EEVblog Electronics Community Forum
Electronics => Beginners => Topic started by: JayMan07 on September 24, 2020, 12:45:43 am
-
Ok so I've been doing some experiments with a VFD that caused me to start looking at the waveform.
This pic is of the waveform with probe at x1
(https://www.eevblog.com/forum/beginners/can-anyone-make-sense-of-this-vfd-waveform/?action=dlattach;attach=1072942;image)
This pic is of the waveform with probe at x10
(https://www.eevblog.com/forum/beginners/can-anyone-make-sense-of-this-vfd-waveform/?action=dlattach;attach=1072946;image)
WTF is going on here...the VFD is set at 220VAC. It doesn't matter whether there is a load or not the waveform is the same. The x1 looks horrible and the x10 completely looks normal. Is it my scope or what?
-
Based on the detailed information you’ve posted, any responses will be nothing more than wild ass guesses.
-
Let me know what details you need and I'll try and provide them.
The drive is set at 220VAC at 60Hz. I don't understand how a x10 attenuation would show a sine wave when a x1 does not. Again any details you need I'll be happy to provide them! I'm just not sure what to think hence the post in the beginner section lol.
-
You could start by describing the device that is shown. Name, Model number. What is the bandwidth of the vertical amplifier?
What probe are you using for X1 and for X10 measurements?
What vertical scale factor is used in each case?
Do you specify the X1, X10 to the scope or do you need to take the probe factor into account when interpreting the amplitude?
-
I’m gonna guess the issue is the scope since it is a very low resolution lcd with a display that appears to be some kind of dedicated device rather than a general purpose scope. But, since we know really nothing other than what can be seen in two photos, it’s all pure speculation.
Let’s start with:
What are the scope horizontal time base settings for the two photos?
And if that doesn’t answer the mystery, then we’re into details like:
What model scope (so we can have some idea of its specs)?
What model probes (to know if they’re compatible with the scope)?
What brand/model drive (so we have some idea what the output should look like)?
What points on the vfd or motor are you measuring? Phase to phase? Phase to ground?
After we slog though all that, the final question will probably be something along the lines of why are you actually bothering to measure this signal? What are you doing that isn’t going according to plan that has you putting a scope on a PWM waveform that typically doesn’t mean squat in the bigger scheme of things.
If the drive is a commercial device of reasonable quality and in working order, the only thing that really matters is that the drive is outputting the correct volts to Hertz ratio for the motor and there is little to nothing to be learned by scoping something that you can and should just measure with a DMM (so long as it is a true rms meter).
If you can learn something out of this, great, I’m all for it. But, if you’re connecting a $50 no name Chinesium scope with $5 probes to a device that has upwards of 400 volts dc potential between the output terminals of the drive and ground, you’re into lethal voltages which does not play well together with the wrong equipment and practices. All joking aside, mains voltage and scopes don’t mix well at all and this is especially true for new players.
-
Upon some further study, it appears the “probes” are just banana test leads plugged into the top of the device.
What make makes one set x10 and the other x1?
-
Thanks for the replies! And yes I hear you about the lethal voltages. I work on lethal voltages every day so I do get that. I was using my portable device initially for isolation purposes. Anyway, since most of the questions seem to be about set up, device, and probe, I went ahead and isolated my bench top scope since it'll provide more information. BTW I know you're supposed to isolate the DUT but I have no way of doing that so I had to isolate my scope. The scope is a Siglent 1104X-E using the probe it came with (i.e. x1/x10 probe) set to x10.
Here are some captures. I guess I'm just trying to understand why it is hard to get an accurate reading on these things, whether it's by DMM or scope. For example, even using a fluke 87V I have to use the low pass filter to get an accurate voltage reading. I'm assuming the Siglent is accurate, but the waveform is different than my portable scope. So far I am seeing something different on each device/setting. Assuming the siglent is accurate it's a very peculiar waveform so I guess I was just wanting to know more about what is going on.
-
Upon some further study, it appears the “probes” are just banana test leads plugged into the top of the device.
What make makes one set x10 and the other x1?
I had a banana to bnc adapter that was connected to the siglent x1/x10 probe
-
Nothing suspicious with the waveforms on the bench scope.
