Measuring current after VFD and DC-drives

I have done some tests to measure the current at the output of a variable frequency drives. Of course you should look at the current that the drive displays, but we had some doubts if the drive was still ok and wanted to compare the current indicated
by the drive with the current measured with a measuring clamp.
I could see that if the motor is running at speed (and not blocked mechanically or at speed 0), the current measured with the clamp is approximately the same as the one indicated by the drive. And this was even with a non true rms clamp.
However if the drive was blocked the current measured with the clamp was completely different from the current indicated by the drive.
Could I conclude that if the drive is running at speed the current measured with a clamp is approximately accurate, and the influence of a true rms or not a true rms clamp will be minimal??
What if the motor is not blocked but running at lower speed then the setpoint (torque reduced), will the measurement then be accurate???
How about measuring at the output of a dc-drive, is a true rms clamp then a real necessity, or will a non true rms clamp also do the job (like for the vfd).
Share me your experiences in measuring current with a measuring clamp.
Thanks.

First, why would you like to measure using a clamp? VFD's have display of the current and i think that should be your basis. i also have different readings with clamps at the output side of the VFD or motor side. I'm not sure about it, but since frequency is altering the motor speed, using a clamp to measure the current is inaccurate and VFD's can measure them electronically. in my case i base my readings on the display. Besides, VFD's are designed also to protect motors. Or if your concerns are the motors, you should install EOCR for added protection both to VFD and motor.

You could measure the current to the input of the drive.

Paul B said:

You could measure the current to the input of the drive.

Click to expand...

This will only work when you subtract the drive inefficiency, and the motor is running at line frequency, and the motor is running a near "stock" V/Hz curve. Too many and's in my book

My experiences have been that you "might" get somewhat accurate readings, when the motor is running lightly loaded. I have only used the clamp-on meter to verify the current existed and was somewhat close to what the drive reported, but I would not depend on it for accuracy.

We have used the Load Control UPC to get accurate load values from drives and motors, even with VFD outputs. These were used in full time power (not just current) monitoring applications, not as a testing tool.

I see they now offer a portable test unit that you might be interested in:
http://www.loadcontrols.com/downloads/PPC3datasheet.pdf

because a VFD makes stanges sinewave with multi small square wave,(i don't know the english term sorry for my origin) the resulting curent average would differt from quick amps reading. A true RMS clamp will average the noise to a useful current. Another clamp may not be true rms but may react slow enough to average the result and give a good reading too.

A good practice would be to install a choke inductor between VFD and motor and if you read the amperage on the motor side, you have better chances to get a good reading as the choke reduce harmonic current a little bit

In most case, i suggest to use the VFD reading instead as even a true RMS clamp that may still be fooled by the VFD noise....

The only fully accurate way to know the amperage is to put a scope on the reading and do the average yourself by calculating the "area under the curve" This action will help you undertand why a clamp may get false result too!!!

Reading the input of the vfd will be very inacurate because the power factor is kept close to 1 (100%) The "VAR" power portion (reactive current) is cancelled by the VFD capacitors so you almost just read the watts...And like at the output, Event if you wanna know the watts amperage, the input current is chopped at high frequency so the sinewave isn't clean enought to be read by a cheap clamp either...

PLCUser, as I have stated many times, use of measuring devices on the drive output is totally unreliable. You seem to have found a meter that has some correlation between actual and measured at certain loads and frequencies but what use is that?

A more interesting test would be to take ten different brands and styles of meters and do the same thing you just did. My experience would make me expect to see each different unit tune into an accurate reading at different points and tune back out at different points.

Interesting, but not very useful, especially when an accurate measurement is available from the drive.

And what if it concerns a DC-drive?
What if you use a true rms or a not true rms clamp at the output of a dc-drive?

In essance an AC drive is three DC drives. Same basic stuff switching off a DC bus. SO you still have PWM output on a DC drive and the same measuring difficulties.

I teach a class on AC motor with VFD applied to electric vehicles. I use a small "model car" driven with 100 W motor. I wanted to measure and display motor line current. The "best" ?? I have found is a regular darsinol (splelling I know) movement meter. I do not believe it is accurate but it is good to show differing current (roughly torque) readings at differing loads. BUT in this case all I am looking for is a change in the reading and using that to relate to increased load.

