help with converter waveforms

[jimmy2times]

We have measured quite high levels of even current harmonics in the input to our VSI npc drive system (24 pulse, obtain by two 12 pulse transformers shift by 7.5deg with respect to one another). Attached are the 6 pulse current waveforms on the secondary side of one of those transformers. Does someone have guidance, or otherwise an online reference, with assisting to interpret the converter waveforms and what we could further investigate to find out source of such harmonics. All references I find with harmonics is related to the characteristic harmonics that are expected in such systems, e.g. 5th, 7th, 11th, 13th etc but no guidance or information on waveforms found in practise that might not represent the textbook.

thanks.

 

[davidbeach]

Even harmonics in a case like that are generally a symptom of asymmetric firing of the power electronics (active or passive).

 

[HamburgerHelper]

Yeah, I don't know of anything else that generates even harmonics or interharmonics. Maybe, you got a bad MOSFET on your drive.

Here is a class I took at University of Idaho that is still online that might help.

http://www.ece.uidaho.edu/ee/power/ECE528/

Here are the two text that went with the class. An old pdf of the Dugan book can be found online and I think if you have an IEEE membership you have access to it, too.

Primary Text: Electrical Power Systems Quality, Third Edition by Dugan, McGranaghan, Santoso,
and Beaty. (McGraw-Hill, 2012, ISBN 978-0-07-176155-0). (Abbreviated PSQ in class)

Secondary Text: Fundamentals of Electric Power Quality by Surya Santoso. Available at
Amazon.com (Abbreviated FPQ in class) Note: the ISBN number of the latest edition is:
978-1440491023. This is the Winter 2010 edition. The 2009 edition is also acceptable

 

[jraef]

On a VSI drive, your input (rectifier) power devices are either diodes or SCRs. I'd say you might have SCRs and one of them is not firing.


" We are all here on earth to help others; what on earth the others are here for I don't know.
" -- W. H. Auden

 

[LionelHutz]

Well, the double hump waveforms show that all the rectifiers are working. The offset from 0 is very odd and I can't come up with any kind of theory to explain it.

Typically a single rectifier fails but it's very obvious in the patterns with pulses missing.

 

[jimmy2times]

Front end rectifiers are naturally commutating diodes. Sorry I should have said

HH: Thank you for the course link, from first glance it looks like good reference material

 

[LionelHutz]

You only provided data for one of the transformer outputs.

What did the other one look like?

Are the rectifiers in series or parallel?

 

[jimmy2times]

Unfortunately we didn't measure the other 12 pulse transformer secondary waveforms. We opted to look at the motor output current from both inverters instead. by that time we had run out of analysers. We are doing some other tests in a few weeks would you recommend looking a the other transformer as well? Any other suggested measurements to take?

the rectifiers are in series.

we were using flexible rogowski coils for current measurements. could that result in a false interpretation of zero offset?

 

[LionelHutz]

With the rectifiers in series, I'm just wondering if a failed rectifier on the other transformer could cause that offset current you're seeing.

 

[jimmy2times]

Just to clarify previous statement. The motor is dual wound. The rectifiers of each 3 winding transformer are in series. So i guess that makes the rectifiers of each converter isolated from one another, though both converters are working in unison to deliver torque together at the motor winding output (one conveter master one converter slave). We will look to measure all 4 secondaries windings next time to see if issue is also occuring on 2nd converter input.

 

[MatthewDB]

I've seen rectifier waveforms like this a few times. What we found was that the DC link inductance and capacitance were resonant at the 2nd harmonic. When the line voltage was balanced, there was no problem. But a slight line voltage imbalance, the imbalance produced a 2nd harmonic at the rectifier output. That caused the DC link to ring and the harmonics were reflected to the input.

 

[jimmy2times]

I found a useful paper that reinforced some suggestions/advice above. Thought i'd share link here. As stated by others it says main reason for even harmonic generation is mis firing/ unequal firing. Doesn't help us much though with diode front rectifcation stage though.

I'm wondering if this is could lead to other issues in the driven mechanical system? (Electric propulsion application), we have other issues such as cavitation. I've been reading up on torque pulsations and wonder if these measurements show our system is unheathy?

https://www.researchgate.net/profil...rharmonics-of-power-electronic-converters.pdf