Acorn Products Z80 User's manual

Acorn Products Z80 User's manual
Z80 SECOND PROCESSSOR SERVICE MANUAL
Part No. 0409,015
Issue 2
August 1984
Within this publication the term 'BBC' is used as an abbreviation for
'British Broadcasting Corporation'.
°Copyright Acorn Computers Limited 1984
Neither the whole or any part of the information contained in, or the
product described in, this manual may be adapted or reproduced in any
material form except with the prior written approval of Acorn
Computers Limited (Acorn Computers).
The product described in this manual and products for use with it, are
subject to continuous development and improvement. All information of
a technical nature and particulars of the product and its use (
including the information and particulars in this manual) are given by
Acorn Computers in good faith. However, it is acknowledged that there
may be errors or omissions in this manual. A list of details of any
amendments or revisions to this manual can be obtained upon request
from Acorn Computers Technical Enquiries. Acorn Computers welcome
comments and suggestions relating to the product and this manual.
All correspondence should be addressed to:Technical Enquiries
Acorn Computers Limited
Newmarket Road
Cambridge
CB5 8PD
All maintenance and service on the product must be carried out by
Acorn Computers' authorised dealers. Acorn Computers can accept no
liability whatsoever for any loss or damage caused by service or
maintenance by unauthorised personnel. This manual is intended only to
assist the reader in the use of this product, and therefore Acorn
Computers shall not be liable for any loss or damage whatsoever
arising from the use of any information or particulars in, or any
error or omission in, this manual, or any incorrect use of the
product.
This manual is for the sole use of Acorn Computers' authorised dealers
and must only be used by them in connection with the product described
within.
First published 1984
Published by Acorn Computers Limited
CONTENTS
Page
1
Introduction
1
2
Packaging and Installation
2
3
Specification
3
4
Disassembly and Assembly
4
5
5.1
5.2
5.3
5.4
5.5
5.6
5.7
5.8
5.9
5.10
Circuit Description
CPU
Clock
ROM Latch
Wait-States
Reset
Interrupt Handling
DRAM Control
DRAM R e f r e s h
Desynchronising Logic
The Tube
5
5
5
5
6
6
7
7
9
10
11
6
6.1
6.2
6.3
Fault Finding on the Z80 Second Processor
General
Fault Conditions
Circuit Checks
17
17
17
19
Appendix
Diagnostic Flowcharts
Z80 Second Processor - Block Diagram
Circuit Diagram
Z80 PCB Component Layout
Silk-Screen
Power Supply Unit Circuit Diagram
Z80 Second Processor - General Assembly
Parts Lists General Assembly
Z80 PCB Assembly
25
37
39
41
43
45
47
49
49
WARNING: THE Z80 SECOND PROCESSOR MUST BE EARTHED
Important: The wires in the mains lead for the. Z80 second processor
are coloured in accordance with the following code:
Green and yellow
Blue
Brown
Earth
Neutral
Live
As the colours of the wires may not correspond with the coloured
markings identifying the terminals in your plug, proceed as follows:
The wire which is coloured green and ye11ow must be connected to the
terminal in the plug which is marked by the letter E, or by the
safety earth symbol 4 or coloured green, or green and ye11ow.
The wire which is coloured blue must be connected to the terminal
which is marked with the letter N, or coloured black.
The wire which is coloured brown must be connected to the terminal
which is marked with the letter L, or coloured red.
If the socket outlet available is not suitable for the plug supplied,
the plug should be cut off and the appropriate plug fitted and wired
as previously noted. The moulded plug which was cut off must bedisposed of as it would be a potential shock hazard if it were to be
plugged in with the cut off end of the mains cord exposed. The moulded
plug must be used with the fuse and fuse carrier firmly in place. The
fuse
carrier
is
of
the
same
basic colour* as the coloured insert in the base of the plug. Different
manufacturers' plugs and fuse carriers are not interchangeable. In the
event of loss of the fuse carrier, the moulded plug MUST NOT be used.
Either replace the moulded plug with another conventional plug wired
as previously described, or obtain a replacement fuse carrier from an
authorised BBC Microcomputer dealer. In the event of the fuse blowing
it should be replaced, after clearing any faults, with a 3 amp fuse
that is ASTA approved to BS1362.
*Not necessarily the same shade of that colour.
1. Introduction
This manual is intended to provide the information required to
diagnose and repair faults on the Z80 second processor (a part of the
BBC Microcomputer system) which was designed by Acorn Computers Ltd.
of Cambridge, England.
The information contained in this manual is aimed at Acorn dealers and
service engineers who will be servicing the Z80 second processor on
behalf of Acorn Computers Ltd.
Z80 is a trademark of Zilog Inc.
CP/M® is a registered trademark of Digital Research Inc.
The Tube is a trademark of Acorn Computers Limited.
2. Packaging and Installation.
The Z80 second processor is supplied in a two-part moulded polystyrene
packing which is further packaged within a cardboard sleeve. Supplied
with the second processor is a DNFS ROM with fitting instruction
sheet, a set of reminder cards for the red function keys, 7 floppy
disks, an end-user licence and a guarantee card. For BBC
Microcomputers fitted with MOS ROMs below version 1.2, a voucher
redeemable against replacement of lower version ROMs, is also
supplied.