For this application the best I have found is a Kilawatt meter (Home Depot thing) on the 120 VAC line to VFD. This has had a pretty close correlation to volt and amp display on VFD. But I do not believe I was getting accurate readings.

In short based on DickDV and others recommendations to rely on the VFD display and my own attempts to do better than what the VFD can do
I conclude
TRUST THE VFD DISPLAY

Dan Bentler

There are two main types of DC drives, those that use SCR switching to rectifier the incoming AC to variable DC, and those that rectify the incoming AC to a fixed DC bus and then use PWM switching techniques to give a variable output to the motor.

On the SCR-switched drives, pretty much any ammeter that can read DC is ok to use since the current to the motor is essentially DC with varying amounts of ripple.

I haven't had much experience with PWM output DC drives but I would have to wonder about the accuracy of measurements there due to the mix of high frequencies and DC. Anyone have something better to offer than this?

Sorry for digging up old threads but it fits my issue. I have several sites that run Wellons Kilns and they all have VFD's on the fans. For example we have one with (10) 15HP motors and are ran off of a Powerflex 700 VFD. WE are trying to make sure that we are getting all we can from our fans so I need to know what the amperage is on all the fans which we can get close with an amp clamp. My question is the manufacture installed a single CT on the load side of the drive then is split to all the motors. Why would you do this if you can pull the amps from the drive? We can sometimes see a 10% difference between the drive and the CT. When I was looking at most of the sites drives, most of the CT's actually were higher then the reading on the VFD. I contacted our local AB supplier and they said this is due to the frequency of the drive. How can you have more amps on the load side of the drive then the line side? They told me that I would need to measure the amps with a Fluke 87 with a frequency compensated amp clamp. Has anyone heard of this?

I want to move all the amp readings from the CT's and use the drive analog output but I want to make sure that we don't have any issues with the motors.

Thanks,

Kevin

Most meters are designed to "sense" sine wave only. Any other wave shape and they will not give accurate reading.

It appears to me that with 10 motors driven off one VFD that what you are really trying to measure is are we getting enough cooling and is each motor / fan pulling its share of the load?

I would suggest taking air flow measurements on each fan to ensure each is moving the same amount of air give or take 10%.

Dan Bentler

You could try one of these.

Thanks Dan,

You are right on what we want and our general practice has been to set them the same with an amp clamp. What I need is to have the kilns operate as close to FLA as we can. The kiln system drives the fan speed based on the CT readings so we might be giving up some fan speed due to the CT's. if the FLA is 155A, I want to be close.

Now what Bit_Bucket_07 suggested looks interesting. Might be a good alternative if we want to monitor all the motors. I will find out why Wellons installs the CT's after he drives tomorrow.

Thanks,

Kevin

I believe the 700 has built in common mode chokes which is probably helping you get close with your handheld meter.

I am pretty sure there is a CT on each output buried in there somewhere otherwise how does the drive detect ground faults?

The drive readout, >99% of the time, is going to be more accurate although an average of the three phases.

I can think of maybe two or three times dealing with a drive running weird and giving me bogus current readout.

Use the drive display for your baseline it's more accurate and it's the number your drive is going to compare with its limit parameters in order to allow you to get the most from your fans without tripping.

It usually doesn't take much above base speed to make the current skyrocket, so if you simply program in the current limit you want and set the drive up to reduce frequency to stay under the current limit, then raise your speed command above where it can actually run, the VFD will keep the fan running as hard as it can all the time and allow the maximum amount of current to be used by the motor.

This might meet your needs at the expense of motor life, but if you don't go above the motor FLA rating with your programming, it may do just fine. I recall one application I had to set up about 25% above the motor rating in order to make it work at all. I let the boss know, hey we either lie to it and see if it burns up, or sit down for two weeks while we replace the whole drive train...Two years later, that 10HP motor still thought it was a 13HP and was still hot to the touch 24/7/365 but hadn't failed.

Paul