Make Faster Frequency Measurements with Your

Keysight Technologies
10 Hints for Getting the Most
from Your Frequency Counter
Application Note
Introduction
Counters can be a plug and play instrument and seem fairly simple from the outside. You connect a
signal to the input, and a digital readout tells you the frequency or some other parameter. However, to
achieve the best results, whether that means speed or quality, attention to how you set up the counter
measurement is important.
Choosing the Best Counter
Selecting which counter will best meet your needs is the irst step. There are several related products
that perform a variety of tasks at various frequencies:
– Universal counters
Both frequency and time interval measurements, as well as a number of related parameters.
– RF frequency counters
Precise frequency measurements, up to 3 GHz and beyond.
– Microwave frequency counters
Precise frequency measurements, up to 40 GHz and beyond.
– Time interval analyzers
Optimized for precision time interval measurements.
– Modulation domain analyzers
Designed to show modulation quantities, such as frequency versus time, phase versus time, and time
interval versus time.
Hint
1
Recognize the difference between resolution and accuracy
Assuming a large number of digits
equates to a very accurate measurement may not be correct. It is a
common mistake to equate resolution and accuracy. They are related, but different concepts.
The resolution of a counter is the
smallest change it can detect in
closely spaced frequencies. All other
things being equal (such as measurement time and product cost),
more digits are better—but the
digits you see on the display need
to be supported by accuracy. Digits
can be deceptive when other errors
push the counter’s resolving
ability away from the actual frequency. In other words, it’s possible for a
counter to give you a very accurate
reading of an incorrect frequency.
Random and systematic errors both
determine a counter’s accuracy.
Random errors are the source of
resolution uncertainties and include:
– Quantization error
When a counter makes a measurement, a ±1 count ambiguity
can exist in the least signiicant
digit. This can occur because of
the non-coherence between the
internal clock frequency and the
input signal.
– Trigger error
Noise spikes can be triggered
by noise on the input signal or
noise from the input channels of
the counter.
– Timebase error
Any error resulting from the difference between the actual time
base oscillator frequency and
its nominal frequency is directly
translated into measurement
error.
Systematic errors are biases in the
measurement system that push
its readings away from the actual
frequency of the signal. This group
includes effects on the time base
crystal such as aging, temperature
and line voltage variations.
Compare the two counters in Figure
1. Counter A has good resolution
but a serious systematic error, so
its displayed result in most cases will
be less accurate than those of Counter
B, which has poorer resolution but a
smaller systematic bias error.
Random
Counter A
Systematic
Random
Counter B
Systematic
Actual f = 100 Hz
99.6
99.8
100
100.2
100.4
Figure 1. Simplified view of resolution vs.
accuracy. Systematic errors related to the
timebase “push” the displayed frequency away
from the actual frequency. The random errors
create a range of frequencies inside of which the
counter can’t distinguish different signals.
3
10
Understand counter measurement methods
This method is simple and inexpensive, but it means that the direct
counter’s resolution is fixed in Hertz.
For example, with a 1 second gate
time, the lowest frequency the counter
can detect is 1 Hz (since 1 cycle of the
signal in 1 second is 1 Hz, by definition). Thus, if you are measuring a 10
Hz signal, the best resolution you
can expect for a 1 second gate time is
1 Hz, or 2 digits in the display. For a 1
kHz signal and a 1 second gate, you
get 4 digits. For a 100 kHz signal, 6 digits, and so on. Figure 2 illustrates this
relationship. Also note that a direct
counter’s gate times are selectable
only as multiples and sub-multiples of
1 second, which could limit your measurement flexibility.
Reciprocal counters, in contrast,
measure the input signal’s period,
then reciprocate it to get frequency.
Given the measurement architecture
involved, the resulting resolution is
fixed in the number of digits
displayed (not Hertz) for a given gate
time.
6
4
2
10
In other words, a reciprocal counter
will always display the same number
of digits of resolution regardless of
the input frequency. Note that you’ll
see the resolution of a reciprocal
counter specified in terms of the number of digits for a particular gate time,
such as “10 digits per second.”
You can determine whether a counter
is direct or reciprocal by looking at
the frequency resolution specification.
If it specifies resolution in Hertz, it’s a
direct counter. If it specifies resolution
in digits-per-second, it’s a reciprocal
counter.
The counter industry has standardized
on measuring relative
to a 1 second gate time. Figure 3
compares the resolution of direct and
reciprocal counters. In the lower frequency spectrum, reciprocal counters
have a substantial advantage over
direct counters. As an example, at 1
kHz, a direct counter gives a resolution of 1 Hz (4 digits). A 10 digit/
second reciprocal counter gives a
resolution of 1 µHz (10 digits).