Note: Care should be taken when unpacking and repacking this unit to
ensure that all items are positioned correctly, especia11y the floppy
disks which should first be packed in plastic bags and laid flat.
The Z80 Second Processor User Guide and accompanying literature is
supplied packed separately.
A mains power switch is located at the rear of the second processor.
A 250mA. type T (slow blow) fuse is located at the rear of the second
processor. Before removing this fuse, the second processor must be
disconnected from the mains supply. Access to the fuse may be gained
by undoing the round cover with the slot in it using a screwdriver.
The mains supply must not be reconnected until the fuse is relocated
in its holder with the cover screwed home.
Do not use the second processor in conditions of extreme heat, cold,
humidity or dust, or in places subject to vibration. Do not block the
ventilation under or behind the second processor. Ensure that no
foreign objects are inserted through any openings in the second
processor.
3. Specification
3.1 The Z80 second processor
A second processor for the BBC Microcomputer model B, operating
through the Tube, providing the ability to run sophisticated software
under the CP/M 2.2 operating system.
The second processor is housed in a rigid injection moulded
thermoplastic case and contains the fo11owing:
A 6MHz Z80B CPU
64K of read/write Random Access Memory
4K Read Only Memory ("shadow ROM") providing a boot function on powerup and to handle Non-Maskable Interrupts (NMI) from the Host processor
via the Tube.
The Tube - a fast asynchronous communication path connecting the
second processor to the I/O processor (BBC Microcomputer).
A mains-operated integral power supply comprising a mains transformer
and power supply board.
3.2 Power Supply
Max. AC Input
MIN. AC Input
Power Rating
Supply Frequency
Max. Output Current
264V AC
216V AC
14 watts
47-63Hz
lA at +5V
4. Disassembly and assembly
To service the Z80 second processor, disconnect it from the BBC
Microcomputer and the mains supply and undo the three fixing screws;
two at the top of the back panel and one underneath the unit, nearest
the front and equidistant between the two rubber feet. (The assembly
diagram is given in the Appendix). The lid can now be removed
revealing the transformer and power supply
board, held in place by six screws, and the Z80 PCB. It is recommended
that the transformer and power supply board are not removed unless
absolutely necessary.
To remove the Z80 PCB from the case, pu11 off the two "fast on" tabs
which connect the power supply (brown +5v and black 0v leads) and
remove the four screws which hold the PCB in place.
5. Circuit Description
The circuit may be split into a number of sections by their specific
function. These are dealt with under separate headings. Reference
should be made, where necessary, to the block diagram and circuit
diagram in the appendix.
5.1 CPU
The microprocessor used in this unit is a Z80B, running at a clock
frequency of 6MHz from a crystal oscillator. A11 memory and I/O cycles
are performed at fu11 speed, with the exception of those to the "boot"
ROM, for which a Wait-State is inserted by external logic.
5.2 Clock
A 12MHz crystal controls the frequency of the oscillator formed by the
inverters 1C 24D,E. A "D" type latch, IC17A, is used to divide the
frequency to the required 6MHz. Transistor Q1 provides an active pullup for the clock signal, after inversion by IC 24B, to compensate for
the high dynamic input current of the Z80 on this signal. The NAND
gate IC19D and associated network , provide a shaped clock signal for
the "NMI Service Detect" logic. Since the output of the "D" latch is
inverted before being used as CPU clock, then the "D" output is
available for use as an inverted clock by the DRAM control and the
desync. logic.
5.3 ROM Latch
The Z80 second processor features a "shadow" ROM to boot the system
upon power-up and also to ensure proper handling of NMI interrupts
from the host processor via The Tube. The ROM is enabled at the proper
times by the latch IC15A.
1. After power-up, the reset signal from IC24F to the Z80 is used to
clock the latch IC15A and produce the ROM signal. On any memory read
cycle, while the Rom signal (TP2) is active, IC22B&C wi11 produce an
output-enable signal to the ROM (IC3 pin 20).
The initial instructions following RESET are executed from ROM and
initiate the copying of ROM into high RAM. This is followed by an
instruction-fetch cycle to memory over 8000H which is detected by the
AND gate IC20B and used to clear the ROM latch, remove the shadow ROM
from the memory-map, and allow normal running in RAM.
2. The NM1 signal to the Z80 processor is used by system software in
Disc handling, however, the Z80 interrupt vector to 66H is not
compatible with standard CPM, which has its default file-control block
in this area. The solution used is to bring the shadow ROM temporarily
into the memory-map when an instruction fetch from 66H
is detected. IC26 and IC27 provide a clock signal, NMISERV during such
a cycle, which presets the ROM latch IC15A, latching the ROM in until
an instruction-fetch from high RAM. In the ROM, 66H contains a jump to
the destination expected by standard CPM.