What’s the underlying problem you’re troubleshooting?
-
Nothing suspicious with the waveforms on the bench scope.
What’s the underlying problem you’re troubleshooting?
I'm not suspecting anything is wrong with the vfd so I'm not troubleshooting it. I just noticed it was difficult to read voltage accurately with a DMM so that got me interested, since I obviously don't know much about how they actually work. I expected to see a crude sine wave or something similar to what an inverter put out. I have worked on a lot of 3 phase equipment that runs off mains not vfds. Because mains is a perfect sine wave I assumed a vfd would have to simulate that. Instead with the bench top scope I'm seeing positive pulses followed by negative pulses. It's almost more like dc than ac and I don't understand that given that 3 phase equipment is really designed to run off clean ac.
-
Your Siglent shots just show how sine-wave output is produced
It is created by pulse width modulation. Wider pulses close to the crest of the curve, narrow pulses close to zero.
For smooth voltage, you would want some filtering.
For VFD, nobody cares about filtering, as motors don't care about voltage, they need sinusoidal current, and their inductance makes a good enough filter for current.
Your portable device tries to show the same waveforms, but due to very limited resolution these look very strange.
-
Your Siglent shots just show how sine-wave output is produced
It is created by pulse width modulation. Wider pulses close to the crest of the curve, narrow pulses close to zero.
For smooth voltage, you would want some filtering.
For VFD, nobody cares about filtering, as motors don't care about voltage, they need sinusoidal current, and their inductance makes a good enough filter for current.
Your portable device tries to show the same waveforms, but due to very limited resolution these look very strange.
Thanks that's the clarification I was looking for! And helps this next waveform make sense. I can adjust the voltage of the VFD.
The REF waveform is the VFD set at 220V (the scope is reading RMS of 287V)
Channel 1 waveform is the VFD set at 120V (the scope is reading RMS of 215V)
What's cool is that it seems like all that is really happening is that for the 220 there is a longer on time but peak to peak and everything else is the same.
-
Correct, there is only 3 states available at any output, positive bus voltage, open circuit or negative bus voltage. Thus the swings high and low, and the bits in the middle is the average of the other 2 phases being reflected through the motor and the winding inductance, appearing to be around 0V, plus some voltage developed by the motor running as an alternator so the VFD can, if intelligent enough, determine slip and load on the motor to compensate for it.
To read the voltage accurately you need a true RMS meter that can handle both the frequency and crest factor ( how high the peak is compared to the RMS voltage) correctly. Otherwise you use a resistive load and measure temperature, which integrates the pulses. You can connect 3 incandescent lamps in place of the motor, using (assuming you are in a 230VAC country, and are using this to run something meant for 60Hz operation on 50Hz mains) 3 230VAC incandescent lamps in star, where they will light up brighter or dimmer depending on the output voltage you select, though they are being fed 340V pulses only.
-
As said before, nothing going on here I would not expect.
Some 3-phase motors have some special features for running from a VFD. Some of those features are:
* Thinner laminations to mitigate effects of eddy current because of switching frequency.
* Electrically isolated bearings to prevent induced current pass through them.
* Forced cooling by external fan if high torque at low RPM is needed.
But 1.5kW motors are relatively expensive parts, so I just used what was already there on my 30+ year old mill, and it still works.
Some of the industrial type DMM's have extra features for measurements on VFD's.
For example the Brymen BM869s
http://www.brymen.com/images/ProductsList/BM860s_List/BM860s-manual-print1-IEC61010-1-3rd-r3.pdf (http://www.brymen.com/images/ProductsList/BM860s_List/BM860s-manual-print1-IEC61010-1-3rd-r3.pdf)
I would not like to use current measurements with a DMM on any such equipment. Opening a lead between a VFD and a motor will lead to instant fault on the VFD, and may even damage it. Instead, for current measurements, add some shunt resistors in the cable. Something like 0.1 Ohm 10W seems a nice value, and then measure the voltage drop over the shunt with your DMM. If you do these measurements more often, then make a box with the shunt resistors, isolated banana plugs and screw (or other) connectors on both sides for the motor wires.