If precision resolution is not a
priority, the reciprocal counter still
offers a significant speed advantage:
the reciprocal counter will give 1 mHz
resolution in 1 ms, while a direct
counter needs a full second to give
you just 1 Hz resolution (Figure 4).
Reciprocal counters also offer continuously adjustable gate times (not
just decade steps), so you can get the
resolution you need in the minimum
amount of time.
4
1K
100 K 10 M 1 G
Frequency (Hz)
100 G
Figure 2. The number of digits displayed by a
direct counter versus frequency (for a 1 second
gate time).
Figure 3. Comparing resolution for direct and
reciprocal counters (for a 1 second gate time).
12
Reciprocal
10
8
6
re
ct
Direct counters simply count cycles
of the signal for a known period – the
gate time. The resulting count is sent
directly to the counter’s readout
for display.
8
Di
Frequency counters fall into two
basic types: direct counting and
reciprocal counting. Understanding
the effects of the two different
approaches will help you choose the
best counter for your needs and use it
correctly.
Number of digits
2
12
Number of digits
Hint
4
2
10
1K
100 K 10 M 1 G
Frequency (Hz)
100 G
Figure 4. Here are the gate times needed to
yield various resolutions with a 10 digits/
second reciprocal counter.
The choice comes down to cost versus performance. If your resolution
requirements are flexible and you
aren’t too concerned with speed, a
direct counter can be an economical choice. A reciprocal counter is
required for the fastest, highest
resolution measurements.
Hint
3
Choose the appropriate timebase
The measurement accuracy of
a frequency counter is strongly
dependent on the stability of its
timebase. The timebase establishes
the reference against which the
input signal is measured. The
better the timebase, the better
your measurements can be. The
frequency at which quartz crystals
vibrate is heavily influenced by
ambient temperature, and timebase
technologies fall into three
categories based on the way they
address this thermal behavior:
– Room Temperature Crystal
Oscillator (RTXO)
This type of timebase does not
utilize any temperature compensation or control. These
types of oscillators have been
manufactured for minimum
frequency change over a range
of temperature–typically
between 0°C and 50°C. This
is accomplished through the
proper choice of the crystal
cut during the manufacturing
process. A high quality RTXO
would vary by about 2.5 parts
per million (ppm) over that
temperature range. This works
out to ±2.5 Hz on a 1 MHz
signal, so it can be a significant
factor in your measurements. It
carries the advantage of being
inexpensive, but results in large
frequency errors.
– Temperature Compensated
Crystal Oscillator (TCXO)
One method of compensating
for frequency changes due
to temperature variation is
through externally added
components that have
complementary thermal
responses to obtain a more stable
5
frequency. This approach can
stabilize the thermal behavior
enough to reduce timebase
errors by an order of magnitude
relative to RTXO (approximately 1
ppm (±1 Hz on a 1 MHz signal)).
– Oven Controlled Crystal
Oscillator (OCXO)
In this technique, the crystal
oscillator is housed in an oven
which holds its temperature at
a specific point in the thermal
response curve. The result is
much better timebase stability,
with typical errors as small as
0.0025 ppm (±0.0025 Hz on a
1 MHz signal). Additionally,
oven-controlled timebases also
help with the effects of crystal
aging, which means you don’t
have to take your counter out
of service for calibration as
often.
Hint
4
Schedule calibration to match performance demands
The frequency at which you calibrate
your counter depends on several
factors:
– The type of timebase
– The conditions the counter
is subjected to during
measurement
– How much accuracy you need
from the measurement
Calibration can be a complex issue
and is directly related to counter
accuracy in general. The quality of
the measurement you see on the
display depends on four factors:
1. Time-invariant counter
performance factors (such
as the temperature stability
of the timebase as discussed
in Hint #3)
2. Time-variant counter performance factors (such as the
aging rate of the timebase)
Item #2 is where calibration plays a
role. Although counters are electronic instruments measuring
electrical signals, the quartz crystal
that is the heart of every counter’s
time-base is a mechanical device.
Since it is a mechanical device, the
crystal is susceptible to physical
disturbances that can change the
frequency at which it vibrates which
ultimately affects the counter’s
accuracy. The cumulative effect
of these various disturbances
is known as crystal aging, and
it is this aging that you are
compensating for when you
calibrate the counter.
Aging is a factor that is fairly easy
to predict and easy to compensate
for through calibration. You can
determine if calibration is required
by looking at the aging rate
specification in your counter’s data
sheet.
