Converting solidly grounded MV system to impedance grounded

[rockman7892]

A wastewater plant that I am currently involved with operates their 4.16kV system solidly grounded. Two of their other plants are impedance grounded however for whatever reason this particular one is solidly grounded. (70's vintage plant)

I have approached the customer recently with discussions regarding converting the solidly grounded system to a LRG system and explained the benefits of doing so (reduced equipment damage, personnel safety, etc....)

The service entrance transformers to the plant are utility owned and are 13.2kV - 4.16kV Y-Gnd/Y-Gnd transformers. They are actually four 2500kVA transformers in parallel which make up a 10MVA service (There are two identical service entrances into M-T-M Switchgear).

Before I pursue this any further I have a few concerns that I was hoping others may have experience with.

1) Are you able to impedance ground the secondary of a Y-Gnd / Y-Gnd transformer ore are the implications that prevent you from doing so? All the LRG/HRG systems I have ever seen have been from a Delta/Wye transformer.

2) Can an LRG be implemented with a case like this where there is not a single transformer but rather 4 transformers each most likely having a neutral? Is it possible to bring all of the neural points to a common point before inserting a LRG? (I'm not that familiar with paralleled transformer configurations for utility services)

3) Seeing that these are utility owned transformers do utilities typically allow impedance grounding of the secondary if it is possible with the Y/Y configuration?

Thanks

 

[Skogsgurra]

If there are thyristors or VFDs, you will need at least one delta winding to let triplenes circulate. But since such drives usually are on a 480 V or similar grid, the delta winding will be on the 4.16/.48 kV transformer.

So, my experience doesn't say that you will have any problems with a grounded secondary. But others may have other insights. What about code? Nothing there?

Gunnar Englund

http://www.gke.org

--------------------------------------
Half full - Half empty? I don't mind. It's what in it that counts.

 

[rockman7892]

Skogsgurra, there are several 4.15kV MV drives at this plant for large pump motors however I believe each one of these drives has an internal isolation transformer which I'm assuming would have a delta winding. Good point, and something I will need to confirm.

This location is in North America so I don't believe there are any code issues to my knowledge, but perhaps I am overlooking something.

I'm curious why you need a delta winding to let the triplens circulate, can you explain this to me? What happens if there is not delta winding to let the triplens circulate? Is allowing the triplens to circulate only an issue with and LRG on the neutral or is this an issue anytime there is an absence of a delta winding?

 

[Skogsgurra]

The need for a delta winding is to keep impedance for 3rd, 9th etcetera harmonics low. If you don't do that, there will be a reduction of the RMS feeding the inverters and that reduction increases with higher loads so that the inverters cannot reach full output voltage at higher loads. That could be a problem especially with pumps with their high power consumption at higher speeds and consequent need for full motor voltage.

Mathematically, the explanation can be written simply as 3x120=360 degrees. Very simplistic, but it says it all. If there is no low impedance path for the triplenes, the transformer output will have a high reactance in series and that reduces the voltage. All other harmonics, including the fundamental, will not see that high reactance.

I once had a large pony motor for a still larger synchronous refiner motor and that one could never be taken up to full speed until we got a delta/wye transformer in place. Will look for some links on that. Are the inverters Robicon? If so, they have internal transformers.

Gunnar Englund

http://www.gke.org

--------------------------------------
Half full - Half empty? I don't mind. It's what in it that counts.

 

[waross]

Gunnar, May I ask for a clarification?
I understand that this is an operating plant that is presently fed by a grounded wye primary and a grounded wye secondary.
Do I understand that a delta winding may need to be added if the grounded wye secondary is converted to a high resistance grounded secondary?
If there is a problem with triplen harmonics, will it also present on a grounded wye, grounded wye system?
Thank you for the heads up on possible issues with harmonics.
Thank you sir.

Bill
--------------------
"Why not the best?"
Jimmy Carter

 

[Skogsgurra]

Sorry, I read more into the question than there was. I was thinking along a path that Rockman didn't describe, namely that there would be a need to change a delta winding to a wye winding.

The reason is that it is very common to have two transformers with different groups to achieve twelve-pulse rectification and that only one of them is solidly grounded (the delta cannot be grounded, for obvious reasons), but when I re-read, I see that that isn't the case. My error.

Bill, if the primary is grounded and the feeder is also grounded, there are already paths that can handle the triplenes, so I can understand why you ask this question. No confusion meant, but probably created. OTOH, I am not so sure if a grounded feeder is a good idea, tends to create interference in old-fashioned analog telecommunications.




Gunnar Englund

http://www.gke.org

--------------------------------------
Half full - Half empty? I don't mind. It's what in it that counts.

 

[waross]

Misunderstanding aside, Thanks again for the information on potential harmonics issues.

Bill
--------------------
"Why not the best?"
Jimmy Carter

 

[rockman7892]

Another reason for considering converting to LRG is due to the fact that the plant has several zero sequence CT's on several of their relays. With a solidly grounded system these CT's may saturate on ground fault and not provide the proper relaying protection.

The plant also has emergency generators that are resistance grounded. Although these generators do not operate in parallel with the utility, there is a fast transfer scheme when going from generators back to utility where they will be momentarily paralleled. One of the additional concerns that I have is that the ground fault protection is setup for a resistance grounded system when operating from backup generators but operates on a solidly grounded system when operated form utility. So ground fault relaying may be set up for one or the other but operating from both.

 

[rockman7892]

Looking into this further I notice that with a wye-gnd/wye-gnd transformer zero sequence current will pass through from the primary to the secondary and thus the zero sequence impedance of the primary system will be in series with the secondary zero sequence network which includes the NRG. Does that mean that the ground fault on the secondary will be further limited below the NGR value by the primary zero sequence impedance? For instance if a 400A NGR is chosen will the secondary L-G fault current be significantly less than 400A since the NRG is in series with the primary zero sequence network?

Since there are several 4.16kV Delta / 480V Wye unit substation transformers located throughout the plant does this mean that these transformers will source zero sequence current from the primary delta's for a fault occurring on the wye primary of the utility transformer?