Text preview for : English _ 2013-03-22 _ PDF 426 KB 5990-3218EN c20130617 [4].pdf part of Agilent English 2013-03-22 PDF 426 KB 5990-3218EN c20130617 [4] Agilent English _ 2013-03-22 _ PDF 426 KB 5990-3218EN c20130617 [4].pdf
Back to : English _ 2013-03-22 _ PD | Home
MEASUREMENT TIPS
Volume 5, Number 1
Improving
Digital Multimeter
Throughput
REDUCING TEST TIME
Snapshot: Testing Automotive ECUs
Whether your electronic test is in A major supplier for the automobile industry was making and testing
a manufacturing, design validation, electronic control units (ECUs). During final test of the ECUs, dozens of
or R&D environment, reducing your DMM voltage and resistance measurements needed to be made. The test
test time translates to lower cost time per ECU was running about 43 seconds. By rearranging the tests to
and shortened product develop- group all of the voltage measurements together and all of the resistance
ment schedules, both of which are measurements together, the manufacturer was able to reduce test time
clear benefits. The vast majority of per ECU by about 3 seconds. The manufacturer then evaluated the tem-
electronic tests involve using a digital perature of the environment that housed
multimeter (DMM) at one time or the test equipment and found it to be very
another. There are a variety of ways stable. As a result, test engineers decided
to reduce DMM measurement times to do a single zeroing measurement at the
to improve overall test throughput. start of the voltage measurements and a
Of course, test time improvements single zeroing measurement at the start
sometimes require compromises in of the resistance measurements in
other areas, but knowing the trad- order to turn off auto zero. This saved
eoffs involved in throughput improve- an additional 6 seconds for a total
ments and identifying what is impor- savings of 9 seconds. The test time was
tant in your specific test situation will reduced from 43 seconds to 34 seconds,
help you determine which tradeoffs a 21% reduction that increased throughput
make the most sense. by more than 25%!
Auto zero: Reduce the number of changes Auto range variations
Accuracy versus test time
Changing functions or measurement Auto range time can sometimes
Auto zero is a DMM feature that ranges also requires extra time contribute to longer test times, but
helps you improve accuracy. When in most DMMs. Try to group your not always. The time to auto range
you use the auto zero feature, the measurements to minimize function varies with the DMM design. DMMs
DMM makes an additional zeroing changes and range changes. For using flash A/D converters and paral-
measurement with each measure- example, if you make some voltage lel gain amplifiers can actually reduce
ment you make, thereby eliminating measurements and some resistance test times by using auto ranging,
the offsets of the amplifier and measurements, try to do all of the since the time to change ranges is
integration stages inside the DMM. voltage measurements together and zero. In these cases, the time to issue
However, turning this feature off cuts all the resistance measurements a range change command from a host
the measurement time in half. These together instead of changing back computer and parse the command in
offsets are initially calibrated out, and forth from one function to the the instrument will be slower. Manual
but the offsets can drift slightly with other. Also, try to group your low- ranging of integrating DMMs is still
a change in temperature. Therefore, voltage measurements together and the fastest way to take a measure-
if your measurements are taken your high-voltage measurements ment. Manual ranging also allows
in an environment with a stable together to minimize range changing. you to keep the DMM on a fixed
temperature, or if there are several Voltage ranges above 10 V use a range, which eliminates unwanted
measurements taken in a short period mechanical attenuator that takes time zero measurements and prevents the
of time (temperature changes occur to switch in and out. Grouping your mechanical attenuator from need-
over longer periods of time), the measurements by function and range lessly actuating. Note that the I/O
improvements in throughput by turn- will reduce your measurement times speed and range command parse time
ing auto zero off will far outweigh any considerably. for the Agilent 34410/11A multimeter
slight compromise in accuracy. For is significantly faster than the auto
example, with auto zero off in a stable range algorithm.
environment, the Agilent 34410A/11A MEASUREMENT TIP
multimeter typically adds only an Integration time versus noise
additional 0.0002% of range + 2 V to Minimizing range changes
the DC voltage accuracy specification. improves throughput not only Integration time is another parameter
Note that with auto zero off, any because it eliminates the time over which you have direct control,
range, function, or integration time it takes for the DMM to change but there is a clear tradeoff. DMMs
setting change can cause a single ranges, but also because it elim- integrate their measurements over
auto zero cycle to be performed on inates the need for an additional a set period of time: the integration
the first reading using the new set- auto zero measurement that time. The biggest benefit to choos-
ting. Consequently, turning auto zero would be performed because of ing a longer integration time is it
off and constantly changing settings the different offsets encountered eliminates unwanted noise from con-
defeats the time savings advantage. using different attenuation and tributing to your measurement, espe-
Check your DMM auto zero operation amplifier stages in a newly set cially AC mains line voltage noise.
to be sure of the circumstances lead- range. Minimizing range chang- However, longer integration times
ing to an advantage from this change. es can also lengthen the life of obviously increase your measurement
the DMM, since some range times. For example, if the integration
changes activate a mechanical time is set to an integral number of
relay that can wear with a large power line cycles (NPLCs) such as
number of activations. 1, 2, 10, or 100, the power line noise
contribution will be minimized due
to averaging over a longer period of
time and due to increasing the normal
mode rejection (NMR). With an NPLC
setting of 10 in a 60-Hz environment,
the integration time is 166 ms (200
ms for a 50-Hz line). The larger the
integral NPLC value, the larger the
NMR (for example, 60 Hz rejection),
but the longer the measurement time.
2
AC filter Input frequency Settling time ACV Settling time ACI
Slow 3 Hz to 300 kHz 2.5 sec/reading 1.66 sec/reading
Medium (default) 20 Hz to 300 kHz 0.625 sec/reading 0.25 sec/reading
Fast 200 Hz to 300 kHz 0.025 sec/reading 0.025 sec/reading
Table 1. Settling times for AC filter selections using the Agilent 34460A, 34461A and 34410A DMMs
Note that shorter integration times Optimizing trigger delay Summary
can yield lower resolution. And
changing the integration time will Whenever a signal is applied to the DMMs are used in virtually all
trigger another zero measurement input of a DMM, there is a certain electronic test systems; therefore,
if auto zero is turned on, so try to amount of time that must pass before making conscious choices about
group integration time measurements the signal completely settles in how to make DMM measurements
together if possible. Note that lower- order to make a valid measurement. can save large amounts of test
voltage measurements such as those This is especially true when the time, thereby increasing throughput.
performed on thermocouples usually applied signals are routed through Choosing appropriate settings for
need better noise rejection than high- a switching system, as is very com- auto zero, auto range, and integration
er-voltage measurements performed mon. For example, a voltage to be times as well as reducing function
on batteries or power supplies. measured will have to charge up any and range changes are just some of
capacitance in the switching path the opportunities you have to improve
AC settings and settling time when the signal is first connected to throughput. And reducing test times
the DMM input. Therefore, there is a will translate directly into lower
When making AC measurements, trigger delay setting that is dependent costs and faster time-to-market, both
be sure to select the appropriate AC on function, range, and integration important goals in today's fast-paced,
filter setting to match the signal you time and/or AC filter setting. The competitive marketplace.
are measuring. Select the bandwidth trigger delay is a time that must pass
setting to include the lowest fre- between the time a measurement has
quency you expect to encounter. For been requested (or triggered) and the
TIP SUMMARY CHECKLIST
example, the Agilent 34460A/61A and actual measurement is made. You can
34410A/11A multimeters have three optimize your measurement time by For maximum throughput:
AC filter settings, as shown in Table matching your trigger delay setting to
1, above. The settling time is longest correspond to the type of signal you