3. The input signal clarity and
level of noise
4. Counter set-up and configuration
6
For example: If the aging rate is
4 x 10-8 per day and it has been 300
days since calibration, aging will
add a timebase error of 1.2 x 10-5 into
the overall accuracy calculation.
If this uncertainty (±12 Hz on a 1
MHz signal), in addition to the other
inherent errors reviewed earlier is
acceptable for your measurements,
calibration is not required.Otherwise,
calibrate.
Hint
5
Make the most accurate measurements
– Select the best arming mode
If you want to make quick measurements, using your frequency
counter s automatic arming mode
is a simple way. However, of the
four typical arming modes (automatic, external, time, and digits),
automatic mode is the least
accurate. You can improve
resolution and systematic uncertainty (both elements of
measurement error, as discussed
in Hint #2) by increasing gate
time with either the external,
time, or digits arming modes.
– Use the best timebase available
and calibrate frequently
The quality of the timebase and
how often you calibrate will affect
your measurement accuracy. For
most applications, you can make
a tradeoff between accuracy, timebase quality, and calibration period.
If you purchase a higher-quality
timebase, you can lengthen the
time between calibrations. If you
calibrate more frequently, you
may be able to meet your accuracy
requirements with a less-costly
timebase. The timebase does not
need to be housed within the
frequency counter. You can use
a precision source or a house
standard external to the counter
to improve measurement accuracy.
– Keep your counter’s timebase
warm
As discussed in Hint #3, most
precision frequency counters
rely on a temperature compensated frequency oscillator
(TCXO) or an oven controlled
frequency oscillator (OCXO).
Keeping the frequency oscillator
continuously powered up will
avoid retrace and a shift in the
output frequency. Removing the
power to an oscillator, even for
a short length of time, means
that the oscillator will go through
its power on cycle of fluctuation
(retrace) before coming to rest at
a stable frequency. To ensure the
most stable operation of the crystal:
– Keep your counter in a spot
where you don’t have to unplug
it, so it can alternate between
on and standby mode.
– When you calibrate the timebase,
bring the calibration equipment to the counter, rather
than the other way around,
so you don’t have to unplug
the instrument.
– Keeping your frequency counter
out of drafts and protecting it
from changes in temperatures
will also improve its stability.
When you remove power from
the counter, however briefly,
the aging rate must start over
from the daily aging rate.
7
– Monitor trigger level timing error
When you make timing measurements (time interval, pulse width,
rise time, fall time, phase, and
duty cycle), you need to consider
the effects of the trigger level
timing error. There are several
factors to consider: resolution
and accuracy of the trigger level
circuit, fidelity of the input
amplifier, slew rate of the input
signal at the trigger point, and
width of the input hysteresis band.
To reduce these effects, trigger at
the offset value of the sine wave
or square wave signal. Doing so
will give you the highest slew rate,
and it also will minimize errors
of the hysteresis band. If you
measure from offset-to-offset
(such as a complete period,
0 degrees phase between two
signals) then the effects of the
hysteresis window may actually
cancel out. Note that most counters are optimized for a 0 V
trigger level setting.
– When possible, lock all
timebases to a single clock
The skew and/or jitter that occurs
between two independent timebases will add to error. Using
independent timebases is like
watching a movie with the video
and the audio tracks on different
systems. At the beginning of the
movie, the audio and video may
be synchronized, but as time
passes, small differences between
the two become more noticeable.
In many applications using
modern test and measurement
equipment, this skew is negligible.
Hint
6
Make the fastest measurements
You can configure a modern
frequency counter to make hundreds
of readings per second, which can
be useful for characterizing a signal
that changes over time. Keep in
mind that frequency counters are
optimized for measuring a stable or
slowly changing signal. Also
remember, for making accurate
readings, it is better to make a
single good reading than trying
to average lots of readings.
– Set the counter to a known state
After sending a reset command,
it is a good practice not to send
any additional commands until
the instrument has come back to
a ready state. Adding a wait or
delay of 1 second to a program is
enough for most instruments to
return to a ready state. If the
instrument receives a command
while it is resetting, the command
may be lost.
– Set the output format to match the
data type used in the instrument
This will prevent a delay as the
instrument converts the data to
a different format during post
processing.
– Disable all post processing
and printing operations
When you disable these functions, the processor dedicates
its resources to making the
readings and sending them
to the computer, rather than
responding to extra interrupts,
such as updating the display.
– Configure the expected frequency
The Keysight Technologies, Inc.
53100 series of counters has
the ability to optimize their
configuration based on the
frequency you are measuring.
The actual signal being measured
must be within 10% of the value
you provide in the command.
