VFD Output Voltage, Current, Frequency 7

[RogerMM]

I would like to know what the actual output curves look like from a VFD. I have read several comments and from what I have read there is a difference of opinion. If the PWM voltage is constant and only the frequency is varied what is going on with the volts/hertz ratio? It may be that we must include the motor inductance to be able to see what the curves the motor is seeing. I have read much about VFd's but, it is still confusing to what is actually taking place at the motor. The motor seeing constant voltage with a varying current frequency?

 

[TommyOO7]

RogerMM:
The PWM output Voltage is not constant, the max. value (amplitude) is constant, and equal to DC Link voltage, but first harmonic of the voltage varies with frequency. This, first harmonic, is the voltage seen by motor. Motor bahaviour is dictated by this voltage/frequency combination, and other harmonics cause the parasitic effects.

Regards
Tom.

 

[CJCPE]

The first harmonic or fundamental value of the voltage varies because the ratio of "on time" to "off time" is modulated by modulating the width of the pulses that make up the waveform. This picture shows the general idea, but the number of pulses per cycle is actually much higher than shown:

 

[jraef]

Where are you getting the notion that voltage is held constant and only the frequency changes? At opeating speeds under and up to the motor design frequency, voltage abd frequency are changed together to maintain a constant V/Hz ration (except when used on VT applications, but that's a little specialized). Only when you reach the maximum line voltage can the frequency increase without an increase in voltage, a VFD cannot create additional voltage. So when attempting to drive the frequency beyond the motor's design frequency you have that problem, but only then.

Eng-Tips: Help for your job, not for your homework Read faq731-376

 

[RogerMM]

I would like to know what the effective voltage equation is for the PWM the motor sees as the volts/hertz ratio is held constant. If the effective is linear then, is it a direct ratio for volts and hertz? I.E. 230 Volts, 30 Hertz and so forth?

 

[TommyOO7]

V/f ration is constant up to nominal frequency, after that the voltage is constant and V/f ratio is decreasing.
At the vary low frequences, the ratio is increased to compensate the stator resistance (stator voltage drop)

Tom.

 

[TommyOO7]

You are right 230V => 30Hz !

 

[CJCPE]

For most operation below rated motor frequency, the V/Hz ratio is constant, 480V at 60 Hz, 230V at 30 Hz etc. For optimum performance, there could be some increase or decrease from the nominal V/Hz value, but constant V/Hz is the general principle.

 

[Marke]

The one thing to be carefull of, is this is a PWM output waveform. If you try to measure the output voltage with a meter, it will not read correctly in most cases.

Best regards,

Mark Empson

http://www.lmphotonics.com

 

[jraef]

Thats right, many meters read PWM output voltages incorrectly. Old style analog D'Arsonval meters are actually better for that than many digital meters. I keep my old one specifically for that reason. I have tried several digital "true RMS" meters that claim to be accurate on PWM outputs, but have yet to find one that performs as well as my old Simpson.

As others have said, the ratio is held constant for most general purpose applications, and is determined by the motor nameplate voltage and frequency rating (done during setup of the VFD parameters). So for a 460V 60Hz motor, the ratio is 7.6 V/Hz because 460 [÷] 60 = 7.6. So a VFD set up for a 460V 60Hz motor will mantain that ratio throughout the speed range, and preyty much in a linear fashion. There is a slight twist to that in variable torque applications where you want to save extra money. The ratio is tweaked to further reduce energy consumption at lower speeds because less torque is required there.

Coincindentally, a 380V 50Hz motor is essentially the same because 380 [÷] 50 = 7.6V/Hz. That is why you can often use 380V/50Hz motors in the US even if you are not controlling them with a VFD, they just turn faster. On a 230V motor, the V/Hz ratio is 230 [÷] 60, or 3.83V/Hz, but a 230V 50Hz IEC motor is going to be 4.6V/Hz, so they are not as interchangeable.