Because of the slow access time of the "boot" ROM, all memory cycles
to the ROM must be lengthened by the insertion of a "Wait-State" of
one clock cycle. When the ROM is selected, the OR gate IC22C provides
the OE signal to the ROM and this is used to enable the Wait-State
generator IC16A&B. Via the NAND gate IC21C, a low-going pulse of 1
clock cycle is fed to the WAIT input of the Z80 (see timing diagram
above). The Wait generator requires a further two clock cycles after
the end of the lengthened memory
cycle to clear itself. The Z80 samples the WAIT input on the
falling edge of 0 (t1). TP7 a11ows observation of the WAIT signal.
5.5. Reset
The Z80 second processor may be reset at any time, by the host
processor via the Tube.
The Z80 requires that a reset signal should not occur immediately
after an instruction fetch cycle, otherwise corruption of DRAM data
might result. To avoid this, the "D" latch IC15B synchronises the
reset signal from the Tube to the beginning of an instruction-fetch
cycle (M1). A monostable IC14 ensures the reset
signal to the CPU is a pulse of approx. 4µs duration, sufficient to
produce a reset without delaying the refresh to the DRAMS and so
losing data. The reset to the CPU also clears the ROM latch IC15A,
bringing the shadow ROM into the memory-map.
The Schmitt NAND gate IC19C provides a Power-Up reset to' the Z80 from
the delay network R1, C2 (time-constant 100ms). Diode D1 ensures that
the capacitor does not apply a reverse voltage to the NAND gate input
on Power-Down.
5.6 Interrupt Handling
The host processor can interrupt the Z80 with a maskable interrupt via
the Tube. The interrupt output from the Tube is taken directly to the
INT input of the Z80. After detection of an interrupt, the CPU M1 and
IORQ outputs go low to indicate a vector for the interrupt is expected
on the data bus D0 to D7. The buffer IC28 is enabled by M1 and IORQ
and its inputs are permanently tied to logic 0 or 1 to give a vector
address of 0FEH. The Z80 'Boot' ROM places the Z80 internal interrupt
system into Mode 2, with a High-Byte address of OFFH, giving an
address for the interrupt vector of 0FFFEH.
5.7 DRAM Control
5.7.1 Read/Write Cycles
a. RAS
Whenever a memory cycle occurs, the preset signal on the "D" latch
IC17B is removed by the MREQ signal from the Z80. On the next rising
edge of the system clock, the "D" latch output goes low, giving the "
CHOP" signal. During memory read or write cycles, the fa11ing edge of
CHOP produces the row-address (RAS) signal (TP8), via IC20A, to the
DRAMS, causing the row address information to be latched by the RAMS.
Prior to the RAS signal, the row-address buffer 1C5 was enabled by the
high level on RAS and consequently the low level on the inverted RAS
signal from 1C21A, thus allowing the low order address lines A0 to A7
to be passed to the DRAMS. Once RAS goes low, IC5 is disabled and IC4
enabled, to allow the column address through to the DRAMS. The
inverter IC25E ensures a slight delay in the enabling of the Column
buffer, to avoid data conflict with the Row buffer.
b. CAS
The column-address signal to the DRAMS is generated from the RAS
signal by the OR gates IC23C&D. If the shadow ROM output-enable signal
is active, then the CAS signal wi11 not be generated (logic 1 on pin
12 - IC23D). The AND gate 1C20C enables or disables the CAS signal
under certain other conditions.
CAS is enabled if:
i. the memory cycle is a
write cycle (WR
low, to IC20C pin
11).
or ii. the memory cycle is
a read cycle and
not
an
instruction-fetch
cycle (IC18B not
preset by Ml, RED
signal to IC20
pin 9).
or iii. the cycle is an
instruction-fetch
this signal being
synchronised to
the CPOP signal by
OR gate 1C23B (to
IC20 pin 10)
Note: In this case, the
CAS signal will
not be generated
if the ROM is
selected.
Fig. 2 RAM read or write cycles
5.7.2 Instruction-Fetch Cycles
The Z80 CPU handles an Instruction Fetch differently to other memory
read cycles, in that the MREQ signal is active for only 1.5 clock
cycles instead of 2. In order to allow sufficient access time for the
DRAMS in this abbreviated cycle, the Instruction-Fetch signal, Ml, is
used to generate the RAS and CAS signals a half-cycle earlier. The OR
gate IC23A allows the clock signal through to the "D" latch IC18A,
only when M1 is active. The output of the "D" latch, SUE is clocked
low, and generates the row-address latch signal RAS, a half clock
cycle before the CPOP signal would have done. When the CPOP signal
arrives after being generated by the MREQ (see section 5.7.1a), it
clears the SUE latch and holds RAS low itself until MREQ becomes
inactive.
Fig. 3 Instruction-Fetch cycles
5.8 DRAM Refresh
After each instruction-fetch, the Z80 CPU performs a Refresh cycle for
the DRAMS in the period while the instruction is being decoded. A
seven-bit refresh address is output onto the address-bus (A0 to A6, A7
=0) for approx. 2 clock cycles, and the MREQ signal goes low.
