Started guide,
both of which can only be described as crisply produced,
easy-going yet comprehensive documents with great
educational value. Having gone through this I concluded
that I needed to load the chip...configure screen to set up
the parameters for the real-life PIC: set the clock oscillator for a Crystal, turn off the watchdog timer, Brown Out
Detection and Disable Low Voltage Programming. These
settings, by the way, are now the number-1 problem experienced by readers having built an Elektor project based
on a PIC, having burned their own PIC device and finding
that it doesn’t work. Check those PIC config words!
Then I pressed the Download to Chip icon again and I
had my first program running on the real-life PIC. I looked
at my watch — I started 30 minutes ago. For a relative
newcomer who occasionally still gets nightmares in assembly code I did not think that was too bad.
Figure 1.
The E-blocks I connected up for my
baptism of fire, so to speak.
how the major features of Flowcode
can be used. Some
of these are based
on the ‘F88 device
I happened to have
used for my experiment
and others are based on
a larger ‘F877 device.
Flowcode can also operate on the much larger ‘18
series range of PICs as well
as the smaller ‘12 series. I
then spent the next hour or so looking through the example files for the ‘F88, downloading
a number of them to my hardware and launching the
applications.
Many only require the use of simple LEDs, switches and
LEDs, but some call for more advanced components
Learning curve
Having read the Help file I discovered that Flowcode 3 is
shipped with around 30 example files that demonstrate
12/2006 - elektor electronics
Figure 2.
My first program.
89
hands-on
e-blocks
What’s new & better
in Flowcode Version 3
Graphical User Interface improvements
•Zoom: Multiple zoom levels, zoom to fit
•Tiling: horizontal tile, vertical tile
•Smaller PICmicro MCU on-screen device
•Screen icons: new graphics, description now inside icons,
better comments, more icons per screen viewable
•Screen appearance: user selected icon shading, and back•
ground colour
Hardware and software macros now have separate icon
graphics now known as: ‘macro’ and ‘hardware macro’
•New: Print Preview and print to screen zoom setting
•Flowcharts can be exported to JPEG or BMP for incorporation into documents
•Tile horizontal and vertical and auto-arrange for multiple macro viewing
Improvements for migration to C
•All icons have bubble help to display icon function
•Icons can also produce equivalent C code of each icon as
•
bubbles
Students can view the C code equivalent of the whole
program
•Students can view the Assembly code equivalent of the whole
program
•Screen layout is preserved on save to allow educators to
build more relevant examples
•
•Tutorial files exploit features such as labelled components to
Tutorial file descriptions now included
add context
Multilingual support
•Main program and Help file: English, French, Dutch, Finnish,
German, Spanish
•Main program but not Help file: Chinese, Italian, Greek
Earlier in this series
Electronic Building Blocks, November 2005.
E-blocks and Flowcode, December 2005.
E-blocks in Cyberspace, January 2006.
E-blocks – now you CAN, February 2006.
E-blocks Making Waves, March 2006.
E-blocks Making Waves at C, April 2006.
From 2 to 3
Those of you who are currently using Flowcode version 2
will want to know what improvements have been made
to version 3. I have no experience of version 2 but looking in the Help file it states that major improvements have
been made to the graphical user interface, there are new
features to help learners, lots of functionality improvements such as 16-bit numbers, support for strings, better
interrupt handling, improved macros, more components,
and so on. A full list is given in the inset.
90
Figure 3. My final Decision Maker program.
Flowcode flips the coin
(i.e. E-blocks) like keypads and sensors. When you look
through these examples, Flowcode’s two main strengths
become clear. Firstly, it is a very good way for those that
are not front line programmers, like myself, to understand
how programs work and can be created in a short time.
This in turn means that the learning curve for Flowcode
itself is actually very steep.
Secondly, it is not a toy. Whilst it is great for those starting out to program, it is also quite powerful, with fully
supported interrupts and a range of communication protocols like SPI, RS232, CAN, Bluetooth and even TCP/IP
supported — all of which I will grind my teeth on — in
due course!
So, having trained myself up it was time to get down to
developing an application that, although serious in intention, is the tongue-in-cheek equivalent of flipping a coin: a
PIC-assisted Yes/No decision maker.
Some people, mostly managers, attribute great authority and powers of decision to PCs and microcontrollers,
arguing that they are 100% digital hence have no ‘grey
areas’ or room for ‘discussion’.
What I need to confirm these people in their beliefs (as
well as making them forget the simple coin in their wallet)
is a program where they could press a button and get a
‘yes’ or ‘no’ decision to display on the LCD. So I set about
making it — as an exercise, of course.
You can see the final program in Figure 3. It took me
about an hour in the end. I have put up the file ‘Decide.
elektor electronics - 12/2006
Software functionality improvements
•Better range of simulation speeds to check working program
before downloading it to the PICmicro microcontroller
Alter variables whilst simulation is paused
Support for 16-bit numbers and arithmetic, choice of types
includes CHAR, INT and STRING
Support for hexadecimal and binary numbers in all dialogue
boxes
Full support for strings including string manipulation commands like ADD, LEFT, RIGHT
Variables are now case sensitive
New string process icon supports string manipulation
Interrupt icon supports a larger range of interrupts as well as
custom interrupt definition. Each interrupts run a macro of
your choice.
Improved Delay icon with a much greater range of delays
While icon can operate for a defined number of times
Subroutines can now have parameters passed to them, and
returned
Larger range of supported devices now includes 18 series
PICmicro microcontrollers (technical specification for full list).
•
•
•
•
•
•
•
•
•
•
•
•Undo and Redo commands
•Improved C compiler
Component improvements
•LCD: range supported now includes 40-character 2-line; 20•
character 4-line etc.
Full LCD functions now supported with scroll and other
features.
•Switches can now be labelled, options for display vertical
•
and horizontal, left to right or right to left.
LEDs can now be labelled, options for display vertical and
horizontal;, left to right or right to left.
•New PWM (Pulse Width modulation) component for motor
•
control.
Analogue components available now include thermometer,
dial, or slider
•Target communications components now include RS232,
I2C, Internet web server, Internet TCP/IP, Bluetooth, CAN
bus and LIN bus
E-blocks for Prototyping, May 2006.
E-blocks and X-10, June 2006.
E-blocks Easy ARM Pack, September 2006.
Articles may be downloaded individually
from www.elektor.com.
An overview of available E-blocks and software is
available on the SHOP pages at www.elektor.com
fcf’ for free downloading with this article — the file
number is 065096-11.zip.
Unfortunately Flowcode does not have a random number
generator so I created a simple counter and then used the
MOD feature in a calculation icon to detect whether the
count was odd or even. This outcome was used to answer
Yes or No to whatever weighty decision the user had in
mind.
What impressed me is just how easy this was. The LCD
was very easy to incorporate in the program and it really
adds a lot of functionality to an electronic system.
The hardest part of the job was deciding how I was going to structure the program itself in terms of the logical
flow and the variables I needed to track the status of the
program. Once I had done this on paper, transferring it
to Flowcode was quite painless. The thing that most impressed me was that once I had got the program working
using Flowcode’s simulation mode, when I transferred it
to the hardware (this time having loaded the PIC config
bits!) it worked just like the simulation did.
Conclusion
To a newcomer, Flowcode 3 ‘does what it says on the
tin’. It was easy to get started and make a program, was
intuitive to use, and produced code that worked. There
are a few minor disappointments such as the lack of a
random number generator, which I happened to stum-
12/2006 - elektor electronics
Figure 4. An example of a more advanced Flowcode program.
ble on for the Decision Maker program, and the fact that
floating-point numbers are not supported as variables. It
does however seem to have more features and capabilities than you could shake a stick at. I was also struck by
the thought that it was more fun than the day job. Then I
realized that this was my day job!
(065096-I)
A full featured but 30-day limited demo version of Flowcode 3
can be downloaded free of charge from www.elektor.com.
Follow: magazine → volumes → 2006 → december → A New
Flowcode for E-blocks
91
infotainment
modding & tweaking
Intelligent Voltmeter
Miniature 3-channel A/D converter
Jeroen Domburg & Thijs Beckers
This month we’ll make use of the ATTiny13 again. This versatile microcontroller is used here
in a practical, compact voltmeter with an RS232 link for use ‘on the road’.
92
In this article we show you how to build a small device
that wouldn’t be out of place in the laptop case of an
electronics fanatic. It is a three-channel A/D converter/
voltmeter, connected via the serial port. Whether you
want to measure transistor characteristics, battery discharge characteristics or just measure the voltage at a
few points, this circuit can help you out as long as the
voltages are between 0 and 5 V.
protects the circuit from damage if the DTR line is somehow in the wrong state and outputs a negative voltage.
At the hart of the circuit is an Atmel ATTiny13. This IC has
the all-important 10-bit A/D converter and 1.1 V reference voltage integrated on the chip. The ATTiny13 is
available in a DIP package as well as an SMD package.
If there is a need for it, it’s therefore possible to construct
an even tinier version.
Description
Interface
One of the most important design criteria of this circuit
is the amount of space it takes up. To keep this as small
as possible we have left out as many components as we
could. The final design consists of just five components,
which were coaxed into a DB9 plug (see photos). The circuit uses RS232 for connecting to a PC, although the voltages don’t fully comply with the standards, as this saves
on the number of components required. There is one section that is conspicuous by its absence: the power supply.
