Current TDD limits and Harmonics Mitigation

Status
Not open for further replies.
E

eengr

Member
Location
GA
Occupation
Engineer
Hello all! I have a question regarding current harmonics. I have an AC drive (Siemens SM150) feeding four AC motors (1070 KW, 3300 V each) that is injecting current harmonics to the system at 2900 V. I have modeled it in SKM as a lump load and added a harmonic library to it. This load has a current THD of 20.9%. I do not have any other harmonics sources in the entire model and I do not have any filters in the model. I only have two reactors say M1 and M2 below the utility (13.8 KV). When I do harmonic analysis for the whole system and try to look harmonics at the PCC, I see that M1 has (going through) a TDD% of 54%. I do not understand how this level went so high when all I have is one harmonic source in the entire system.

If this can happen, how do I bring current TDD below IEEE 519 limit? As I already have two reactors, what else can I do to bring the harmonics down without impacting the system? Thank you.

Note: 1. Voltage harmonics (THDv) is under limit and is looking good.
2. I do not have exact demand factor for the loads so I assumed that all the loads are connected and running.
 
I did take out the reactor M1 and replaced it with a cable (with very low impedance). Current TDD% went up, a lot higher. I have not tried taking out both the reactors as M2 is looking good in harmonics perspective.
 
Have you compared the distortion spectrum of the current through M1 to that from the load to check whether certain harmonic orders are being enhanced more than others?
Also, a plot of the current waveforms might provide some insights into what's going on.
 
synchro said:
Have you compared the distortion spectrum of the current through M1 to that from the load to check whether certain harmonic orders are being enhanced more than others?
Also, a plot of the current waveforms might provide some insights into what's going on.
Click to expand...
Yes, I have looked at the distortion spectrum and it does not have any enhanced harmonic orders. This is how the current waveform and the spectrum looks like through the series reactor. But it has %TDD of 85% (sorry it is not 54% as I mentioned in my first post)

1626900812069.png


The harmonic source that I have modeled at the end load (there are several other loads but I modeled only one load as a harmonic source) has some higher 5th and 7th orders (0.11 PU and 0.16 PU) compared to others which has 20.9% THDi.

Does this has anything to do with the demand factor? I don't understand it because I was using it as a worst-case (all loads connected and largest motor has a factor of 1.25 and remaining as 1) which seems fine to me. I do not have any capacitors as well so I am assuming that there is no issue of resonance.

Also, the software shows that TDDi through the series reactor is 85% where as THDi is 3.71% which looks very weird as it implies that there is huge difference between fundamental current and load current.
 
Last edited:
I don't exactly what you are achieving but that's OK.
My field was in power electronics .We often used twelve or twenty four pulse systems up to 4MVA. That stopped the harmonics in their tracks.
Yes, it needed transformers.
 
Besoeker3 said:
I don't exactly what you are achieving but that's OK.
My field was in power electronics .We often used twelve or twenty four pulse systems up to 4MVA. That stopped the harmonics in their tracks.
Yes, it needed transformers.
Click to expand...
I have modeled my load now as an IEEE 24 pulse source and this has brought down the TDDi to 13.83% and THDi to 0.6%. My question is, at full load THD should be equal to TDD and if the total load changes to 75% or 50% or 25% etc. the TDD value goes down relative to THD. But in this case it looks so weird as the TDD is very high compared to THD. Is this possible?
Even when I used my harmonic spectrum from the drive, TDD was at 85% and THD at 3.71%. Is that normal? I don't understand and for me it seems that something is wrong. If it is not normal, I want to discuss what could be the issue or where should I look at to see if there is a mistake.
 
And thank you everyone for taking your time and expressing your views. I would love to see a discussion on THD and TDD and if my case looks weird or if it is normal to have values like that etc.
 
eengr said:
And thank you everyone for taking your time and expressing your views. I would love to see a discussion on THD and TDD and if my case looks weird or if it is normal to have values like that etc.
Click to expand...
THD is the harmonics at a certain load level. TDD is the current harmonics at the max demand. A system could have a high THD at a load level but will be very low at the maximum demand (TDD). Say you have a THDi of 40% at 100A demand. The TDD at full demand of 500A will just be 8% (TDD= THDi x (100A/500A)! Check your load current if it decreased.
 
topgone said:
THD is the harmonics at a certain load level. TDD is the current harmonics at the max demand. A system could have a high THD at a load level but will be very low at the maximum demand (TDD). Say you have a THDi of 40% at 100A demand. The TDD at full demand of 500A will just be 8% (TDD= THDi x (100A/500A)! Check your load current if it decreased.
Click to expand...
Thank you for your reply. Yes, I agree that. So you are implying that at the max demand, THD = TDD and for other loaded conditions THDi has to be greater than TDD, right? My case looks weird then. I don't know why the software is showing very high TDD and so little THDi. Also, the above waveform looks almost clean (I have all loads in service and connected) but I don't get why TDD is at 85% and THDi is at only 3.71%. Because, now this is out of the IEEE 519 limits and I have to do some type of mitigation. I don't want to propose any solution with this being an invalid case.
 
eengr said:
Thank you for your reply. Yes, I agree that. So you are implying that at the max demand, THD = TDD and for other loaded conditions THDi has to be greater than TDD, right? My case looks weird then. I don't know why the software is showing very high TDD and so little THDi. Also, the above waveform looks almost clean (I have all loads in service and connected) but I don't get why TDD is at 85% and THDi is at only 3.71%. Because, now this is out of the IEEE 519 limits and I have to do some type of mitigation. I don't want to propose any solution with this being an invalid case.
Click to expand...
Nope. TDD is the THDi you get during your harmonics scanning multiplied by the ratio of the currents at scanning time and the max demand current. If you were fully loaded during scanning, your THDi result is your TDD. The reason why I asked for the load current is because there is no other way for your TDD to get higher that the THDi except when the demand load to have decreased! Your software could have treated the reactors as restraining impedances, hence a lower current. Suggest to specify pure reactors in your model (no resistance whatsoever).
 
Status
Not open for further replies.
Top