Announcement

Collapse
No announcement yet.

vfd and multimeter

Collapse
X
 
  • Filter
  • Time
  • Show
Clear All
new posts

  • vfd and multimeter

    why does a multimeter not display anything reasonable when trying to measure the output voltage of a vfd with a motor running? (i would have thought the motor smoothes out what the vfd generates.)

  • #2
    It is pseudo sine wave signal made up of pulse width modulated content, so nothing reasonable for the meter to indicate.
    Maybe when a 3PH choke is used, it may be more reasonable?
    Max.

    Comment


    • #3
      Hi Guys,

      In the absence of a oscilloscope a digital meter won't cut it ! You need a good old analogue meter.
      Best Regards:
      Baron J

      Comment


      • #4
        Consider that a digital meter is sampling periodically (every second, half second maybe). Even if this was in sync with what it's fed, the output would be ...dubious. The mechanical movement in an analogue meter is averaging or damping out the waveform or pulse train that is fed to it, so while what it gives you is limited in accuracy (like any measuring tool) it is going to show you something more or less meaningful.
        "A machinist's (WHAP!) best friend (WHAP! WHAP!) is his hammer. (WHAP!)" - Fred Tanner, foreman, Lunenburg Foundry and Engineering machine shop, circa 1979

        Comment


        • #5
          Well, sort-of.

          Usually, a decent RMS responding digital meter will read what the motor "sees".

          That depends on the "crest factor", the factor indicating how much the peak voltage is larger than the RMS. Too high and the reading may be low, because some peaks do not get measured as high as they really are. Most VFDs will clip peaks at the power supply voltage, so the meter generally reads right as to what the motor gets.

          If you measure between an inductor and the motor, then the spikes may be larger by some unknown (to me) amount.
          CNC machines only go through the motions.

          Ideas expressed may be mine, or from anyone else in the universe.
          Not responsible for clerical errors. Or those made by lay people either.
          Number formats and units may be chosen at random depending on what day it is.
          I reserve the right to use a number system with any integer base without prior notice.
          Generalizations are understood to be "often" true, but not true in every case.

          Comment


          • #6
            Hi Guys,

            What you are trying to measure is a low frequency modulation of a very high frequency square wave, 100 Khz or there about. The low frequency modulation is the 25 to 100 Hz that controls the motor speed. A digital meter, even an RMS reading one, wont give you a sensible reading !

            Best Regards:
            Baron J

            Comment


            • #7

              The inductance of the motor (and any series inductors) will smooth the current, but may introduce higher voltage spikes due to inductive "kick".

              I made a PWM "Variac" that modulates a sine wave power source with various duty cycles to produce a variable AC signal. The following images show the raw signal as well as what it looks like with a simple R-C filter with a knee of about 1 kHz. The first image shows the input sine wave and the modulated PWM output.

              Click image for larger version

Name:	PWM_Vernier_4226.jpg
Views:	250
Size:	130.4 KB
ID:	1909954
              Click image for larger version

Name:	PWM_Vernier_4230.jpg
Views:	249
Size:	115.5 KB
ID:	1909953
              Click image for larger version

Name:	PWM_Vernier_4231.jpg
Views:	253
Size:	122.1 KB
ID:	1909955
              Click image for larger version

Name:	PWM_Vernier_5077.jpg
Views:	254
Size:	110.1 KB
ID:	1909952
              Click image for larger version

Name:	PWM_Vernier_5080.jpg
Views:	252
Size:	96.3 KB
ID:	1909951
              http://pauleschoen.com/pix/PM08_P76_P54.png
              Paul , P S Technology, Inc. and MrTibbs
              USA Maryland 21030

              Comment


              • #8
                I think the fluke 1587 fc with low pass filter can read vfd output
                https://www.flickr.com/photos/csprecision

                Comment


                • #9
                  Originally posted by BaronJ View Post
                  Hi Guys,

                  What you are trying to measure is a low frequency modulation of a very high frequency square wave, 100 Khz or there about. The low frequency modulation is the 25 to 100 Hz that controls the motor speed. A digital meter, even an RMS reading one, wont give you a sensible reading !
                  The usual modulation in a VFD, even a lower power VFD, is between 4 kHz and 16 kHz. Most decent RMS reading meters can do that, and will give a good reading. I have used VFDs, and I have designed VFDs, and I can assure you that we had no particular trouble reading the output accurately with Fluke meters.

                  There are also straight analog meters which can read the RMS output as well, and both agreed.

                  Where you have a problem is with the AVERAGE RESPONDING meters (every digital meter that is not true RMS), which will generally fail to read the correct voltage, and give very misleading results. They normally read the average, and then apply a mathematical correction factor to arrive at the RMS. That factor only applies to a true sine voltage wave, and the VFD does not put that out. Therefore the results are generally just wrong, when using an average responding meter.

                  Ther may be RMS responding meters that do not have the bandwidth to handle that range, and those might not give a good reading either. The Fluke meters I personally have all have at least a 10 to 15 kHz bandwidth if they are true RMS (one IIRC is 50 kHz), and are accurate in the normal range of VFDs.

                  Many VFDs DO have a 32 kHz setting, but it is rarely used, because it gives no particular advantage in quietness, but it does have a distinct disadvantage due to increased heating at the 32 kHz setting. If used, a lower bandwidth meter might fail to give a good reading.
                  Last edited by J Tiers; 11-11-2020, 11:38 PM.
                  CNC machines only go through the motions.

                  Ideas expressed may be mine, or from anyone else in the universe.
                  Not responsible for clerical errors. Or those made by lay people either.
                  Number formats and units may be chosen at random depending on what day it is.
                  I reserve the right to use a number system with any integer base without prior notice.
                  Generalizations are understood to be "often" true, but not true in every case.