The "
RFSH" signal from the Z80 is not used, and no other
memory control signals go active. The Refresh address is incremented
by the CPU after each time.
Once the MREQ signal goes active, the "D" latch 1C18B produces CPOP
and hence RAS as normal. The CAS is not required for a Refresh cycle,
and is not enabled since none of the conditions listed in section 5.7.
1b are true (AND gate IC20C).
When the shadow ROM is being read, the CAS signal to the DRAMS is
disabled, but the row-address latch signal, RAS, sti11 occurs. This
has the effect of a refresh cycle to the DRAMS.
5.9 Desynchronising Logic
To prevent ambiguous events, i.e. a register status change during a
status read, this circuit produces a "WA1T" signal to the Z80
processor when the PCS and HCS signals occur simultaneously.
When this happens, a low signal from IC29B pin 6 appears at IC30A pin
2. Q on IC30A goes high and, via IC29A, maintains a logic 1 signal
upon pin 12 of IC30B, thus, by the end of one clock cycle a high is
sent from pin 9 of IC30B to disable PCS. Simultaneously, a WAIT signal
is generated for the second processor via IC19A&B.
As soon as HCS (TP6) is removed, the next rising clock edge removes
the WAIT signal from the Z80 and, as PCS is still low, a
low signal is sent
through IC30B to the
Tube. This is then
maintained by the
low signal upon pin
4 of IC30A until the
PCS is complete.
Thus, if PCS (TP5)
is already running,
it will continue
despite an HCS, but
if HCS began first
then PCS is
prevented from
acting.
*At no time is PCS
affected, as it
would not be
possible to 'stop'
the BBC processor.
Fig. 4 Timing Diagram - HCS/PCS
5.10 The Tube
The Tube (IC1) is an Acorn custom IC which provides parallel
asynchronous communication between two processor systems, the BBC
Microcomputer (Host) and the Z80 second processor (Parasite). To each
processor system, it resembles a conventional peripheral device
comprising 4 read-only and 4 write-only, 8-bit registers. The Z80
accesses these registers via its I/O structure.
Fig. 5 Tube concept
5.10.1 Tube Registers
Each register has its own status byte, with a separate I/O address,
containing Register-Fu11 and Data-Available flags. The status byte for
Register 1 contains additional control bits that may be set by the
Host computer to enable interrupts or to reset the Z80. These control
bits may be read, but not set, by the second processor.
Fig. 6 shows the Tube registers in more detail
Fig. 6 Schematic diagram of Tube registers
The fo11owing tables show the relative address and type of each
register in the Tube, firstly for the Host system, and secondly for
the parasite system (second processor).
Table 1 Host system registers
Address
000
001
010
011
100
101
110
111
Read
Status flags and Register 1 flags
Register 1 (24 byte FIFO read only)
Register 2 flags
Register 2 (1 byte read only)
Register 3 flags
Register 3 (2 byte FIFO read only)
Register 4 flags
Register 4 (1 byte read only)
Address
000
001
010
011
100
101
110
111
Write
Status Flags
Register 1 (1 byte write only)
Register 2 (1 byte write only)
Register 3 (2 byte FIFO write only)
Register 4 (1 byte write only)
Table 2 Parasite system registers
Address
000
001
010
011
100
101
110
111
Read
Status flags and Register 1 flags Al Fl P V M J I Q
Register 1 (1 byte read only)
Register 2 flags
Register 2 (1 byte read only)
Register 3 flags
Register 3 (2 byte F1FO read only)
Register 4 flags
Register 4 (1 byte read only)
Address
000
001
010
011
100
101
110
111
Write
Register 1 (24 byte FIFO write only)
Register 2 (1 byte write only)
Register 3 (2 byte FIFO write only)
Register 4 (1 byte write only)
As can be seen from Fig. 6 and Tables 1 and 2, each numbered register
(e.g. register 1) is actually two registers, one for reading and one
for writing. The register selected is determined by R/W on the Post
system and by NRDS/NWDS on the Parasite system (see Tube Pinout
Diagram).
Only registers 2 and 4 are simple latches; register 3 is a 2-byte
FIFO in each direction and register 1 is a 24-byte FIFO from the
Parasite (Z80) to the Post, but a simple latch from Post to the '
Parasite. The Tube produces maskable and non-maskable interrupts to
the Parasite (see sections 5.6 and 5.3) and a reset signal (section
5.5).
The Z80 IORQ and M1 signals are decoded to detect an I/O cycle by the
OR gate IC22A, which provides the signal which, via the De-sync
circuit, initiates the chip-select, PCS, to the Tube. The Tube thus
occupies all of the Z80 I/O map, the four data registers and four
associated status registers reflecting throughout the possible 256
addresses.