This is in fact drawn from the serial port.
There is no UART present inside the chip, so we had to
implement this in software. For this reason there is also no
need to use a MAX232 or similar to invert the signals or
convert them into 12 V. The latter has been done in the
cheapest way possible: we’ve left it out altogether. The
RS232 signal generated by the microcontroller is connected directly to the RxD line of the PC. Although this isn’t according to the official specifications, in practice it seems
to work most of the time.
The circuit diagram shows how simple this circuit is. The
supply section consists of D1, IC1 and C1. When the serial port is in a ready state, the voltage on the DTR line is
usually somewhere between 8 to 12 V. The circuit gratefully makes use of this. The 78L05 and the electrolytic capacitor ensure that IC2 is supplied with a stable 5 V. D1
To convert the ±12 V of the TxD line from the PC into a
TTL level for the microcontroller we use a single 10 k resistor (R1). The internal ESD diodes of the GPIO pin of the
microcontroller short signals above 5 V and below 0 V to
the supply lines. R1 limits the current to a safe level.
The UART is, as mentioned earlier, emulated inside the microcontroller. A UART normally requires a stable clock to
You only need five electronic parts, one DB9 connector & cover and some
cable to build this project.
In this case a PCB would take up valuable space. Besides, it’s much more
fun to construct it like this.
elektor electronics - 12/2006
About the author
r in a Plug
operate, such as a crystal. This component is also missing
from the circuit. Here we’re using the internal RC oscillator of the ATTiny13 as the clock, which is not very accurate. To obtain a stable timing signal we make use of the
bit-clock of the serial signal from the PC. The first character received from the PC is analysed and the result is used
to generate the bit-clock for the UART.
Apart from the routines for the UART and the A/D converter, there is also a bit of software that implements an autoranging function. If required, the software can automatically switch the reference voltage for individual channels
between the built-in 1.1 V and the 5 V supply voltage.
Because the 5 V supply is a lot less stable than the 1.1 V
reference voltage, it is possible to turn off this feature.
In use
Because the circuit uses a microcontroller and RS232
signals, it is fairly straightforward to control. When it is
connected to a serial port you can use a terminal emulation program such as Hyperterm (Windows) or Minicom
(Linux). The serial port settings are: 1200 baud, no parity,
8 data bits and 1 stop bit. Pressing ‘enter’ will display the
initial menu. From the menu you can adjust several settings. Choosing ‘run’ will make the measured samples of
the three channels scroll down the display. These can be
looked at straight away, or they can be stored in a file by
the terminal program, so that they can be analysed at a
later date, using Excel for example.
Jeroen Domburg is a student at the Saxion Technical University in Enschede, the Netherlands. Jeroen is an enthusiastic hobbyist, with interests in microcontrollers, electronics and computers.
In this column he displays his personal handiwork, modifications and other interesting circuits, which do not necessarily have to be useful. In most cases they are not
likely to win a beauty contest and safety is generally taken with a grain of salt. But
that doesn’t concern the author at all. As long as the circuit does what it was intended for then all is well. You have been warned!
8
K1
1
6
1
2 RxD
6
7
5
3 TxD
8
4 DTR
9
5 GND
R1
10k
D1
1N4001
RST
IC2
PB2
PB1
PB3
PB0
PB4
7
2
3
ATTiny13
IC1
78L05
C1
C2
C3
GND
4
C1
1
SUB D9
065120 - 11
websites of the author [1] or Elektor Electronics [2]. You
are therefore free to modify the software if you wish to
change or add some extra functions.
(065120)
Web links:
[1] http://sprite.student.utwente.nl/~jeroen/projects/serad
[2] www.elektor.com
On the other hand…
An ‘instrument’ with so few components also has some
disadvantages. The inputs, for example, are not galvanically isolated from the serial port. Keep this in mind when
you take measurements from devices connected to the
same PC or even the same mains ring.
The circuit is also fairly easily affected by the voltage levels on the serial port. If the voltage on the DTR line drops
below 6 V, or the RxD line doesn’t accept a signal of only
5 V you’ll find that the circuit won’t function reliably.
The source code may be freely downloaded from the
And this is the end result.
Nobody would guess that this contains a three-channel voltmeter.
It’s a perfect fit.
This way we don’t need to buy a special enclosure either.
And this is what the computer display looks like. The three columns can
also be logged, for analysis at a later date.
12/2006 - elektor electronics
93
hands-on
propulsion
A Wire with Total Recall
Exploring the properties of ‘memory wire’
Burkhard Kainka
Memory metal is often thought
of as ‘a solution looking for
a problem’. It’s an intriguing
material; warming it makes
it return to its original shape.
The application described here
is academic in that it doesn’t
provide a solution to any great
problem but instead produces a
novel visual effect.
This design should provide some relief for those of you who are irritated
by the sight of continually blinking
LEDs. This circuit grabs attention
by swaying an LED gently back and
forth like a tree bending in the wind.
The movement could have easily
been produced by a servo or electric
motor but memory wire has the advantage of moving silently. This type
of wire is also known as Nitinol, indicating that it is an alloy of nickel and
titanium. The wire can be deformed
into any shape which is retained
until heat is applied then it returns
to its original shape. The original
shape is ‘programmed’ into the wire
by heating it to a much higher temperature. A bent wire returns to its
original straight shape or a straight
wire returns to its original pattern.
This looks particularly bizarre if the
original shape is something like a paper clip; straighten it out then apply
a little heat and watch it fold back up
into a paper clip!
94
Memory wire can also be used to provide mechanical propulsion; in this
setup the wire is fixed between two
points and then deformed by tensioning the wire. When heat is applied it
shrinks back to its original length producing a pulling force between the two
points. The wire is available in different diameters and from a number of
suppliers [2] [3] [4].
The Mechanics
This design repeatedly swings the LED
along a trapezoidal shaped path. The
motive force is provided by a 15 cm
length of 0.15 mm diameter memory
wire. It is necessary to put the wire
under sufficient tension so that when
heat is applied it shrinks by 4 to 5 %
back to its original length. Heating
is achieved by passing a current of
around 300 mA through the wire.
The wire is fixed at either end and
in the middle forming an angle of 90
degrees. The shortening and lengthening of these small lengths of wire
is barely perceptible so the movement is magnified by the mechanical leverage produced in the LED
fixture (see Figure 1). Two lengths
of insulated wire with hooked ends
provide tension to the memory wire
from a 1 cm diameter spring made
up of four turns of 0.5 mm copper
wire mounted on the circuit board.
A small steel spring could be substituted to provide the tension.
Each time current passes through one
section of the memory wire it contracts
and pulls the hook which makes the
LED sway. A 10 cm long lever produces around 1 to 2 cm of movement in the
LED which can be further increased by
lengthening the lever. The complete
mechanical structure and electronics are contained on the prototyping
board. Figure 2 is a side view of the
assembly and shows the memory wire
particularly well.
elektor electronics - 12/2006
Figure 1.
The memory wire is clamped at three points and tensioned
from a spring via two hooked wires.
The LED is fitted to the top of the spring.
Electronic control
The memory wire used in this application is called Flex-150 [2]; it has a
resistance of 50 ohms/m and can carry a maximum of 400 mA. Each 7.5 cm
length of wire has a resistance of almost 4 Ohm and should be used with a
voltage supply of less than 1.6 V. A single 1.5 V battery or 1.2 V rechargeable
battery can be used for test purposes. A microcontroller is used to switch
current through the wires so that the
LED repeatedly moves along a twoaxis path defined by values stored in
the software (Figure 3).
LED
A
C
+5V
060144- 12
B
The switching waveforms have a relatively small ‘on to off’ ratio so it is possible to switch current through the
memory wire directly from the 4 to
6 V supply (four NiMH cells would be
a suitable power source). The BC337
transistors can handle up to 1 A and
the PWM switching waveforms are
arranged so that there is never more
than one transistor conducting at any
one time. The program causes the LED
to sway between four positions.
Source code for the Tiny11 program
can be downloaded free of charge from
the Elektor Electronics website [5]. You
can also access additional memory
wire information along with some suggestions for further experimentation.
(060144)
Web links
Figure 2. A side view of the assembly shows the memory wire fixing.
Current through the memory wire causes it to pull the LED in that direction.
[1] www.stiquito.com
[2] www.mikromodellbau.de
+5V
[3] www.robotstore.com
[4] www.memory-metalle.de
C1
[5] www.elektor.com (month of publication)
100µ
1
2
3
PB5
IC1
R1
PB2
PB3
PB1
PB4
PB0
ATTiny11
Figure 3.
Current through the wires is switched by a microcontroller, this
ensures consistent, repeatable LED movement.
12/2006 - elektor electronics
100 Ω
A
8
7
6
5
R3
B
C
T2
220 Ω
R2
T1
BC337
D1
220 Ω
BC337
4
060144 - 11
95
TECHNOLOGY LAB TALK
Smaller is not Always Better
Faultfinding on inaccessible IC connections
Karel Walraven
Most of our readers will appreciate that miniaturisation has brought many benefits. It is
certainly the case that those small, modern mobile phones look better and are more
practical than those old-fashioned bricks you had to lug around (see Retronics elsewhere
in this issue). At least as long as they work. Should something go wrong, it seems hopeless to repair it, but it isn’t impossible. We’ll look into this in more detail here.