8
– Set the trigger level
The input signal will create a
trigger condition as it passes
through the level set in the
command. Set the trigger level
so that it intersects the signal
at its maximum slew rate, this
will minimize the amount of
time it takes to satisfy the trigger
condition. A sine or a square
wave has the maximum slew
rate at the zero crossing
(assuming a 0 V offset).
– Set triggering to make
immediate readings
When instruments use dual-level
triggering, both triggering conditions must be met before a
reading can be made. Setting
the trigger arm condition to
immediate will satisfy the first
level of triggering.
Hint
7
Adjust sensitivity to avoid noise triggering
Modern counters are broadband
instruments with sensitive input
circuits. However, to a counter, all
signals look the same. Sine waves,
square waves, harmonics, random
noise all look like a series of zero
crossings. A counter measures the
signal’s frequency by triggering on
these zero crossings. If your signal
is clean, the process is relatively
straightforward. Noisy signals, however, result in the counter triggering
on spurious zero crossings. When
this happens, you will not get the
measurements you expect.
Fortunately, there are approaches
around this issue.
– Counters require the signal
to pass through both lower
and upper hysteresis thresholds before they register a zero
crossing. The gap between these
two levels is referred to as trigger
sensitivity, the hysteresis band,
or the trigger band.
that are causing problems with
the measurement. The trigger
band is fairly narrow, so both
the unwanted noise (at points
1 and 3) and the real signal
(at points 2 and 4) cause the
counter to trigger. What is really
just two cycles of the signal get
counted as four.
By adjusting the trigger band to make
the counter less sensitive, you can
avoid these spurious triggers. In
Figure 6, the trigger band is wide
enough (the sensitivity is low
enough) that the spurs don’t get
counted as zero crossings. The
counter registers two valid zero
crossings and goes on to compute
the appropriate frequency. If you
think your signal might have some
noise problems, try switching your
counter into low sensitivity mode.
If the displayed frequency changes,
chances are you were triggering
on noise.
– High quality counters let you
adjust this band to minimize
unwanted triggering. Figure 5
shows a signal with some noise
9
2
1
4
3
0 V Trigger
band
Figure 5. The two small peaks (spurious signals
in this case) generate unwanted triggers at point
1 and point 3 because the trigger band is set too
narrow.
1
2
0 V Trigger
band
Figure 6. Lowering the trigger sensitivity by
expanding the trigger band produces the desired
count.
Hint
8
Reduce jumpy displays
A jumpy display, where the last several digits fluctuate rapidly, can be a
challenge if you’re trying to adjust a
circuit in real time or perform some
other task based on the counter’s
display. Depending on your counter’s capabilities, you have several
options:
– Reduce the number of displayed
digits
Most counters have a “Fewer
Digits” function. While this can
quiet the display, it might hide
information you need to make
decisions about circuit behavior.
Note this is strictly a display
function that does not have any
effect on the actual measurement.
– Use limit testing
If you only need to know whether
a signal is within a certain band
of frequencies, use limit testing
with a visual indicator if your
counter has this capability.
10
– Use signal averaging
Averaging (also labeled as “Mean”
on many counters) is a good
option to consider any time your
signal is jumping. Unlike simply
reducing the number of displayed
digits, averaging actually improves
the quality of your measurements.
By reducing the effects of random
variations in the signal, it reduces
the number of display changes.
Hint
9
Improve low frequency measurements
Hint #7 discussed the problem of
triggering on noise in your signal.
This problem can be even more
acute with low frequency signals
(roughly 100 Hz and below), since
the chance of spurious triggering on
irrelevant high-frequency components is increased. In addition, the
signal’s slew rate affects trigger accuracy – the lower the slew rate, the
more chance there is for error.
Some steps you can take to help
improve the quality of counter measurements on low-frequency signals:
– Utilize the low-pass filter
If the option is available, this
can reduce the chance of
triggering on harmonics and
high-frequency noise.
– Use manual triggering
When a counter is set to use auto
triggering, it estimates the peakto-peak level of the signal and
computes the midpoint to establish a trigger level. This approach
generally leads to good results
but can cause trouble on lowfrequency signals. The problem
is that the auto trigger algorithm
can take less time than the signal
takes to transition between its
minimum and maximum values.
As a result, the auto trigger can
wind up following the signal level
up and down, rather than setting
a single trigger level based on
a consistent estimate of the
minimum and maximum values.
The solution is to turn off auto
trigger and set the trigger level
manually.
– Use DC coupling
Many counters offer a choice
between DC and AC coupling
on their primary input channel.
AC coupling removes any DC
offset from the signal, whereas
DC coupling admits the entire
signal, offset and all. The issue
with AC coupling is that it also
attenuates lower frequencies.