5.10.2 Tube Pinout
Fig. 7 Pinout diagram for Tube IC
DESCRIPT1ON OF PINS (ref. Fig.7):
Power Supply
GND
VCC1
VCC2
VCC3
0V supply rail
Parasite main +5V supply
Parasite secondary supply (+2-3v)
Host +5V supply
Data buses
HD0-7
PD0-7
Address signals
HA0-2
PAO-2
8-bit data bus to Host processor
8-bit data bus to Parasite
processor
3 register select lines from Post
3 register select lines from
Parasite
Post chip select
Parasite chip select
Host $2 - high level signifies
valid address bus
Post read/write line - determines
whether read or write register is
selected on address specified by
PAO-2, and direction of data flow
on PD0-7
Parasite read strobe (active-low)
Parasite write strobe (active-low)
Timing signals
PCS
PCS
P42
PR/W
PNRDS
PNWDS
Interrupt lines
PRST
PRST
PNM1
PIRQ
DMA lines
DRQ
DACK
Host reset (RST) -initialises Tube
to known state and generates PRST
Reset (RST) line to parasite
processor
Non-maskable interrupt to parasite
Interrupt to Host (not used by Z80
second processor)
Request for DMA transfer
DMA acknowledge from DMA controller
DMA facility is not used by the Z80 second processor
5.10.3 Tube Timing Diagram
Fig. 8 Tube Timing Diagram
N.B.The timing reference for the Post is 0 and R/W gives the
direction of transfer.
For the parasite, the PCS direction is given by PNRDS
or PNWDS, and timing by PCS.
1
2
3
4
5
6
7
8
9
10
11
R/W SET UP TO 02
TIMING STROBE PULSE WIDTH
ADDRESS SET UP TIME
ADDRESS & CHIP SELECT POLD TIMES
DATA OUT DELAY TIME
DATA OUT POLD TIME
DATA IN SET UP TIME
DATA 1N HOLD T1ME
R/W POLD T1ME
CYCLE TIME
CS SET UP T1ME
MIN
35ns
110ns
35ns
10ns
10ns
5Ons
20ns
10ns
250ns
20ns
MAX
70ns
6. Fault Finding on the Z80 Second Processor
6.1 General
a)
The Z80 second processor has three socketed IC's (IC 1 - 3) these may easily be replaced if necessary.
b)
Test points are provided on the Z80 PCB, as fo11ows:
TP1 CLOCK
6MHz clock signal 0 to pin 6 of Z80
processor
TP2 ROM
disable/activate signal to ROM, pins 18 &
20
TP3 M1
Z80 generated clock signal indicating an
instruction-fetch cycle. Also used in
interrupt handling.
TP4 MREQ
goes low to indicate memory addressing
TP5 PCS
indicates successful parasite chip select
to Tube, via de-sync logic circuit
TP6 HCS
indicates Post chip select to tube
TP7 WAIT
Occurs during reads from the ROM and as
result of simultaneous HCS/PCS event.
Enables refresh cycles
TP8 RAS
Row Address Signal, used in ALL memory
addressing, both for accessing and
refreshing
TP9 CAS
Column Address Signal, used only for
memory accessing, disabled during
refresh
6.2 Fault Conditions
A Z80 Second Processor failure can usually be related to one of
five fault conditions:
"BBC Microcomputer 32k..." message displayed.
Predominantly caused by either power failure, misconnected or
damaged plugs and/or interconnecting cable.
"Acorn TUBE Z80 64k ..." message displayed; no response to
Keyboard.
If this message is displayed, the ROM has been copied
completely to RAM , the ROM disabled and the Boot procedure
begun. Failure to respond to the keyboard means that the
system has crashed due to either hardware or software failure.
Possible causes may be:
Software
Incompatible or failed
DNFS or Boot ROM
Hardware
Tube register fault
Z80 fault
DRAM error
A further consideration is that a component or components may
have
become
temperature
sensitive
and
are
failing
intermittently. Check by first replacing IC1, 1C2 and IC3 in
turn, then check operation once warm; use a freezer spray to
locate temperature-sensitive components.
Flashing Cursor in the top left corner of an otherwise blank
screen (Total Failure)
This is the most usual result of plugging in a broken Z80
second processor as the majority of
faults manifest
themselves in this way.
Normal Operation until BREAK reset attempted; system fails.
Most probably caused by failure of reset circuit (IC15b,
IC14 and associated components).
BBC Microcomputer fails when Z80 Second Processor connected to
Tube socket
Probable causes are misconnected or damaged plugs/sockets
and/or damaged ribbon cable; Tube IC failure or IC29 failed on
Z80 PCB.
Diagnostic Flowcharts for the above conditions are given in the
Appendix. These should be read in conjunction with the following
Circuit Checks:
6.3 Circuit Checks
6.3.1 Clock
Using an oscilloscope, check that a 12MHz signal is being generated
at pin 13 of IC22D. If not, check the crystal X1, resistor values and
operation of inverters IC24D/E. Trace the signal to pin 9 of IC17A
where it should appear as a clearly defined 6MHz square wave (0). 4)
should appear from the driver Q1 to supply IC2 pin 6. Check that the
clock signals, 4) and appear at all the expected points shown on the
circuit diagram. If not, check for loading caused by failed IC's and
track short-circuits. Pin 11 of IC19D should also be generating a
delayed clock required for the NMISERV circuit. If no delayed clock
is found, check the values of C9 and R9.
