CR4® - The Engineer's Place for News and Discussion®

 Previous in Forum: Allen Bradley FTV Display Problem Next in Forum: Magnetic Level Transmitter
Associate

Join Date: Jul 2011
Posts: 42

How to Measure a PWM Current / Voltage Output?

04/06/2012 4:40 AM

what does a digital multimeter measures? peak/rms/or average value?

same question for Analog PMMC type meters.

What of I need to measure the average cuurent of a 33hz pwm waveform? what except CRO can be used?

Interested in this topic? By joining CR4 you can "subscribe" to

Check out these comments that don't yet have enough votes to be "official" good answers and, if you agree with them, rate them!
Guru

Join Date: Oct 2008
Location: Newcastle Oz
Posts: 954
#1

Re: how to measure a PWM current / voltage output?

04/06/2012 5:33 AM

Unless stated in the specifications Digital and Analogue multimeters will read Peak voltage, Only Certain Digital meters which state AC RMS will display the measured voltage as RMS.

As long as you are aware what your meter is showing you, then you can interpolate that measurement into the format you need.

__________________
There are two reasons for a man to do a thing, One that sounds good, and the real one...
Guru

Join Date: May 2006
Location: The 'Space Coast', USA
Posts: 11172
#2

Re: how to measure a PWM current / voltage output?

04/06/2012 7:32 AM

I don't think that RMS meters would work as I think they measure RMS for a sine wave, not an asymmetric square wave.

This LINK may be useful for understanding the problem.

In the case cited by the original poster you are looking for average current of a pulse wave and that will not be the same as an RMS current, you need pulse amplitude.

Guru

Join Date: Sep 2009
Posts: 967
#3

Re: How to Measure a PWM Current / Voltage Output?

04/06/2012 8:35 AM

I would prefer a good CRO for repetitive signals, but you can always use a DSO. S.M.

__________________
Life is complex. It has a real part and an imaginary part.
3
Guru

Join Date: Feb 2008
Posts: 1325
#4

Re: How to Measure a PWM Current / Voltage Output?

04/06/2012 10:46 PM

I don't know what you mean by CRO. Try to avoid special acronyms...

I have done many tests with RMS voltage measurements using different instruments from leading manufacturers and can tell you that as soon as you depart from a sinusoidal waveform at 50-60 Hz, the results diverge greatly (10-50%). I have used digital oscilloscopes, clamp ammeters, and multimeters of various precision (lab and field precision).

Even if a meter uses the correct equation to calculate the RMS, it is often at loss to figure out the correct cycle length of a complex waveform. The best results were obtained with an oscilloscope where the calculation window can be forced between the cursors (as long as I placed them correctly).

I have found that a simple RC filter with cutoff frequency of 1KHz (or less) helped many multimeters but you must compensate for the insertion losses and make sure it can take the voltage used.

My best results were with a Tektronix DPO3xxx oscilloscope working on a window manually defined using the cursors. Other oscilloscopes of that level or above should have similar features.

__________________
Experienced is earned, common sense is taught, both are rare essentials of life.
Guru

Join Date: Dec 2006
Location: Panama
Posts: 4289
#5

Re: How to Measure a PWM Current / Voltage Output?

04/06/2012 11:08 PM

It all depends on what the manufacturer claims the meter measures; not all "true RMS" meters are true. As noted by others, the only way to determine for sure what you are measuring is with an oscilloscope. I wasn't aware that Cathode Ray Oscilloscopes were still available on the market- I thought everyone had switched to DSO's with LED screens...

5
Power-User

Join Date: Sep 2011
Location: Bellingham, WA
Posts: 204
#6

Re: How to Measure a PWM Current / Voltage Output?

04/07/2012 11:02 AM

Digital meters vary by brand and model. Fluke true-RMS multimeters happen to use a thermal-based sensing element to measure true-RMS value for any wave-shape. This is one of the many reasons I prefer the Fluke brand for precision work.

Analog meters are generally calibrated to register the RMS value for a sinusoidal wave-shape. However, the natural response of a moving-coil meter mechanism is to respond to the average value of the wave-shape, not the RMS value, and so trying to measure a non-sinusoidal wave-shape with an analog meter will result in incorrect measurements. For more information on this subject, research "crest factor" for different wave-shapes.

If your task is to measure the duty cycle of a PWM wave-shape, you can use an oscilloscope, or you can use a multimeter that measures duty cycle directly. Most Fluke brand multimeters can do this directly, and display the duty cycle as a percentage. Just one more reason I like that brand . . .

__________________
They call me "lightning" when wielding a hammer, because I never strike twice in the same place
Guru

Join Date: Dec 2006
Location: Panama
Posts: 4289
#7

Re: How to Measure a PWM Current / Voltage Output?

04/07/2012 1:46 PM

I once was an avid fan of Fluke, but the past few years has seen way too many Fluke failures in my shop. I no longer respect their quality...Although I know of a number of older Flukes still going strong 20 to 25 years into their lives. Seems it is only the lastest versions (say, the last 5-7 years or so) that die prematurely...

Off Topic (Score 4)
Guru

Join Date: Dec 2010
Posts: 1263
#8

Re: How to Measure a PWM Current / Voltage Output?

04/09/2012 5:45 PM

Most digital meters measure the average value during a sampling window with a length which is a multiple of the local mains period [often they have a 50/60Hz switch preset inside].

For AC mains the reading is adjusted automatically to indicate the root mean square (rms) value of a sinusoidal waveform N.B. rms/average = 1.1107.

If the window is not a multiple of period, incomplete cycles are measured and reading fluctuates.

If I guess right, PMMC = Permanent Magnet Moving Coil. This kind of meter measures the average value and the good multimeters (AVO, VOM) and panel meters have heavy damping - a time constant of about 1 second. They will indicate average value reliably - even with a flicker, the average needle position is good. At 33 Hz, flicker would not be a problem.

You do not write whether your PWM is "chopped DC" or two polarity. If it is DC, you can use the moving coil meter directly, bearing in mind that the shunt full scale voltage may not match the meter full scale.

If AC, the usual full-scale is 2.5 V rms sine minimum - because the rectifiers are not linear for lower voltages. This is too high for most shunts although source resistors on parallel FETs might give enough volt drop.

It would be possible to use an amplifier to increase the voltage to the meter.

Power-User

Join Date: Feb 2007
Location: Gdansk, Poland
Posts: 100
#9

Re: How to Measure a PWM Current / Voltage Output?

11/28/2012 4:14 PM
• If current is one polarity (switched DC, not alternating), almost any digital multimeter at DC will do the job. Most digital multimeters at Voltage DC/Current DC measure average voltage/current.
• However, measurement period is fixed to 1...n periods of 50Hz /60Hz. As expained by Tonykuphaldt, measurement of 33Hz PWM will give fluctuations in readings. Therefore mulimeter with high n (long measuring period), or with averaging (over many readings) should be used - will give more stable readings.
• There are multimeters with integration time triggered by external signal, most suitable for this job. Triggered by Your 33Hz, divided by:(2,4,.......2n), they would allow for most stable readings.

Readings of Digital Multimeters on AC : some are measuring avg(rectified voltage), some are measuring peak voltage, some RMS based on heat generated (eg Fluke , mentioned by 67model), some RMS obtained by fast analog squareing/rooting circuits (Solartron 70xx series).

Interested in this topic? By joining CR4 you can "subscribe" to