Have you ever wondered why it is that when you’re looking for a place on a map, you often find it nearer the
edge? That is because the edge of a map takes up a
much larger area than you would expect. For example,
say you have a map of 1 by 1 m. Half the area of this
map is taken up by a square in the middle with dimensions of 70 by 70 cm, and the other half is taken up by
a strip along the edge, which is just 15 cm wide. It doesn’t seem much, a strip of just 15 cm, but in practice you
find yourself staring at those wretched 15 centimetres for
half the time...
In a similar way, the introduction of ‘pinless’ components
96
has had a big impact on miniaturisation. The space taken
up by the pins along the edge is just wasted space really.
It would be much better if we could mount the ICs right
next to each other, with all connections made on the
underside. The so-called ‘Square Packs’ have therefore
proved to be very popular (see main photo). The area
gained becomes relatively bigger as the chips get
smaller. This is because the connecting pins are of a
fixed size, so smaller chips have a
relatively larger area
elektor electronics - 12/2006
taken up by the pins compared to larger chips. The difference in area can be as much as a factor of two. Just like
our example with the map, a lot of area is ‘lost’ by the
use of connecting pins along the edge.
One of the main rules in faultfinding is that you should
always take measurements at the pins of components.
After all, you can never be certain that there is a perfect
connection between the pin and the PCB track. With
square packs it is no longer possible to physically access
these connections, so we have to find some other way.
However, all is not lost when you’re unsure of one of
these connections. Although you can’t directly access this
connection, you can use a handy trick to test the connection with a multimeter (!). This technique also comes in
very handy for those of you who solder SMDs, for example, with our SMD Oven from the January 2006 issue.
When the circuit works straight away there is nothing to
worry about, otherwise there is a possibility that there
is a bad connection between one of the pads underneath a square pack (or another difficult to reach IC)
and the IC itself. Even boards manufactured on an
industrial scale have a certain percentage of failures,
and with some patience and a bit of luck it is possible
to repair these PCBs.
1
2
V+
Most multimeters have a setting that measures diode
properties. In this setting the forward voltage drop of the
diode is measured when a small test current (usually
about 1 mA) is passed through it. That’s exactly what
we’re looking for. First, check that your multimeter comes
equipped with this setting (Figure 1). Then try it out
with a normal silicon diode and then a Schottky diode.
These should give readings of about 0.65 and 0.35 V
respectively.
And now for the big trick: how can we use the diode-test
setting of a multimeter to check the connection of a pin
that is inaccessible? Fortunately, virtually every pin on an
IC has protection diodes built in (there are a few exceptions, such as oscillator circuits and open-collector outputs). Usually, there is a reverse-biased diode between
the pin and ground, and another between the pin and
the positive supply (Figure 2). We therefore connect the
positive probe of the multimeter to the ground of the circuit and connect the negative probe to the track going
to the pin. When there is a good connection between the
track and the chip we get a measurement of the internal
protection diode and the meter usually gives a reading of
about 0.6 to 0.7 V. On the other hand, if there is an
open circuit the meter won’t give a reading. It is fairly
obvious that when the fault is in the ground connection to
the IC, all measurements to the pins will show a faulty
connection.
Unfortunately, the readings don’t always give a clear-cut
result. This is because the track doesn’t only go to the
suspect pin, but also to other components, and these
extra connections will often influence the measurement.
You should therefore refer to the circuit diagram and estimate how much of an effect these connections have
(Figure 3). If there is a high-impedance resistor or FET
then they can be ignored. Should the track go to another
pin on an IC then it is impossible to know from which
pin you’re measuring the protection diode. In such cases
you’re left with just one option: you have to (temporarily)
cut the track.
V-
065103 - 12
3
IC
IC
IC
10k
IC
065103 - 13
(065103-I)
12/2006 - elektor electronics
97
infotainment
retronics
SSB receiver for 20 and 80 m (1987)
Jan Buiting, PE1CSI
Elsewhere in this issue you’ll find an article describing a shortwave receiver (150 kHz
- 30 MHz) with contemporary components like
a DDS synthesizer, an
ARM microcontroller
and an LCD. Valuable
as they are and adding
considerably to ease of
use, the micro and the
LCD are really just peripherals around the
real thing: the receiver
proper, i.e., the RF and
AF circuitry. Out of curiosity I delved into the
Elektor relics cabinet
and pulled out an almost intact example of
the 20/80 m SSB receiver we published in the
November 1987 issue.
About 20 years ago
(time flies!) no eye catching ‘design highlights’ or
‘quick specs’ boxes were
included in technical articles, which look grim and
dry by today’s standards,
but are really a wonder
of technical comprehensiveness. The style of the article
text is ‘techie’ throughout. I recall
writing it, the way it was done in
1987, and a success it was.
For the RF enthusiasts and old
hands among you, a quick spec
of the 1987 design would read
like this: single-conversion USB/
LSB receiver with 9-MHz IF; single
free-running DG-MOSFET VCO;
DG-MOSFET input stages with soft
bandpass and IF notches, manual
20/80 m antenna switching; DGMOSFET mixers; AGC acting on IF
amp, 2.2-kHz audio.
There were not too many design
quirks in this receiver. The single
(!) varicap-controlled VCO has a
frequency range of 5.0-5.5 MHz.
Mixing is additive for the 80-m
band (3.5+5.5 = 9 MHz) and
subtractive for the 20-m band
(14–5 = 9 MHz). Three parallel-connected 27-MHz quartz
crystals operating at their funda-
mental frequency (9 MHz) form
the IF filter with a bandwidth of
about 2 kHz.
The design has no fewer that
five dual-gate MOSFETs from
the BF98x series, which were extremely popular at the time, approaching – to a considerable
extent — the response and behaviour of valves. However the
devices were also notorious for
the ongoing confusion about their
pinning. Depending on the manufacturer and sometimes even the
production batch, the drain and
source pins on these devices were
identified in different ways (studs,
longer/shorter pin, etc.). To help
constructors of the receiver (and
prevent lengthy technical discussions over the telephone), the
DG-MOSFET circuit symbol and
all three pinout variants of the
BF981/982 were clearly printed
in the circuit diagram.
I realise the 1987 receiver has
just two amateur radio bands
as opposed to ‘general coverage’ for this month’s DDS/ARM
version but then the 20-m and
80-m bands are by far the most
popular hotspots in the shortwave spectrum and SSB is the
default mode for voice communications, local (80 m) and DX
(20 m). Also, a block diagram
and concluding piece of text in
the 1987 article explain how the
receiver can be used as the tuneable IF section of a 0-30 MHz
communication receiver. Food
for thought for advanced constructors of RF gear.
Having fitting two knobs that
were missing from the front panel of the 1987 receiver I gave it
a quick ‘electrical’ inspection —
everything was still in place as
assembled 20 years ago by Jan
Barendrecht our RF expert. The
receiver came alive when connected to a 20-m/80-m com-
bined dipole I had access to at my local radio
club. The results were
not bad, but not convincing either, particularly when compared to
a vintage Yaesu FRG7 I had also brought
along. The first, glaring, shortcoming is the
lack of a frequency readout — you really have
no idea where you are
on the band when turning that multiturn tuning
control. Not admitting
defeat, on 20 m I simply pitched reception of
a mid-strength Balkanish SSB station against
the FRG-7. The frequency stability, BFO and
AGC of the Elektor receiver were beyond reproach but the receiver
as a whole was let down
quite badly by none other than its audio section, more specifically,
the loudspeaker fitted
in the case! A world of
difference when I connected headphones,
and even better results
when I bypassed the
flaky LM386, tapping the audio
signal at the wiper of the volume
pot and feeding it to a dedicated
headphone amplifier.
Although the receiver has poor
prestage filtering, the BF982
MOSFETs go a long way in preventing intermodulation and
other effects owing to RF overloading. None the less it will require an antenna tuner / attenuator when the bands open up
due to propagation. Also, the
2.2-kHz bandwidth afforded by
the 27 MHz xtals appears to be
too wide nowadays and the IF
filter slopes are not accurately
defined. But then, the price tag
compared to an FRG-7… in
1987, of course!
(065089-I
A scanned copy of the 1987 article is available
free of charge from our website.
Go to MAGAZINE → December 2006.
Retronics is a monthly column covering vintage electronics including legendary Elektor designs. Contributions, suggestions and requests are welcomed; please send an
email to editor@elektor.com, subject: Retronics EE.
100
elektor electronics - 12/2006
puzzle
Hexadoku
Participate!
Please send your solution (the numbers in
the grey boxes) by email, fax or post to:
Puzzle with an electronic touch
Time flies when you’re having fun but also when... solving an Elektor
Hexadoku! Here’s the last puzzle for this year 2006 — we hope you or
your family members enjoy solving it.
If it’s your first attempt, don’t give up — persist and win a super prize!