Some counter’s performance
is not specified below a certain
frequency with AC coupling
due to this fact.
– Decrease the counter’s sensitivity
A low frequency signal may have
a low slew rate – meaning the
signal is slow to change states.
The slower the slew rate, the
harder it is to create a repeatable trigger. Decreasing the
counters sensitivity will help.
In order for a counter to successfully trigger, the signal will need
to pass through a lower and an
upper threshold. The trigger
band, the delta between the
upper and lower threshold is
determined by the counter’s
sensitivity. Decreasing the counter
sensitivity will increase the
difference between the upper
and lower threshold, widening
the trigger band.
11
– Monitor the status register
A low frequency measurement
can take time to complete. If
you are controlling the counter
from a computer, you may want
to check the status register before
requesting a reading. The counter
will continue to make a measurement until it receives a second
valid trigger condition, indicating
the end of the measurement. If
the input signal becomes disconnected, the counter will wait
indefinitely for the measurement
to complete. If you request a
measurement the computer will
be stuck waiting until the counter
measurement finishes before
responding to the query. To avoid
this, start the measurement and
then check the status register
to be certain a measurement
has completed before requesting
the reading.
Hint
10
Utilize limit testing capability
It is not uncommon for a counter to
produce a reading with 10-12 digits
every second. Limit testing can
enable you to interpret the readings
easier. You can configure and implement limit tests several
different ways:
– A visual indication can be
lit on the display to indicate
an out-of-limit reading.
– You can set the counter to
stop taking readings when
a limit is reached.
– You can instruct the counter
to send an SRQ over the
GPIB interface to indicate
a reading is out of limits.
– A hardware line is provided
that indicates an out-of-limit
reading has occurred.
– You can set the counter to
omit out-of-limit readings
from statistical measurements.
You can combine limit testing with
your counter's statistics, scale and
offset features. Scale and offset are
often used to convert a frequency
measurement to a physical measurement (for example, speed or rpm).
12
Lastly, you can configure the counter
to continue or to stop taking readings after a limit has been exceeded. If
your counter seems to stop
triggering, it may be because it is
configured to stop after an out-of-limit
reading. Also, when you configure
your counter to output an external
signal, it will cycle power and come
up in a default state, so make sure
you save and recall the frequency
counter setup.
13 | Keysight | 10 Hints for Getting the Most from Your Frequency Counter – Application Note
myKeysight
www.keysight.com/find/mykeysight
A personalized view into the information most relevant to you.
Three-Year Warranty
www.keysight.com/find/ThreeYearWarranty
Keysight’s commitment to superior product quality and lower total cost
of ownership. The only test and measurement company with three-year
warranty standard on all instruments, worldwide.
Keysight Assurance Plans
www.keysight.com/find/AssurancePlans
Up to five years of protection and no budgetary surprises to ensure your
instruments are operating to specification so you can rely on accurate
measurements.
www.keysight.com/go/quality
Keysight Electronic Measurement Group
DEKRA Certified ISO 9001:2008
Quality Management System
Keysight Channel Partners
www.keysight.com/find/channelpartners
Get the best of both worlds: Keysight’s measurement expertise and product
breadth, combined with channel partner convenience.
For more information on Keysight
Technologies’ products, applications or
services, please contact your local Keysight
office. The complete list is available at:
www.keysight.com/find/contactus
Americas
Canada
Brazil
Mexico
United States
(877) 894 4414
55 11 3351 7010
001 800 254 2440
(800) 829 4444
Asia Paciic
Australia
China
Hong Kong
India
Japan
Korea
Malaysia
Singapore
Taiwan
Other AP Countries
1 800 629 485
800 810 0189
800 938 693
1 800 112 929
0120 (421) 345
080 769 0800
1 800 888 848
1 800 375 8100
0800 047 866
(65) 6375 8100
Europe & Middle East
Austria
Belgium
Finland
France
Germany
Ireland
Israel
Italy
Luxembourg
Netherlands
Russia
Spain
Sweden
Switzerland
United Kingdom
0800 001122
0800 58580
0800 523252
0805 980333
0800 6270999
1800 832700
1 809 343051
800 599100
+32 800 58580
0800 0233200
8800 5009286
800 000154
0200 882255
0800 805353
Opt. 1 (DE)
Opt. 2 (FR)
Opt. 3 (IT)
0800 0260637
For other unlisted countries:
www.keysight.com/find/contactus
(BP-09-23-14)
This information is subject to change without notice.
© Keysight Technologies, 2008 - 2014
Published in USA, July 31, 2014
5989-8431EN
www.keysight.com
Download PDF

advertising