6.3.2 RAS/CAS Generator Circuits
Both of these are best traced back from the RAM. RAS is always
present and should be seen at TP8 and also inverted at pin 3 of
IC21A. If only one appears, then check for loading, either on the
address buffers or on the DRAMS.
RAS is generated by both IC18A and IC17B (SUE and CPOP signals
respectively), independently of each other, but both are required to
be operating for full RAS ability. RAS may therefore be appearing due
to only one of the two Dtypes working, so check that pins 1 and 2 of
both IC20A and IC21A are operating. If not, check the operation of
the Dtypes according to inputs; RAS wi11 fail if the CPU is not
operating as it requires Ml, and MREQ, as we11 as the clock signal.
The operation of CAS is dependant upon the functioning of RAS and
also the correct decoding of a memory access. Check that memory RD, W
or Instruction Fetch (Ml) signals appear then check that this is
properly decoded from IC23B to IC23C via IC20C, and not disabled by
an incorrect signal from 1C23D.
6.3.3 Wait State Generator
IC2 pin 24 should predominantly be high; WAIT should only be active
under two conditions:
i)
ii)
During ROM read, TP2 goes low for approximately 0.25sec. This is
visible as a low on a logic probe applied to IC16A pin 2 after
the BREAK key has been pressed.
When PCS and PCS occur simultaneously; this WAIT pulse is
generated frequently during data transfer via the Tube. Note
that, in this second case, the WAIT signal is produced by the
desync. logic.
If WAIT is permanently low, or high, check TP7 after pressing BREAK;
the WAIT signal should go low and then high. If not, check that (0) is
clocking IC16A and B and IC30A and B (for HCS/PCS WAIT) and that the
desync. circuits are producing the correct WAIT outputs. See section
5.9.
6.3.4 ROM Signal/Break
On power-up, the RC network R1, C2, D1 provides a low to high
transition of approximately 0.1 second duration to pin 9 of IC19C. If
power-up reset fails and the low to high transition time is found to
be incorrect, check these component values and replace as necessary.
After power-up, pressing and releasing the BREAK key on the host
keyboard causes PRST to appear on pin 37 of Tube IC1. This is clocked
through IC15B by Ml. Thus, if the CPU is halted for any reason, M1
will not be present and a BREAK reset will not be possible, i.e. a
successful power-up reset is necessary to a11ow any further resets to
work.
The low signal should clock to monostable IC14 which should produce a
signal of approximately 10µs duration. If not, check the values of RC
network R7/C8 and replace if necessary.
The output of IC14 appears at pin 10 of IC19C; from here on the reset
function is common to both power-up and BREAK, as follows:
The reset signal from IC19C is inverted by IC24F and appears at pin
26 of IC2 (CPU reset) and also at pin 3 of IC 15A (ROM latch) so
that, if 1C15A is functioning correctly, a reset should cause a lowgoing pulse to appear at IC2 pin 26 (reset active low), followed by a
low on TP2 (ROM). This signal must appear at pin 18 of IC3 and
requires both MREQ and RD to be both active low to pass IC22C and
output enable IC3 on pin 20 (and disable CAS at IC21D, pin 13).
The active-low ROM signal at IC22C also appears at pin 2 of IC16A
which enables WAIT states at pin 24 of IC2 (see Wait State Generator,
above). Using an oscilloscope, check that all these events occur,
replacing any failed components.
After ROM on TP2 has remained low for approximately 0.25s, the CPU
executes an instruction fetch from high memory, M1 and MREQ both go
to active low and their inverted signals appear at pins 3 and 5
respectively of IC20B. This, combined with A15 high should produce a l
ow at pin 6 which, via IC24C will clear IC15A at pin 1 and remove the
ROM
signal.
Again,
check
all
conditions
with
an
oscilloscope and correct any failed logic.
6.3.5 Desynchronising Logic and PCS Disable HCS
After power-up, check that PCS is active at pin 18 of IC1. If not,
then either the Tube 1C1 or IC29 has failed on the second processor
side, there is a ribbon cable/connector fault, or the Host is faulty.
PCS
After pressing BREAK, check that a low signal appears simultaneously
at pins 1 and 2 of IC22A and that this appears at pin 3. Check that a
low then appears at pin 21 of IC1; if not, the
Desync. logic circuit is faulty. Check the clock signal at pin 11 of
IC30 and the inverted clock signal at pin 3. Whilst referring to the
Circuit Description, check that a11 signals are operating correctly in
the Desync. logic circuit.
With HCS checked to be functioning correctly, PCS REQ, via IC29A,
should always reach Pin 12 of IC30B when HCS is high. If not, check
operation of IC30A and that pin 5 of IC30A only produces a low signal
when HCS is active.
6.3.6 NMISERV
When an active signal appears at pin 17 of IC2 (NMI), the address
lines should be seen to address 066H (instruction fetch). This should
decode through IC26/27 to give the NMISERV signal. Failure of this
circuit wi11 prevent disk access. If this occurs, check 0D clock
circuit, that NMI from IC2 pin 17 appears at pin 5 of IC21B and that
NMISERV from pin 9 of IC27 appears at pin 4 of IC15A and IC21D pin 12.