The instructions for this puzzle
are straightforward. In the diagram composed of 16 x 16
boxes, enter numbers such
that all hexadecimal numbers
0 through F (that’s 0-9 and
A-F) occur once only in each
row, once in each column
and in each of the 4x4 boxes
(marked by the thicker black
lines). A number of clues are
given in the puzzle and these
determine the start situation.
All correct entries received
for each month’s puzzle go
into a draw for a main prize
and three lesser prizes. All
you need to do is send us the
numbers in the grey boxes.
Prize winners
The puzzle is also available
as a free download from our
website (Magazine → 2006 →
December).
The E-blocks Starter Kit
Professional goes to:
Tomas Bakke
(Horte, Norway).
2 3 4
1
B
C 3
5 6 9 F
8
4
5
5 8
3 A
0
F
6
9 D
A
E 7
1
9
9
A F 0
3 4
6
B 2
B
3 4 C
E
C
2
4 0
B
F 5
12/2006 - elektor electronics
The solution of the
October 2006 Hexadoku is:
754C1.
0 C A 8
D
7
4
0 D E
B 3
5
B E
4 6
A 7 9 8
2
9
F 3
4
B
1
8
6
2
C
C
3
0 A
2
4
5 C
1
D
8 D
1 A
1
5
F
9 0
F E
C
2 8
D
1
infotainment
5
8 1
Elektor Electronics Hexadoku
Regus Brentford
1000 Great West Road
Brentford TW8 9HH
United Kingdom.
Fax (+44)(0)208 2614447
Email:
editor@elektor-electronics.co.uk
Subject: hexadoku 12-2006.
The closing date is
1 January 2006.
The competition is not open to employees of Segment b.v., its business partners and/or associated
publishing houses.
An Elektor SHOP voucher
worth £35.00 goes to:
L. Bailey (Wheatley);
Andrew Larkin (Wivenhoe);
Jim Orchard (Chelmsford).
Congratulations everybody!
Solve Hexadoku
and win!
Correct solutions received
7 2
E
D
7
1
F 8
B 6
C
9 5
4
3 9
C
enter a prize draw for an
E-blocks
Starter Kit Professional
worth £248.55
and three
Elektor Electronics SHOP
Vouchers worth £35.00 each.
We believe these prizes should
encourage all our readers to
participate!
101
ELEKTORSHOWCASE
To book your showcase space contact Huson International Media
Tel. 0044 (0) 1932 564999 Fax 0044 (0) 1932 564998
ATC SEMITEC LTD
www.atcsemitec.co.uk
Thermal and current-sensitive components
for temperature control and circuit protection;
• NTC Thermistors
• Current Diodes
• Thermostats
• Re-settable Fuses
• Thermal Fuses
• Temperature Sensors
Call today for free samples and pricing
Tel: 0870 901 0777
Fax: 0870 901 0888
CONFORD ELECTRONICS
http://www.confordelec.co.uk
Lightweight portable battery/mains audio units
offering the highest technical performance.
Microphone, Phantom Power and Headphone
Amplifiers. Balanced/unbalanced signal lines
with extensive RFI protection.
FUTURE TECHNOLOGY DEVICES
http://www.ftdichip.com
FTDI designs and sells
USB-UART and USB-FIFO
interface i.c.’s.
Complete with PC drivers,
these devices simplify the task of designing or
upgrading peripherals to USB
DANBURY ELECTRONICS
AVIT RESEARCH
www.avitresearch.co.uk
USB has never been so simple...
with our USB to Microcontroller Interface cable.
Appears just like a serial port to both PC and
Microcontroller, for really easy USB connection to
your projects, or replacement of existing RS232
interfaces.
See our webpage for more
details. Only £29.99 inc vat.
BAEC
http://baec.tripod.com
"The British Amateur Electronics
Club Archive Website. Archiving
extracts from 140+ Newsletters from 19662002. Currently have interesting and useful
selected articles from 12 Newsletters. Also a
section about built electronics projects with
schematics and photos. Plus useful info.,
downloads and links. NO ADVERTS!"
BETA LAYOUT
www.pcb-pool.com
Beta layout Ltd Awardwinning site in both
English and German
offers prototype
PCBs at a fraction of the cost of the usual
manufacturer’s prices.
COMPULOGIC LTD
www.compulogic.co.uk
Internet Remote Control Starter Kit £139.99
Create a simple web based remote control
interface for many applications
• Miniature Web Server Module
• Analogue/Digital Module
• PSU
• Manuals, software, example HTML code
102
http://www.DanburyElectronics.co.uk
Transformer manufacturers since 1983. Visit our
new site! Also link directly to Mike Holme’s Valve/Tube DIY amplifier site, featuring our standard Audio
Transformers (Mains, Output, Chokes, PP, SE, etc).
EASYSYNC
http://www.easysync.co.uk
EasySync Ltd sells a wide
range of single and multiport USB to RS232/RS422
and RS485 converters at competitive prices.
ELNEC
www.elnec.com
• device programmer
manufacturer
• selling through contracted
distributors all over the world
• universal and dedicated device programmers
• excellent support and after sale support
• free SW updates
• reliable HW
• once a months new SW release
• three years warranty for most programmers
FIRST TECHNOLOGY TRANSFER LTD.
http://www.ftt.co.uk/PICProTrng.html
Microchip Professional C
and Assembly
Programming Courses.
The future is embedded.
Microchip Consultant / Training Partner developed
courses:
• Distance learning / instructor led
• Assembly / C-Programming of PIC16, PIC18,
PIC24, dsPIC microcontrollers
• Foundation / Intermediate
FUTURLEC
http://www.futurlec.com
Save up to 60% on
• Electronic Components
• Microcontrollers, PIC, Atmel
• Development Boards, Programmers
Huge range of products available on-line for
immediate delivery, at very competitive prices.
HEROS TECHNOLOGY LTD
www.herostechnology.co.uk
Introducing Modular Concept for
microcontrollers.
Suitable for Developers, Pre-production,
Educational and Hobby applications.
• WinPIC2006 USB full speed programmer.
• CPU microcontroller modules.
• Peripheral modules for all microcontrollers.
JLB ELECTRONICS
www.jlbelectronics.com
Suppliers of electrical / electronic parts and
consumables. Including:
• Cable ties / bases
• Tools / hardware
• Bootlace ferrules
• Connectors
• Solvent sprays & cleaners
• PVC Tape
• Heat Sink compound
KMK TECHNOLOGIES Ltd.
http://www.kmk.com.hk
Low Cost DIY Robotic Kits
and Computer
Controller Boards.
elektor electronics - 12/2006
products and services directory
LONDON ELECTRONICS COLLEGE
NEW WAVE CONCEPTS
SOURCEBOOST TECHNOLOGIES
MODular ElecTRONics
www.new-wave-concepts.com
Software for Hobbyists:
• Livewire - circuit simulation
software, only £34.99
• PCB Wizard - PCB design
software, only £34.99
• Circuit Wizard - circuit, PCB and breadboard
design software, only £59.99
Available from all Maplin Electronics stores and
www.maplin.co.uk
http://www.sourceboost.com
Next generation C compiler and
development products at highly
affordable prices:
• C, C++, and Basic compilers for PIC12, PIC16,
PIC18
• Modern IDE, with PIC simulator, source level
debugger and virtual devices.
• RTOS for PICmicro.
• PIC based controller and Development boards.
• Download and try for Free from
http://www.sourceboost.com
http://www.lec.org.uk
Vocational training and education for national
qualifications in Electronics Engineering and
Information Technology (BTEC First National,
Higher National NVQs, GCSEs and Advanced
Qualifications). Also Technical Management and
Languages.
www.modetron.com
• Plug and Program
• FREE application s/w
• Hobbyist ease-of-use
• Professional finish with enclosure
and LEXAN faceplate
• We will design and brand your
custom application
• Growing range of PSU’s, i/o modules, displays
and microcontrollers
MQP ELECTRONICS
http://www.mqpelectronics.co.uk
Leaders in Device
Programming Solutions.
• Online shop
• Low Cost Adapters for all
Programmers
• Single Site and Gang Programmers
• Support for virtually any Programmable Device
PCB WORLD
http://www.pcbworld.org.uk
World-class site: Your magazine project or
prototype PCB from the artwork of your choice
for less. Call Lee on 07946 846159 for details.
Prompt service.
ROBOT ELECTRONICS
http://www.robot-electronics.co.uk
Advanced Sensors and Electronics for Robotics
• Ultrasonic Range Finders
• Compass modules
• Infra-Red Thermal sensors
• Motor Controllers
• Vision Systems
• Wireless Telemetry Links
• Embedded Controllers
SHOWCASE YOUR COMPANY HERE
Elektor Electronics has a feature to help customers
promote their business, Showcase - a permanent
feature of the magazine where you will be able to
showcase your products and services.
• For just £220 + VAT (£20 per issue for eleven
issues) Elektor will publish your company name,
website adress and a 30-word description
• For £330 + VAT for the year (£30 per issue
for eleven issues) we will publish the above plus
run a 3cm deep full colour image - e.g. a product
shot, a screen shot from your site, a company
logo - your choice
Places are limited and spaces will go on a strictly
first come, first served basis. So please fax back
your order today!