Eng-Tips: Help for your job, not for your homework Read faq731-376

 

[ozmosis]

To confuse matters more, there are more modern VFD's on the market that, as well as having the pre-defined linear and variable V/F characteristics, can also be configured for customised V/F curves. This is particulary useful when using a 'standard' PWM VFD to control synchronous motors where the V/F curve needs shifting somewhat to your specific motor.
Terms such as 'torque boost' that allow a few more volts at the lower frquency range also act to shift the volts higher than the expected, so measuring volts at lower frequencies can seem higher if you are not aware that the feature 'torque boost or voltage boost' has been enabled.
Ultimately you are trying to provide an optimum volt/frequency to the motor to ensure the magnetising current (flux gap) is adequate to ensure the motor can produce the required torque.

 

[RogerMM]

I have a VFD that the displays shows twice the current reading versus a clamp on meter on the motor leads. I have contacted the manufacturer and they have not resolved the problem. What do you think?

 

[CJCPE]

A lot of older type clamp-on meters work something like a transformer. As the frequency approaches DC, the reading should go to zero. Near 50/60 Hz they should give a reasonable reading. Hall effect and iron vane (Columbia Tong-test) meters should work at low frequencies.

 

[TommyOO7]

The difference in current reading could come from the motor parameters entered into the VFD. Did you check that motor size, speed, etc. ... entered into VFD are correct?

Tommy

 

[ozmosis]

typically, the current measured within the VFD should be more accurate than a meter as they are specifically designed to measure a PWM waveform and the displayed current on the VFD should be the figure that is used internally within the drive to monitor for protection and detection levels. I think I'd be looking at the meter first!

 

[Skogsgurra]

More confusion.

A vector controlled VFD adjusts its voltage to suit the actual load. So it is not at all unusual to see a rather low V/Hz ratio on lightly loaded motors. I have had a few calls from people that were convinced that something was wrong with there drives - took some time to explain this to them over the phone.

Also, to measure the motor voltage it is not enough to have a True RMS meter. It will show the RMS of the PWM waveform and that is a lot more than RMS of fundamental. I use a simple filter with a 1 Hz corner frequency to check motor voltage. The 1 Hz corner frequency makes its output voltage = U/f so I read 8,0 V if the motor is a 400 V/50 Hz one. The good old ITT/mw/Metrix MX210 clamp-on has this filter built in and the measured value is recalculated (times f) before being displayed. There are filters for current as well as voltage.

Many recorders have filters with 1 or 5 Hz corner frequencies that can also be used. But beware of high voltage and short circuit currents!

Gunnar Englund

http://www.gke.org

 

[ozmosis]

there are a few good documents issued by Fluke about measuring VFD waveforms:

http://us.fluke.com/usen/support/ap...lukeProducts)&parent=APP_NOTES(FlukeProducts)

you have to register but it is free. A good document is called: "Why true-rms? Non-linear loads need a true-rms
current clamp for accurate readings"

 

[jraef]

I have a VFD that the displays shows twice the current reading versus a clamp on meter on the motor leads. I have contacted the manufacturer and they have not resolved the problem. What do you think?

I think you have found a manufacturer with poor technical support!

Any decent VFD tech support person should know that you cannot get accurate current readings with most clamp-on ammeters on the output of a PWM VFD. The VFD itself probably uses an internal Hall Effect transducer for that very reason. That's why you get the advice to trust the VFD reading rather than the meter; the issues have already been dealt with in the VFD because it needs accurate current readings in order to function properly and protect the motor.

If it is important for you to verify the VFD's reading then skogsgurra's sggestions are appropriate.

Eng-Tips: Help for your job, not for your homework Read faq731-376

 

[RogerMM]

We have resolved the problem. The manufacturing tech was unable to help us. We changed out the VFD and the display agrees with our clamp on meter. I appreciate everyone's help.

 

[itsmoked]

That is wierd.

What model clamp on do you have?