Check for broken tracks and replace IC's 26 and 27 if necessary.
6.3.7 Interrupt OFEH
When operating the Boot ROM, the interrupt vector 0FEP from IC28 wi11
be read. Check the operation of IC28, that M1 and IORQ are appearing,
that the buffer inputs are correctly tied (high or low) and that the
buffer output of "FE" is appearing upon request; if not, check tracks
and power rails. Replace IC28 if necessary.
6.3.8 DRAMS (Dynamic RAM IC's)
The fo11owing should be performed for each DRAM in turn (IC's 6 -13):
Check the power supply pins, +5v to pin 8 and Ov to pin 16. Check
that RAS, CAS, and W are all appearing, then make sure that address
buffers become enabled (active low) at pins 1 and 19 of 1C's 4 and 5,
and providing active address lines to the DRAMS. Check that no
address lines are shorted together and that all data lines are
operating and not tied together.
6.3.9 Power Supply
Check the 250 mA. type T mains fuse, accessible at the rear of the
unit (see section 2).
Check for any loose, disconnected or broken leads.
After making sure that the second processor is disconnected from the
mains supply, check the mains switch at the rear of the unit.
Overload protection of the second processor is provided on the second
processor Z80 board itself. Fuse FS1 protects against overcurrent and
thyristor Till protects against overvoltage by blowing the fuse.
The overvoltage protection circuit functions as follows:
D3 is a 5.1V Zener diode; rail voltages greater than 5.1V appear
across resistor R11 whilst spikes are absorbed by capacitor Cll. If
the rail voltage across R11 exceeds approximately 1V, thyristor TH1,
which is capable of drawing 8 amps., conducts and blows the 1 amp.
fast-blow fuse FS1. Supply rail cut-off is therefore achieved if the
voltage reaches approximately 6.1V.
If fuse FS1 is blown, it could be due to either overvoltage or
overcurrent and there is likely to be either a short circuit
somewhere or the power supply board is faulty (it is supplying too
high a voltage).
Disconnect the two power supply leads, brown and black, from the
second processor PCB and connect a 10 ohm 2.5W resistor between them.
Measure the voltage across the brown lead (+5V) and the black lead (
ground) which should be in the range 4.95 to 5.25V with a maximum of
50mV noise (peak-to-peak, 0-50MPz bandwidth). If the voltage is out
of spec. then set it to 5V exactly using the trimmer which is
accessible through a hole in the power supply board. See Fig. 9.
If 5v cannot be obtained and/or the noise level is out of spec.,
replace the power supply unit.
Now remove the resistor connected across the power supply leads and
reconnect them to the second processor PCB, ensuring correct polarity.
Test the current drawn by the second processor PCB from the +5V
supply. The board should draw 600 - 800mA from the power supply.
Fig. 9 Position of +5V Trimmer on Power Supply PCB.
If the current is zero, the second processor PCB has gone open
circuit. Check fuse FS1 and connectors and tracks.
If the fuse is blown, the fault is a short circuit on the PCB.
If the current is higher than it should be, measure the voltage. If
the voltage is greater than 5.25V, adjust it to 5V using the trimmersee fig. 9. If the voltage is in spec., then one or more of the
components on the second processor PCB is faulty. Switch off the
power supply and feel which of the components is hot.
WARNING TAKE CARE WPEN CPECKING FOR 'POT' COMPONENTS` - TPEY MAY BE
HOT ENOUGH TO CAUSE INJURY.
6.3.10 Checking the Tube power supplies
The Tube (IC1) is powered both from the BBC Microcomputer and from
the second processor. If either of these supplies fails then the
second processor wi11 not work.
With the second processor switched off (ON/OFF switch down), switch
on the BBC Microcomputer (ON/OFF switch up). Check that there is a
potential of approx. 5V between pin 4 (+ve) and pins 1 and 5 (ground)
of IC1. If not, check the ribbon cable and connectors.
Now switch off the BBC Microcomputer (ON/OFF switch down), and switch
on the second processor (ON/OFF switch up). Check that there is a
potential of 5V between pin 2 (+ve) and pins 1 and 5 (ground) of IC1.
Also, check that there is a potential of between +2v and +3v between
pin 3 and both pins 1 and 5. If not, investigate components R2 and
Ll, replacing if necessary.
Diagnostic Flowcharts
Note: The letters in circles refer
to the relevant flowcharts
which follow.
Master Flowchart
27
Power Supply
3 5
Z80 Second Processor Functional Block Diagram
Z80 second Processor PCB Circuit Diag
Z80 PCB Silk Screen
43
Power Supply Circuit Diagram
Z80 Second Processor General Assembly
47
Z80 Second Processor Parts Lists
NOTE: Items indentified by * are norma11y available as spare
parts - please
availability.