I wish to promote my company, please book my space:
• Text insertion only for £220 + VAT • Text and photo for £330 + VAT
NAME:..............................................................ORGANISATION:................................................
JOB TITLE:......................................................................................................................................
ADDRESS:.......................................................................................................................................
SYTRONIC TECHNOLOGY LTD
www.m2mtelemetry.com
Supplier of wireless modules and accessories for
remote monitoring M2M applications.
• GSM/GPRS TCP/IP modules
• Embedded GSM/GPRS modem
• Development Kits
• GPS modules
• GSM/GPS antennas
• Adapter cables
Online ordering facilities.
Tel: 01728 685802
ULTRALEDS
http://www.ultraleds.co.uk
tel: 0871 7110413
Large range of low cost Ultra bright leds and Led
related lighting products. Major credit cards
taken online with same day depatch.
USB INSTRUMENTS
http://www.usb-instruments.com
USB Instruments specialises
in PC based instrumentation
products and software such
as Oscilloscopes, Data
Loggers, Logic Analaysers
which interface to your PC via USB.
..........................................................................................................................................................
...........................................................................TEL:.......................................................................
PLEASE COMPLETE COUPON BELOW AND FAX BACK TO 00-44-(0)1932 564998
VIRTINS TECHNOLOGY
COMPANY NAME .........................................................................................................................
www.virtins.com
PC and Pocket PC based
virtual instrument for
electronics enthusiasts,
students, professionals and
scientists, including sound
card real time oscilloscope,
spectrum analyzer, and signal generator. Free to
download and try.
WEB ADDRESS..............................................................................................................................
30-WORD DESCRIPTION.............................................................................................................
..........................................................................................................................................................
..........................................................................................................................................................
..........................................................................................................................................................
..........................................................................................................................................................
12/2006 - elektor electronics
103
Quasar Electronics Limited
PO Box 6935, Bishops Stortford
CM23 4WP, United Kingdom
Tel: 0870 246 1826
Fax: 0870 460 1045
E-mail: sales@quasarelectronics.com
Web: www.QuasarElectronics.com
All prices INCLUDE 17.5% VAT.
Postage & Packing Options (Up to 2Kg gross weight): UK Standard 3-7 Day
Delivery - £3.95; UK Mainland Next Day Delivery - £8.95; Europe (EU) £6.95; Rest of World - £9.95 (up to 0.5Kg).
!Order online for reduced price UK Postage!
Payment: We accept all major credit/debit cards. Make cheques/PO’s
payable to Quasar Electronics.
Call now for our FREE CATALOGUE with details of over 300 kits, projects,
modules and publications. Discounts for bulk quantities.
Credit Card
Sales
Ho! Ho! Ho! Christmas 2006 is on it's way
BUT DON'T PANIC!!
We have some fantastic gift ideas for young (and old) enquiring minds
Electronic
Project Labs
An electronics course
in a box! All assume no
previous knowledge
and require NO solder.
See website for full
details
Mechanical
Motorised
Wooden Kits
Robot Sensor - £19.95
Order Code EPLR20KT
Electronic Bell - £8.95
Order Code EAKEBKT
Future engineers can
learn about the operation of transmissions
steered through gears
or pulleys. Easy to
build, no glue or soldering required.
Electronic Motor - £8.95
Order Code EAKEMKT
30 in ONE - £15.95
Order Code EPL030KT
Digital Recording
Laboratory - £29.95
Order Code EPLDRKT
Solderless
Electronic
Project Kits
Generator - £8.95
Order Code EAKEGKT
130 in ONE - £37.95
Order Code EPL130KT
AM-FM Radio Kit - £6.95
Order Code ERKAFKT
300 in ONE - £59.95
Order Code EPL300KT
Short Wave Kit - £6.95
Order Code ERKSWKT
Room Alarm - £4.95
Order Code EAKRAKT
Hand Held Metal
Detector - £9.95
Order Code ELMDX7KT
Stegomech - £12.95
Order Code C21-602KT
Tyrannomech - £12.95
Order Code C21-601KT
Automech - £12.95
Order Code C21-605KT
Festive
Electronic
Project Kits
Coptermech - £12.95
Order Code C21-604KT
Musical LED Jingle
Bells - £17.95
Order Code 1176KT
Trainmech - £12.95
Order Code C21-606KT
Flashing LED Christmas Tree - £6.95
Order Code 4080KT
500 in ONE - £149.95
Order Code EPL500KT
Crystal Radio Kit - £6.95
Order Code ERKCKT
Metal Detector - £9.95
Order Code ELMDKT
Robomech - £12.95
Order Code C21-603KT
See our website
for even more
great gift ideas!
Secure Online Ordering Facilities Ɣ Full Product Listing, Descriptions & Photos Ɣ Kit Documentation & Software Downloads
Elektor Electronics (Publishing)
Regus Brentford
1000 Great West Road
Telephone +44 208 261 4509
Brentford TW8 9HH
Fax
+44 208 261 4447
United Kingdom
Email: sales@elektor.com
Order now using the Order Form in
the Readers Services section in this issue.
More information on www.elektor.com
CD-ROM
BESTSELLERS
USB TOOLBOX
This CD-ROM contains technical data about the USB
interface. It also includes a
large collection of data sheets
for specific USB components
from a wide range of manufacturers. There are two ways
to incorporate a USB interface
in a microcontroller circuit: add
a USB controller to an existing circuit, or use
a microcontroller with an integrated USB interface. Included on this CD-ROM are USB Basic
Facts, several useful design tools for hardware
and software, and all Elektor Electronics articles
£18.95 (US$ 34.95)
on the subject of USB.
Home Automation
This CD-ROM provides an
overview of what manufacturers offer today in the field of
Home Networking, both wired
and wireless. The CD-ROM
contains specifications, standards and protocols of commercially available bus and
network systems. For developers, there are datasheets of
specific components and various items with
application data. End-users and hobbyists will
find ready-made applications that can be used
£12.95 (US$ 22.90)
immediately.
ECD Edition 3
Elektor’s Components Database gives you easy access
to design data for over 5,000
ICs, more than 35,000 transistors, FETs, thyristors and
triacs, just under 25,000 diodes and 1,800 optocouplers.
All databank applications are
fully interactive, allowing the
user to add, edit and com£14.95 (US$ 26.50)
plete component data.
Microcontroller Basics
1
Microcontrollers have become an indispensable
part of modern electronics. They make things
possible that vastly exceed what could be done
previously. Innumerable applications show that
almost nothing is impossible. There’s thus every
reason to learn more about them. This book
offers more than just a basic introduction.
It clearly explains the technology using various
microcontroller circuits and programs written in
several different programming languages. In the
course of the book, the reader gradually develops
increased competence in converting his or her
ideas into microcontroller circuitry.
ISBN 0-905705-67-X
230 Pages
£18.70 (US$ 33.70)
Visual Basic
for Electronics Engineering Applications
2
This book is targeted towards those people that
want to control existing or home made hardware
from their computer. After familiarizing yourself
with Visual Basic, its development environment
and the toolset it offers are discussed in detail.
Each topic is accompanied by clear, ready to
run code, and where necessary, schematics are
provided that will get your projects up to speed
in no time.
ISBN 0-905705-68-8
476 Pages
£27.50 (US$ 51.50)
BESTSELLING BOOKS
Top-5
1 Visual Basic
3
for Electronics Engineering Applications
ISBN 0-905705-68-8 £27.50 (US$ 51.50)
2 Microcontroller Basics
ISBN 0-905705-67-X £18.70 (US$ 33.70)
3 PC-Interfaces under Windows
ISBN 0-905705-65-3 £25.95 (US$ 52.00)
4 Modern High-end Valve Amplifiers
ISBN 0-905705-63-7 £25.95 (US$ 52.00)
5 308 Circuits
ISBN 0-905705-66-1 £18.20 (US$ 37.00)
More information on www.elektor.com
Order o
www.ele
Order now using the Order Form in
the Readers Services section in this issue.
USB-stick with ARM
and RS232
GameBoy
ElectroCardioGraph
(November 2006)
(October 2006)
Assembled and tested board
PCB, ready built and tested
060006-91
050280-91
£ 79.90 / $ 149.95
£ 55.20 / $ 103.95
RC Servo Tester / Exerciser
PIC In-Circuit
Debugger/Programmer
(July/August 2006)
(October 2006)
Kit of parts including PCB,
programmed controller and
all components.
Kit of parts including
PCB, programmed
controller and all
components.