                  Comment


                  • #10
                    A typical VOM will treat an AC Voltage as if it it was a simple sine wave. In fact, the movements of mechanical VOMs and the electronics of many digital VOMs actually respond to the AVERAGE value of an AC waveform. BUT they are calibrated to translate this average reading into an RMS value which is what is commonly used to talk about AC Voltages. Only a very few, high end, and expensive meters will read true RMS values of AC waveforms other than pure sine wave. And those true RMS meters WILL provide meaningful readings of almost any AC waveform, including the pulsed outputs of VFDs and other circuits that use digital techniques to control the effect of the output Voltage/current in the devices that it is intended to power.

                    A good demonstration of the difference between average/RMS Voltage values and what is actually going on in the circuit can be seen with a typical VOM and an oscilloscope. Read the Voltage of any 115 VAC socket with a VOM and you will get a value that is close to 115 Volts. But if you read that same Voltage with an oscilloscope, you will see that the waveform will have a Voltage swing from zero to about 160 Volts. The meter is responding to the average value of the AC waveform and displaying it as the equivalent RMS value for a pure sine wave. But the oscilloscope is showing the true Voltage at every point on that sine wave. Overall, the oscilloscope will show the peak-to-peak value.

                    Another, often neglected specification for a VOM is the frequency response. Many may have heard the frequency response talked about with regard to audio or other amplifiers, but VOMs and other items of electronic equipment and test equipment also have a limitation due to frequency response. While a VOM may read correct average/RMS Voltages at low AC frequencies, like 60 Hz, many will have a fall-off in their readings for higher frequencies. Many meters have low readings for frequencies that are short of the often used upper limit of the audio range (20, 000 Hz). Some digital VOMs will not read AC Voltages as low as several thousand Hertz and perhaps even several hundred Hertz. If you are using a VOM to check an AC circuit, if is important to be aware of the frequency range of that meter. Meters with mechanical movements are more likely to cover the entire audio range than digital ones are. But do read the AC specs. for your meter.
                    Paul A.
                    SE Texas

                    And if you look REAL close at an analog signal,
                    You will find that it has discrete steps.

                    Comment


                    • #11
                      As for having the motor connected, a motor is not a proper low pass filter. It may have an integrating effect on the energy in the chopped waveform of a VFD, but that does not mean that it actually converts it to a true sine wave as seen as a Voltage or current. And, as I said above, most VOMs are only accurate with true sine waves.

                      The math for this gets complicated and involves vector manipulation because complex impedances are involved.
                      Paul A.
                      SE Texas

                      And if you look REAL close at an analog signal,
                      You will find that it has discrete steps.

                      Comment


                      • #12
                        The motor generates a back EMF, which is just short of the applied RMS voltage. With a typical sine output VFD the voltage on the motor will look generally like a sine, but will be somewhat jagged, as the pulses are applied and end. It will also have spikes to the peak voltage of the VFD power supply.

                        Depending on the meters and the motor, VFD carrier frequency selection, etc, both average and RMS meters may read fairly close to each other, or may differ considerably..

                        oddly, the higher the frequency, the better the sine wave. The meter may fail to read the high frequency "wiggle" of the voltage, and give a slightly erroneous reading, but the actual voltage wave that exists is not too bad.

                        You really run into trouble with the "modified sine wave" type inverter, which produces ONE pulse per cycle of "line" frequency, because the square wave is of a lower amplitude, so that the area under it is equal to the area under the similar sine (equal energies). That has th largest difference of average and RMS, so the worst difference of meter reading. For a 120V inverter output, the modified sine may get an average meter result of 108V or thereabout, while the true RMS gives a reading of 120V, the desired output (and reading)
                        CNC machines only go through the motions.

                        Ideas expressed may be mine, or from anyone else in the universe.
                        Not responsible for clerical errors. Or those made by lay people either.
                        Number formats and units may be chosen at random depending on what day it is.
                        I reserve the right to use a number system with any integer base without prior notice.
                        Generalizations are understood to be "often" true, but not true in every case.

                        Comment


                        • #13
                          The ultra-cheap (often free) Harbor Freight DMMs apparently use a series rectifier and capacitor, so it actually reads peak voltage and is scaled for RMS based on a sine wave. So it might very well read the peak voltage pulses of the PWM signal which will be the same for all output voltages.
                          http://pauleschoen.com/pix/PM08_P76_P54.png
                          Paul , P S Technology, Inc. and MrTibbs
                          USA Maryland 21030

                          Comment


                          • #14
                            its dangerous that some mutimeters consistently show zero voltage, including an analog one. the one that should be geometric jumps around and only shows values up to 40v (probably the frequency problem) and i found an unexpensive one that shows correctly.

                            how does the didital meter actually measure voltage? (i tried googling it with no result.)

                            Comment


                            • #15
                              Generally, they implement the "root mean square" mathematical operation of calculating "RMS". Can be done analog-fashion and then measured, or it can be done digitally by sampling the voltage and doing the calculation numerically in software. A fast DSP can do that fast enough to satisfy most requirements for bandwidth, sampling rate, etc.

                              Analog meters may have a slow rectifier that does not rectify higher frequencies well. A value that changes all over may indicate a sampling issue, it must sample faster than 2x the highest frequency of interest, or it will "alias" and usually give an unstable reading.
                              Last edited by J Tiers; 11-12-2020, 03:37 AM.
                              CNC machines only go through the motions.

                              Ideas expressed may be mine, or from anyone else in the universe.
                              Not responsible for clerical errors. Or those made by lay people either.
                              Number formats and units may be chosen at random depending on what day it is.
                              I reserve the right to use a number system with any integer base without prior notice.
                              Generalizations are understood to be "often" true, but not true in every case.

                              Comment

                              Working...
                              X