ITEM PART No.
contact
your
supplier
for
details
DESCR1PTION
of
QTY REMARKS
Z80 Second Processor General Assembly (SEE PAGE 47)
2
3
4
5
7
8
9
10
11
12
13
14
15
23
24
26
201,110 *
201,109 *
201,742 *
201,108 *
831,000 *
870,302
870,040 *
800,027 *
800,017 *
880,025
800,037 *
815,900
815,901
815,207 *
805,003 *
882,946 *
882,665
890,000 *
CASE LOWER MOULD1NG
1
CASE UPPER MOULDING
1
CASE LABEL, LOWER
1
(REAR)
CASE LABEL UPPER
1
POWER SUPPLY ASSEMBLY
1
MAINS CABLE C/W PLUG
1
40 WAY RIBBON CABLE, GREY 260mm
1
40 WAY R1BBON CABLE STRAIN RELIEF 1
40 WAY RIBBON CABLE CABLE SOCKET 1
CABLE GROMMET
1
S.P.E. CONNECTOR
1
TOP PALF
FUSE POLDER
1
FUSE HOLDER SHROUD
1
FUSE 20mm x 5mm 250mA TYPE T
1
SLO BLO
MAINS SWITCH
1
SP1RE NUT No.6
2
SELFTAP PAN HD SCREW No6 x 13mm
3 BLACK
'STICK-ON' FOOT
4
Z80 PCB Assembly (SEE PAGE 41)
4
5
6
7
8
9
10
12
13
14
502,103
502,821
502,220
502,221
502,122
502,120
502,391
502,102
628,101
613,101
620,101
16
17
18
19
20
23
24
26
27
28
29
629,010
631,033
630,100
628,470
631,056
800,124
800,140
201,605
700,080
201,644
738,095
11
15
621,470
*
*
*
*
*
*
RESISTOR 10K 1/4W 5%
1
RESISTOR 820R 1/4W 5%
2
RESISTOR 22R 1/4W 5%
1
RESISTOR 220R 1/4W 5%
2
RESISTOR 1K2 1/4W 5%
1
RESISTOR 12R 1W 10%
1
RESISTOR 390R 1/4W 5%
1
RES1STOR 1K 1/4W 5%
1
CAPACITOR 100nF CERAMIC
1
CAPACITOR 10uF 35V TANT
1
CAPACITOR 100uF 6V3 ELEC
1
CAPACITOR 47uF 10V ELEC
1
CAPACITOR 10nF PLATE CERAMIC
1
CAPACITOR 33pF PLATE CERAMIC
2
CAPACITOR 1000pF PLATE CERAMIC
1
CAPACITOR, DECOUPLER
33
CAPACITOR 56pF PLATE CERAMIC
1
IC SOCKET 24 WAY DIL
1
IC SOCKET 40 WAY D1L
2
INTEGRATED CIRCUIT TUBE
1
INTEGRATED CIRCUIT Z80B
1
INTEGRATED CIRCUIT, BOOT ROM
1
INTEGRATED CIRCUIT 81LS95/74LS795 3
R7
R3,4
R8
R6,9
R5
R2
R11
R1
C11
C10
C2
C3
C4
C5,7
C8
A NOM.47nF
C9
FOR IC3
FOR 1C1,2
IC1
, 1C2 CPU
IC3
IC4,5,28
Z80 PCB Assembly (SEE
31
32
33
34
35
36
37
38
39
40
41
42
43
45
46
47
48
50
51
53
54
55
56
50
704,164 *
742,123 *
741,074/
748,074 *
742,074 *
742,132 *
742,011 *
742,000 *
742,032 *
741,004 *
742,004 *
742,260 *
742,133 *
800,037 *
791,000
783,906 *
880,049
820,120 *
815,007 *
860,002 *
815,910
800,200
795,006
794,148
PAGE 41) - cont'd
INTEGRATED CIRCUIT 8264
INTEGRATED C1RCUIT 74LS123
1NTEGRATED CIRCUIT 74S74/74F74
8
1
2
IC6-13
IC14
IC15
INTEGRATED CIRCUIT 74LS74
INTEGRATED CIRCUIT 74LS132
INTEGRATED CIRCUIT 74LS11
INTEGRATED CIRCUIT 74LS00
INTEGRATED CIRCUIT 74LS32
INTEGRATED CIRCUIT 74S04
INTEGRATED CIRCUIT 74LS04
INTEGRATED CIRCUIT 74LS260
INTEGRATED CIRCUIT 74LS133
S.H.E. CONNECTOR
THYRISTOR C122F
TRANSISTOR 2N3906
INSULATOR
CRYSTAL 12MPz
FUSE 20mm x 5mm 1 AMP
INDUCTOR 2u2 10%
FUSE CLIP
FASTON TAB
DIODE ZENER BZY88 - C5V1
DIODE 1N4148
3
1
1
1
3
1
1
1
1
1
1
1
1
1
1
1
2
3
1
2
IC16-18,30
IC19
IC20
IC21
IC22,23,29
IC24
IC25
IC26
IC27
PL1 (PALF)
TH1
Q1
FOR 1TEM 45
X1
FS1
Ll
D3
D1
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

advertisement