050348-71
040259-71
£ 34.50 / $ 64.95
£ 22.70 / $ 42.85
MORE READY-BUILT PROJECTS
ClariTy 300-W Class-T Amplifier
£
$
030217-91 Amplifier board with SMDs pre-fitted; cores for L1 & L2
34.50
55.70
050008-91 PCB, ready built and tested
050008-71 Matching enclosure
50.00
10.25
94.25
19.30
Electrosmog Tester
Flash Microcontroller Starter Kit
010208-91 Ready-assembled PCB incl. software, cable, adapter & related articles 69.00 112.50
Gameboy Digital Sampling Oscilloscope (GBDSO)
990082-91 Ready-assembled board, incl. the PC software and related articles 103.00 183.00
LPC210x ARMee Development System
040444-91 Processor board, ready-made and tested
Micro Webserver with MSC1210 Board
25.50
030060-91 Microprocessor Board, ready-assembled
044026-91 Network Extension Board, ready-assembled
044026-92 Combined package (030060-91 & 044026-91 & related articles)
NO. 361 DECEMBER 2006
Shortwave Capture
030417-1 PCB, bare (receiver board)
030417-2 PCB, bare (control & display boards)
030417-41 AT90S8515-8PC, programmed
NO. 360 NOVEMBER 2006
USB Stick with ARM and RS232
060006-1
060006-41
060006-91
060006-81
PCB, bare
AT91SAM7S64, programmed
assembled & tested board
CD-ROM, all project software
NO. 359 OCTOBER 2006
PIC In-Circuit Debugger/Programmer
48.05
75.90 142.95
44.50 83.95
117.50 220.95
www.thepcbshop.com
www.thepcbshop.com
11.40 21.45
11.00 20.75
27.60 51.95
79.90 149.95
5.20
9.75
050348-1 PCB
050348-41 PIC16F877, programmed
050348-71 Kit, incl. PCB, controller, all parts
5.20
17.90
34.50
050280-91 PCB, ready built and tested
55.20 103.95
GBECG – Gameboy ElectroCardioGraph
9.75
33.75
64.95
ECG using a Sound Card
040479-1 PCB
040479-81 CD-ROM, all project software
5.20
5.20
9.75
9.75
41.50
7.25
8.90
5.20
77.95
13.65
16.85
9.75
060221-11 Disk, all project software
060221-41 ATmega16, programmed
5.20
8.90
9.75
16.85
040398-11 Disk, PIC source & hex code
040398-41 PIC16F628A-20/P, programmed
5.20
5.50
9.75
10.35
060012-11 Disk, all project software
060012-41 PIC16C745, programmed
5.20
6.90
9.75
12.95
040172-11 Disk, project software
040172-41 PIC16F84(A), programmed
040172-71 Kit, incl. PCB, controller, all parts
5.20
10.30
22.70
9.75
19.40
42.85
030190-11 Disk, project software
030190-41 PIC16F873-20/SP, programmed
5.20
16.50
9.75
31.00
050146-11 Disk, project software
050146-41 AT90S2313-10PC, programmed
5.20
6.90
9.75
12.95
050233-11 Disk, project software
050233-41 PIC16F84, programmed
5.20
10.30
9.75
19.40
050259-11 Disk, project software
050259-41 AT90S2313, programmed
5.20
6.90
9.75
12.95
NO. 358 SEPTEMBER 2006
Elektor RFID Reader
060132-91
030451-72
060132-71
060132-81
PCB, ready assembled & tested, with USB cable
Standard back-lit LC display
Matching enclosure
CD-ROM, all project software
Experimental RFID Reader
DiSEqC Monitor
USB/DMX512 Converter
NO. 356/357 JULY/AUGUST 2006
RC Servo Tester/Exerciser
LED Thermometer
Toothbrush Timer
Easy Home Control
Universal LCD Module
Elektor Electronics (Publishing)
Regus Brentford
1000 Great West Road
Brentford TW8 9HH
United Kingdom
Tel.: +44 (0) 208 261 4509
Fax: +44 (0) 208 261 4447
Email: sales@elektor.com
nline at
ktor.com
Due to practical constraints, final illustrations and specifications
may differ from published designs. Prices subject to change.
See www.elektor.com for up to date information.
Kits & Modules
Elektor RFID-reader
(September 2006)
Ready-built and tested PCB with USB port for connection
to the PC. Including USB cable; not including display and
enclosure.
GameBoy Programmable Logic Controller
(July/August 2006)
GBPLC Module
- Read and write 13.56 MHz RFID cards
- MIFARE and ISO 14443-A compatible
- Programmable
Ready-assembled and tested
GBPLC Module and
Programming Interface
050190-91
060132-91
£ 84.95 / $ 159.95
£ 41.50 / $ 77.95
LC-display
GBPLC I2C I/O Box
030451-72
£ 7.25 / $ 13.65
Ready assembled and
tested board
Matching enclosure
060098-91
£ 84.95 / $ 159.95
060132-71
£ 8.90 / $ 16.85
Combined package
CD-ROM (all project software)
GBPLC Module and I/O Extension
060132-81
£ 149.80 / $ 279.00
£ 5.20 / $ 9.75
1-Wire Thermometer with LCD
060090-11 Disk, project software
060090-41 PIC16F84A-04CP, programmed
5.20
10.30
050190-1+2 PCBs, bare, GBPLC Module & Programming Interface
050190-51 Programmed PAL, EEPROM and Flash IC
050190-91 Ready-built and tested GBPLC Module and Programming Interface
11.70 22.00
11.00 20.75
84.95 159.95
060098-1 PCB, bare
060098-91 Ready-built and tested board
17.90 33.75
84.95 159.95
GBPLC - Gameboy PLC
GBPLC - I2C I/O Box
Binary Clock
020390-11 disk, project software
020390-41 PIC6C54-04/P, programmed
No. 355 JUNE 2006
FM Stereo Test Transmitter
050268-1
PCB
Network Cable Analyser
050302-1 PCB
050302-11 Disk, PIC source code
050302-41 PIC16F874-20/P
5.20
8.05
9.75
19.40
9.75
15.10
22.00
8.20
5.20
16.90
15.55
9.75
31.85
NO. 354 MAY 2006
Onboard OBD-2 Analyser
050176-72 Kit of parts, incl. 050176-1, 050176-2, 050176-42, all components,
excl. LCD and Case
050176-73 LCD, 4x20 characters with backlight
050176-74 Case, Bopla Unimas 160 with Perspex cover and mounting plate
050176-42 ATmega16, programmed
050092-71 OBD-2 Analyser: Kit of parts without cable
050092-72 OBD-2 Analyser: DB9 to OBD adapter cable
Mini ATMega Board
050176-1
PCB, includes adapter PCB 050176-2
NO. 353 APRIL 2006
Simple recharable A Cell Analyser
050394-1 PCB, bare
050394-11 Disk, PC Software
Universal SPI Box
050198-41 AT89C2051-24PC, Programmed
No. 352 MARCH 2006
Application Board for R8C/13
050179-92
050179-1
030451-72
030451-73
Ready-assembled board
PCB
LCD with backlight
Poly-LED display
FPGA-Prototyping board
050370-91 Ready assembled board
For subscribers
For non-subscribers
Telephone Eavesdropper
030379-1
11.70
CE !
I
R
P
ED
H
S
SLA
PCB
Versatile FPGA Module
040477-91 Ready assembled plug-on module
For subscribers
For non-subscribers
NO. 351 FEBRUARY 2006
Brushless Motor Controller
050157-41 ST7MC1, programmed
A 16-bit Tom Thumb
24.80
28.80
15.80
10.30
52.50
27.55
46.70
54.50
29.90
19.45
96.95
51.95
8.95
16.85
4.80
5.18
9.04
9.75
050179-91 R8C Starter Kit
050179-C5 Set of 5 pcs. R8C13 microcontroller only
No. 350 JANUARY 2006
95-watt Laptop PSU Adaptor
050029-1
PCB
Automatic Attic Window Controller
050139-11 Disk, PIC source & hex code
050139-41 PIC16F84A-20I/P, programmed
7.25
13.65
48.27
13.77
7.25
25.50
90.94
25.94
13.65
48.05
181.80 333.50
216.30 398.50
9.05
181.80 333.50
216.30 398.50
3.80
7.15
8.30
20.70
15.60
39.00
4.80
9.05
5.20
13.10
9.75
24.65
Products for older projects (if available) may be found on
our website www.elektor.com
home construction = fun and added value
17.05
info & market
sneak preview
Free CD-ROM from
Microchip/Labcenter/Elektor
!
In the January 2007 issue we launch EXPLORER-16, a 16-bit PIC microcontroller project brought to you
jointly by Microchip, Labcenter and Elektor Electronics magazine. The project is based on a number of hardware and software components, exclusively gathered for Elektor readers. The first instalment of the series
covers the installation and practical use of two Microchip software contributions to the free CD-ROM: the
MPLAB development suite for PIC micros and — buckle up — the C30 Compiler! Space allowing we will also
introduce you to the workings of Proteus VSM, Labcenter’s software that’s so powerful it can simulate the interaction between a microcontroller and any analogue or digital hardware connected to it!
Overclocking CPUs and Micros
For as long as we can remember, CPUs in PCs have been subjected to overclocking experiments, sometimes with catastrophic results. Whenever a
new CPU appears on the market, it does not take speed tweakers long to make things run faster. What’s rarely seen however is information on
overclocking of laptop PCs. In our January issue we venture to cover the subject in some detail with a Centrino-based laptop as the guinea pig.
Also…
Nixie Clock in Sputnik Design
Multi-purpose Milling Machine; LED Rotating Text Display;
Berlin Clock Remake; Texas Instruments Clock Generator;
FPGA Course (8); Hexadoku.
Entirely fitting a January issue we have several articles on clocks and clocking
systems. One of the pinnacles, we’re sure, is our Sputnik Clock which uses nixie
tubes for the time readout. The clock is designed to look like a miniature version
of the first man-made satellite to make it into an orbit around the earth.
