Harmonics

[cornbread]

We recently hooked up a power meter to a 480v substation feeding several MCC line up with VFD's. I'm seeing TDH voltage in the 3 to 4 % and current @ 30%. I thought below 10% was "not to worry" and anything above 10% could cause you problems down the road. Looks like the main contributors are the 5th, 7th harmonics, kinda of expected given the number of VFD loads. I can't remember where I got the 10% rule. Any comments, links or thoughts on the matter would be greatly appreciated.

Example of the current data
1 531
2 5.3
3 2.9
4 1.8
5 154
6 0.7
7 57
8 0.4
9 1.0
10 0.7
11 26

 

[mivey]

We recently hooked up a power meter to a 480v substation feeding several MCC line up with VFD's. I'm seeing TDH voltage in the 3 to 4 % and current @ 30%. I thought below 10% was "not to worry" and anything above 10% could cause you problems down the road. Looks like the main contributors are the 5th, 7th harmonics, kinda of expected given the number of VFD loads. I can't remember where I got the 10% rule. Any comments, links or thoughts on the matter would be greatly appreciated.

Example of the current data
1 531
2 5.3
3 2.9
4 1.8
5 154
6 0.7
7 57
8 0.4
9 1.0
10 0.7
11 26

What is the rated load current? We usually express % current harmonics as a function of rated load or historic peak load.


Add: And that would be individual maxes of: 4% for <11th order, 2% 11th to <17th, 1.5% for 17th to <23rd, 0.6% for 23rd to <35th, 0.3% for 35th+. The total TDD should be below 5% total.

 

[mivey]

What is the rated load current? We usually express % current harmonics as a function of rated load or historic peak load.


Add: And that would be individual maxes of: 4% for <11th order, 2% 11th to <17th, 1.5% for 17th to <23rd, 0.6% for 23rd to <35th, 0.3% for 35th+. The total TDD should be below 5% total.

Those are for odd harmonics. The evens would be 25% of the values listed above.

 

[jim dungar]

The problem area is THD-V, the cause is THD-I.

If you do not have voltage problems your current harmonics are probably not much of a concern.

 

[mivey]

Any comments, links or thoughts on the matter would be greatly appreciated.

IEEE Std 519-1992 would be your resource document

Add: I just noticed you said substation. The value I gave were for a short-circuit to demand load ratio below 20. The allowances are higher for higher fault sources.

 

[mivey]

The problem area is THD-V, the cause is THD-I.

If you do not have voltage problems your current harmonics are probably not much of a concern.

...so if you stay below 5% on the THD-V, you are usually ok.

 

[Mayimbe]

The problem area is THD-V, the cause is THD-I.

If you do not have voltage problems your current harmonics are probably not much of a concern.

Depends

THD-I reflects on the K Factor, K Factor reflects on derating transformers, derating transformers cause a resizing of the neutral (2 times bigger than phase conductors, usually). The rumor says that here, a whole important building burn down, because nobody toke in count this "K Factor". The neutral was under dimensionated so it burned.

 

[jim dungar]

Depends

THD-I reflects on the K Factor, K Factor reflects on derating transformers, derating transformers cause a resizing of the neutral (2 times bigger than phase conductors, usually). The rumor says that here, a whole important building burn down, because nobody toke in count this "K Factor". The neutral was under dimensionated so it burned.

The OP said he was dealing with Motor Control Centers and VFDs, so I was pretty sure that there is not a lot of L-N loading.

 

[cornbread]

It's a 1000KVA xfmr feeding AB MCC's with just about every motor on a VFD.

 

[Mayimbe]

The OP said he was dealing with Motor Control Centers and VFDs, so I was pretty sure that there is not a lot of L-N loading.

So? Dont see your point. As far as I know, harmonics are created by non linear loads. VFDs (electronic equipment = non linear loads) and MCC (electronics too). L - N loading stands for Lineal Neutral, or for Lineal Not??

I was making an opinion to help the OP, I wasnt refusing yours.

 

[jim dungar]

So? Dont see your point. As far as I know, harmonics are created by non linear loads. VFDs (electronic equipment = non linear loads) and MCC (electronics too). L - N loading stands for Lineal Neutral, or for Lineal Not??

I was making an opinion to help the OP, I wasnt refusing yours.

3-phase VFD's are connected Line-Line, not Line-Neutral, therefore they do not produce triplen harmonics that are additive on the neutral conductor.

In my opinion, and experience, almost no transformer feeding motor control centers and VFD's needs to be K-rated unless it is continuously being loaded well past 80%.

 

[Besoeker]

So? Dont see your point. As far as I know, harmonics are created by non linear loads. VFDs (electronic equipment = non linear loads) and MCC (electronics too). L - N loading stands for Lineal Neutral, or for Lineal Not??

I was making an opinion to help the OP, I wasnt refusing yours.

Jim is right.
If you look at the harmonic currents given in the original post you will see that the magnitude of the 3rd, 6, and 9th (triple n) they are small compared to others.

On the other hand, 5th, 7th, and 11th have the greatest magnitude by far.
The major harmonics from most three phase VSDs is given by 6n +/-1 where n is an integer.
Thus, you get 5th, 7th, 11th, 13th 17th, 19th etc.

 

[Mayimbe]

Ok. Obviuously you both (Besoeker and Jim) got more experience than me, and if you say that no transformers feeding VFDs and MCCs needs to be K-rated, I will agree entirely. But theres something I dont understand. When you say that the triplen harmonics are smaller than the others, and by that I understand that the neutral wont have any problem, as you say. My doubt is that the formulae to get the K factor that I have, takes in count ALL the harmonics, not only the triplen. So my question is:

If some where in the future I have a transformer Feeding VFDs and MCCs, Should I or should I not see that the K factor is ok? from your experience.

 

[jim dungar]

My experience is that K-rated transformers are not required where the loading is primarily 3-phase loads connected Line-Line.

For the most part, but not always, K-rated transformers are a marketing gimmick. However, I can see a need for 200% rated neutrals where the predominant loading is Line-Neutral.

 

[Besoeker]

Ok. Obviuously you both (Besoeker and Jim) got more experience than me, and if you say that no transformers feeding VFDs and MCCs needs to be K-rated, I will agree entirely. But theres something I dont understand. When you say that the triplen harmonics are smaller than the others

They are smaller than the others in this particular case and for this specific application.
I have had experience of cases where the triple n harmonics were a very serious problem. One was in a large luxury hotel where a considerable part of the loading was single-phase dimmers. The problem wasn't so much for the transformers but the inadequate rating of the neutral busbars in the distribution system. I got some work out of it when the problem surfaced but it should have been spotted by the electrical consultants at the design stage.