RESERVE YOUR COPY NOW!
UK mainland subscribers will receive the magazine between 18 and 20 December 2006.
The January 2007 issue goes on sale on Thursday 21 December 2006 (UK distribution only).
Article titles and magazine contents subject to change, please check www.elektor.com.
newsagents order form
shop save / home delivery
Please save / deliver one copy of Elektor Electronics magazine for me each month
Name:
. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
Address: . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
Post code: . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
Telephone: . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
Date:
. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
Signature: . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
Please cut out or photocopy this form, complete details and
hand to your newsagent.
Elektor Electronics is published on the third Thursday of each
month, except in July.
.elektor.com www.elektor.com www.elektor.com www.elektor.com www.elektor.co
Elektor Electronics on the web
All magazine articles back to volume 2000 are available online in pdf format. The article summary and parts list (if applicable)
can be instantly viewed to help you positively identify an article. Article related items are also shown, including software downloads, circuit boards, programmed ICs and corrections and
updates if applicable. Complete magazine issues may also
be downloaded.
In the Elektor Electronics Shop you’ll find all other products
sold by the publishers, like CD-ROMs, kits and books. A
powerful search function allows you to search for items and
references across the entire website.
Also on the Elektor Electronics website:
•Electronics news and Elektor announcements
•Readers Forum
•PCB, software and e-magazine downloads
•Surveys and polls
•FAQ, Author Guidelines and Contact
108
elektor electronics - 12/2006
12-2006
Order Form
Please supply the following. For PCBs, front panel foils, EPROMs, PALs, GALs, microcontrollers and diskettes, state the part number and
description; for books, state the full title; for photocopies of articles, state full name of article and month and year of publication.
PLEASE USE BLOCK CAPITALS.
Price each
Description
Qty.
Total
(see reverse before ticking as appropriate)
£ 18.95
Bank transfer
CD-ROM Home Automation
£ 12.95
Cheque
(UK-resident customers ONLY)
Visual Basic for Electronics
Engineering Applications
Elektor RFID Reader
Matching enclosure
£ 27.50
£
8.90
Expiry date: .................................................
Verification code: .........................................
SWITCH ONLY:
Start date: ....................................................
Issue number: ..............................................
Prices and item descriptions subject to change.
The publishers reserve the right to change prices
without prior notification. Prices and item descriptions
shown here supersede those in previous issues. E. & O.E.
Sub-total
P&P
Total paid
Email
–
– 2006
Signature
*USA and Canada residents may
(but are not obliged to)
use $ prices, and send the order form to:
Old Colony Sound Lab
P.O. Box 876, Peterborough
NH 03458-0876. Tel. (603) 924-6371, 924-6526,
Fax: (603) 924-9467
Email: custserv@audioXpress.com
✁
EL12
12-2006
Elektor Electronics (Publishing)
Tel.: +44 208 261 4509
Fax: +44 208 261 4447
www.elektor.com
sales@elektor.com
Date
Order Form
Please send this order form to *
(see reverse for conditions)
Regus Brentford
1000 Great West Road
Brentford TW8 9HH
United Kingdom
Tel.
Subscription
Giro transfer
£ 41.50
Address + Post code
✁
METHOD OF PAYMENT
CD-ROM USB Toolbox
Name
✁
Order Code
Yes, I am taking out an annual subscription
to Elektor Electronics and receive a free
1 W Luxeon LED Torchlight.
METHOD OF PAYMENT
(see reverse before ticking as appropriate)
Bank transfer
Cheque
(UK-resident customers ONLY)
Giro transfer
I would like:
Standard Subscription (11 issues)
Subscription-Plus
(11 issues plus the Elektor Volume 2006 CD-ROM)
* Offer available to Subscribers who have not held a subscription
to Elektor Electronics during the last 12 months. Offer subject to availability.
See reverse for rates and conditions.
Expiry date: .................................................
Verification code: .........................................
SWITCH ONLY:
Start date: ....................................................
Name
Issue number: ..............................................
Address + Post code
Please send this order form to
Elektor Electronics (Publishing)
Tel.
Email
Date
* cross out what is not applicable
EL12
–
– 2006
Signature
Regus Brentford
1000 Great West Road
Brentford TW8 9HH
United Kingdom
Tel.: +44 208 261 4509
Fax: +44 208 261 4447
www.elektor.com
subscriptions@elektor.com
ORDERING INSTRUCTIONS, P&P CHARGES
Except in the USA and Canada, all orders, except for subscriptions (for which see below), must be sent BY POST or FAX to our Brentford address
using the Order Form overleaf. On-line ordering: http://www.elektor-electronics.co.uk
Readers in the USA and Canada may (but are not obliged to) send orders, except for subscriptions (for which see below),
to the USA address given on the order form. Please apply to Old Colony Sound for applicable P&P charges. Please allow 4-6 weeks for delivery.
Orders placed on our Brentford office must include P&P charges (Priority or Standard) as follows:
UK: £4.00 Europe: £5.00 (Standard) or £7.00 (Priority) Outside Europe: £8.00 (Standard) or £12.00 (Priority)
HOW TO PAY
All orders must be accompanied by the full payment, including postage and packing charges as stated above or advised by Customer Services staff.
Bank transfer into account no. 40209520 held by Elektor Electronics (Publishing) / Segment b.v. with ABN-AMRO Bank, London. IBAN: GB35
ABNA 4050 3040 2095 20. BIC: ABNAGB2L. Currency: sterling (UKP). Please ensure your full name and address gets communicated to us.
Cheque sent by post, made payable to Elektor Electronics (Publishing) / Segment b.v.. We can only accept sterling cheques and bank drafts
from UK-resident customers or subscribers. We regret that no cheques can be accepted from customers or subscribers in any other country.
Giro transfer into account no. 34-152-3801, held by Elektor Electronics (Publishing) / Segment b.v. Please do not send giro transfer/deposit
forms directly to us, but instead use the National Giro postage paid envelope and send it to your National Giro Centre.
Credit card VISA, Access, MasterCard, JCBCard and Switch cards can be processed by mail, email, web, fax and telephone. Online ordering
through our website is SSL-protected for your security.
COMPONENTS
Components for projects appearing in Elektor Electronics are usually available from certain advertisers in this magazine. If difficulties in the supply
of components are envisaged, a source will normally be advised in the article. Note, however, that the source(s) given is (are) not exclusive.
TERMS OF BUSINESS
Delivery Although every effort will be made to dispatch your order within 2-3 weeks from receipt of your instructions, we can not guarantee this
time scale for all orders. Returns Faulty goods or goods sent in error may be returned for replacement or refund, but not before obtaining our
consent. All goods returned should be packed securely in a padded bag or box, enclosing a covering letter stating the dispatch note number. If the
goods are returned because of a mistake on our part, we will refund the return postage. Damaged goods Claims for damaged goods must be
received at our Brentford office within 10-days (UK); 14-days (Europe) or 21-days (all other countries). Cancelled orders All cancelled orders
will be subject to a 10% handling charge with a minimum charge of £5·00. Patents Patent protection may exist in respect of circuits, devices,
components, and so on, described in our books and magazines. Elektor Electronics (Publishing) does not accept responsibility or liability for failing
to identify such patent or other protection. Copyright All drawings, photographs, articles, printed circuit boards, programmed integrated circuits,
diskettes and software carriers published in our books and magazines (other than in third-party advertisements) are copyright and may not be
reproduced or transmitted in any form or by any means, including photocopying and recording, in whole or in part, without the prior permission
of Elektor Electronics (Publishing) in writing. Such written permission must also be obtained before any part of these publications is stored in
a retrieval system of any nature. Notwithstanding the above, printed-circuit boards may be produced for private and personal use without prior
permission. Limitation of liability Elektor Electronics (Publishing) shall not be liable in contract, tort, or otherwise, for any loss or damage suffered
by the purchaser whatsoever or howsoever arising out of, or in connexion with, the supply of goods or services by Elektor Electronics (Publishing) other
than to supply goods as described or, at the option of Elektor Electronics (Publishing), to refund the purchaser any money paid in respect of the goods.
Law Any question relating to the supply of goods and services by Elektor Electronics (Publishing) shall be determined in all respects by the laws
of England.
January 2006
SUBSCRIPTION RATES FOR ANNUAL
SUBSCRIPTION
United Kingdom
Surface Mail
Rest of the World
USA & Canada
Airmail
Rest of the World
USA & Canada
Standard
£41.90
Plus
£48.80
£54.50
US$ 95.50
£61.40
US$106.50
£68.90
US$120.00
£75.80
US$131.00
HOW TO PAY
SUBSCRIPTION CONDITIONS
Bank transfer into account no. 40209520 held by Elektor Electronics
(Publishing) / Segment b.v. with ABN-AMRO Bank, London. IBAN: GB35
ABNA 4050 3040 2095 20. BIC: ABNAGB2L. Currency: sterling (UKP).
Please ensure your full name and address gets communicated to us.
The standard subscription order period is twelve months. If a permanent change of address during the subscription period means that
copies have to be despatched by a more expensive service, no extra
charge will be made. Conversely, no refund will be made, nor expiry
date extended, if a change of address allows the use of a cheaper
service.
Student applications, which qualify for a 20% (twenty per cent) reduction in current rates, must be supported by evidence of studentship
signed by the head of the college, school or university faculty. A
standard Student Subscription costs £33.50, a Student SubscriptionPlus costs £40.40 (UK only).
Please note that new subscriptions take about four weeks from receipt
of order to become effective.
Cancelled subscriptions will be subject to a charge of 25% (twentyfive per cent) of the full subscription price or £7.50, whichever is the
higher, plus the cost of any issues already dispatched. Subsciptions
cannot be cancelled after they have run for six months or more.
Cheque sent by post, made payable to Elektor Electronics (Publishing)
/ Segment b.v.. We can only accept sterling cheques and bank drafts
from UK-resident customers or subscribers. We regret that no cheques
can be accepted from customers or subscribers in any other country.
Giro transfer into account no. 34-152-3801, held by Elektor
Electronics (Publishing) / Segment b.v. Please do not send giro transfer/
deposit forms directly to us, but instead use the National Giro postage
paid envelope and send it to your National Giro Centre.
Credit card VISA, Access, MasterCard, JCBCard and Switch cards can
be processed by mail, email, web, fax and telephone. Online ordering
through our website is SSL-protected for your security.
January 2006
New
ISBN 90-5381-212-1
£ 18.95 / US$ 34.95
CD-ROM
USB TOOLBOX
Embedded
USB Know How
This CD-ROM contains all the essential
information a designer needs to start working
with the USB interface. It includes a large
collection of data sheets for specific USB components from a wide
range of manufacturers.
There are two ways to incorporate a USB interface in a microcontroller
circuit. You can add a USB controller to an existing circuit, or use
a microcontroller with an integrated USB interface. Both options are
available on this CD-ROM.
Order now using the Order Form
in the Readers Services section
in this issue.
USB Toolbox provides information on all ICs suitable for different
applications. A sub-division has been made in controllers, hubs, microcontrollers and others. What is perhaps more interesting for many
designers however, is the extremely extensive software collection which
contains drivers, tools and components for Windows, Delphi and various
microcontroller families. Of course, none of the Elektor Electronics
articles on the subject of USB are missing on this CD-ROM
Elektor Electronics (Publishing)
Regus Brentford
1000 Great West Road
Brentford TW8 9HH
United Kingdom
Tel. +44 208 261 4509
See also www.elektor.com
INDEX OF ADVERTISERS
ATC Semitec Ltd, Showcase . . . . . . . . . .www.atcsemitec.co.uk . . . . . . . . . . . . . . . . . .102
London Electronics College, Showcase . .www.lec.org.uk . . . . . . . . . . . . . . . . . . . . . . . .103
Avit Research, Showcase . . . . . . . . . . . . .www.avitresearch.co.uk . . . . . . . . . . . . . . . . .102
MQP Electronics, Showcase . . . . . . . . . .www.mqpelectronics.co.uk . . . . . . . . . . . . . . .103
BAEC, Showcase . . . . . . . . . . . . . . . . . . .http://baec.tripod.com . . . . . . . . . . . . . . . . . .102
New Wave Concepts, Showcase . . . . . . .www.new-wave-concepts.com . . . . . . . . . . . .103
Beijing Draco . . . . . . . . . . . . . . . . . . . . . .www.ezpcb.com . . . . . . . . . . . . . . . . . . . . . . . .37
Newbury Electronics . . . . . . . . . . . . . . . .www.newburyelectronics.co.uk . . . . . . . . . . . . .22
Beta Layout, Showcase . . . . . . . . . . . . . .www.pcb-pool.com . . . . . . . . . . . . . . . . . . .7, 102
Number One Systems . . . . . . . . . . . . . . .www.numberone.com . . . . . . . . . . . . . . . . . . . .69
Bitscope Designs . . . . . . . . . . . . . . . . . .www.bitscope.com . . . . . . . . . . . . . . . . . . . . . . .3
Nurve Networks . . . . . . . . . . . . . . . . . . . .www.xgamestation.com . . . . . . . . . . . . . . . . . .22
ByVac . . . . . . . . . . . . . . . . . . . . . . . . . . .www.byvac.co.uk . . . . . . . . . . . . . . . . . . . . . . .69
PCB World, Showcase . . . . . . . . . . . . . . .www.pcbworld.org.uk . . . . . . . . . . . . . . . . . . .103
Compulogic, Showcase . . . . . . . . . . . . . .www.compulogic.co.uk . . . . . . . . . . . . . . . . . .102
Pico . . . . . . . . . . . . . . . . . . . . . . . . . . . . .www.picotech.com . . . . . . . . . . . . . . . . . . . . . . .7
Conford Electronics, Showcase . . . . . . . .www.confordelec.co.uk . . . . . . . . . . . . . . . . . .102
Cricklewood . . . . . . . . . . . . . . . . . . . . . . .www.cctvcentre.co.uk . . . . . . . . . . . . . . . . . . . .69
Danbury, Showcase . . . . . . . . . . . . . . . . .www.DanburyElectronics.co.uk . . . . . . . . . . . .102
Easysync, Showcase . . . . . . . . . . . . . . . .www.easysync.co.uk . . . . . . . . . . . . . . . . . . . .102
Elnec, Showcase . . . . . . . . . . . . . . . . . . .www.elnec.com . . . . . . . . . . . . . . . . . . . . . . .102
Eptsoft . . . . . . . . . . . . . . . . . . . . . . . . . . .www.eptsoft.com . . . . . . . . . . . . . . . . . . . . . . .78
Eurocircuits . . . . . . . . . . . . . . . . . . . . . . .www.eurocircuits.com . . . . . . . . . . . . . . . . . . . .6
First Technology Transfer Ltd, Showcase .www.ftt.co.uk . . . . . . . . . . . . . . . . . . . . . . . . .102
Future Technology Devices, Showcase . . .www.ftdichip.com . . . . . . . . . . . . . . . .98, 99, 102
Futurlec, Showcase . . . . . . . . . . . . . . . . .www.futurlec.com . . . . . . . . . . . . . . . . . . . . . .102
Heros Technology, Showcase . . . . . . . . . .www.herostechnology.co.uk . . . . . . . . . . . . . .102
Intronix Test Instruments . . . . . . . . . . . . .www.pcTestInstruments.com . . . . . . . . . . . . . .41
Quasar Electronics . . . . . . . . . . . . . . . . . .www.quasarelectronics.com . . . . . . . . . . . . . .104
Redline . . . . . . . . . . . . . . . . . . . . . . . . . .www.redlineplc.com . . . . . . . . . . . . . . . . . . . . .78
Robot Electronics, Showcase . . . . . . . . . .www.robot-electronics.co.uk . . . . . . . . . . . . .103
Scantool . . . . . . . . . . . . . . . . . . . . . . . . .www.ElmScan5.com/elektor . . . . . . . . . . . . . . . .6
Showcase . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .102, 103
SourceBoost Technologies, Showcase . . .www.sourceboost.com . . . . . . . . . . . . . . . . . .103
Sytronic Technology Ltd, Showcase . . . . .www.m2mtelemetry.com . . . . . . . . . . . . . . . .103
Ultraleds, Showcase . . . . . . . . . . . . . . . .www.ultraleds.co.uk . . . . . . . . . . . . . . . . . . . .103
USB Instruments, Showcase . . . . . . . . . .www.usb-instruments.com . . . . . . . . . . . . . . .103
Velleman . . . . . . . . . . . . . . . . . . . . . . . . .www.velleman.eu . . . . . . . . . . .i-TRIXX backcover
Virtins Technology, Showcase . . . . . . . . .www.virtins.com . . . . . . . . . . . . . . . . . . . . . . .103
Jaycar Electronics . . . . . . . . . . . . . . . . . .www.jaycarelectronics.co.uk . . . . . . . . . . . . . . . .2
JB Systems, Showcase . . . . . . . . . . . . . .www.modetron.com . . . . . . . . . . . . . . . . . . . .103
JLB Electronics, Showcase . . . . . . . . . . .www.jlbelectronics.com . . . . . . . . . . . . . . . . .102
KMK Technologies Ltd, Showcase . . . . . .www.kmk.com.hk . . . . . . . . . . . . . . . . . . . . . .102
Labcenter . . . . . . . . . . . . . . . . . . . . . . . .www.labcenter.co.uk . . . . . . . . . . . . . . . . . . . .112
Lichfield Electronics . . . . . . . . . . . . . . . . .www.lichfieldelectronics.co.uk . . . . . . . . . . . . .37
12/2006 - elektor electronics
Advertising space for the issue of 22 January 2007
may be reserved not later than 22 December 2006
with Huson International Media – Cambridge House – Gogmore Lane –
Chertsey, Surrey KT16 9AP – England – Telephone 01932 564 999 –
Fax 01932 564998 – e-mail: gerryb@husonmedia.com to whom all
correspondence, copy instructions and artwork should be addressed.
111
Was this manual useful for you? yes no
Thank you for your participation!

* Your assessment is very important for improving the work of artificial intelligence, which forms the content of this project

Download PDF

advertising