Impacts of Series Capacitors

[davidbeach]

I'm analyzing a possible line with series capacitors and I'm not sure I'm understanding the results. Is it possible that when a highly compensated line is in parallel with non-compensated lines that fault currents can be so out of phase that the total fault current is actually reduced?

 

[magoo2]

David,

Under fault conditions, the protective equipment usually bypasses the capacitors. When you did your calcs, did you assume that the capacitors are still in?

Magoo2

 

[davidbeach]

Fault currents do not in all cases exceed the current at which the caps are bypassed (looking at impacts well beyond the terminals of the line) and the caps don't go to full bypass instantaneously. The model accounts for bypass MOV conduction but not for bypass breaker operation.

 

[subtech]

David

Is there sufficient compensation to cause the faulted line angle to be less than 30 deg?

 

[QBplanner]

David
Do you have a simple diagram shown the system and the fault location?

 

[waross]

Hello David.
I have a very old text book that discusses transformers in parallel. There is an example of two transformers in parallel with the same impedance but differing X/R ratios. The current vectors show that the total current is less than the sum of the transformer currents.
I believe that this effect will also be shown with your application in the instance of faults with the capacitors in the circuit.
Further to this, if I was presented with an example of parallel uncompensated lines with differing X/R ratios due to different line spacing or differing conductor resistance, I would expect the sum of the individual line currents to be greater than the actual total current.
In some instances the actual difference may be so little as to be immaterial but it should be possible to calculate a difference however small.
I apologize for the delay in responding, David. I was hoping that someone with direct experience would respond. Failing that I hope that this "Blast from the Past" may be helpful.

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

 

[magoo2]

Bill,

I played with a few values but couldn't get it to come out the way you said it was. On the 2 transformer situation, could you tell me the 2 impedances and R and X values?

Maybe we'll hear back from David before long.

Thanks.

Magoo2

 

[davidbeach]

Was away from the simulation software all weekend.

Can't provide a truly simple one-line; just imagine a geographic region that has 12-15 lines in one "path" presently, 2/3 of which are 500kV, the rest lower voltages, none of which are compensated. Then throw a pair of 70% compensated 500kV lines into the path, as well as all the other interconnecting lines, and I don't know how much of it can be "simplified" without changing the results.

With full compensation I see fault currents about 180 out of phase from when there is no compensation, or the compensation is very small. I need to do some more investigation and will report back if I find something interesting.

 

[davidbeach]

Just gotta do a better job of checking other people's work. The caps were sized to compensate 80% of the X0, not X1. No wonder I'd see a 180 degree swing.

 

[waross]

Hi magoo2;
With transformers it's fairly simple.
Transformer #1; The available short circuit current will depend on the impedance and the phase angle of the current will depend on the X/R ratio of the transformer.
Transformer #2; The same conditions apply but if the X/R ratio is not the same the phase angle of the current will not be the same.
The total available short circuit current at the fault location will be the vector sum of the available short circuit currents of the two transformers.


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

 

[magoo2]

Bill,

You're absolutely correct. Unless the two transformers are identical, the magnitude of the total fault current will be less than the sum of the individual magnitudes.


David,

If I understand what you said about your engineer modelling the series capacitor based on Xo instead of X1, you likely have a compensation of 2 or more times the 80% figure in the model. So your modelled series compensated line is basically an R - C circuit. If you parallel this with the normal R - L circuit (or equivalent), then the fault currents will be roughly 180 degrees out of phase with respect to each other. Like Bill's example, this results in a total fault current magnitude less than the sum of the 2 magnitudes.

Magoo2

 

[QBplanner]

"The caps were sized to compensate 80% of the X0"

that will be over 300% comopensation of X1 for a typical 500kV line in north America.

" geographic region that has 12-15 lines in one "path" presently, 2/3 of which are 500kV"

That is a quite complex system. I don't see it in West coast area.

 

[magoo2]

QBplanner,

He said he made a mistake in the modeling. What is in service is series compensation for 80% of X1 not 80% of X0.

Magoo2

 

[davidbeach]

Lines are years away from being in service, fortunately. They will go into a WECC path that presently has that many lines. Path can be 100 miles wide in places, but it is a recognized path.

 

[QBplanner]

Magoo2 I simply stated the consequences if compensating 80% of X0. Nothing does beyond that.
BTW, I am in WECC system too.

 

[davidbeach]

I had been given values for three sets of series caps:
-0.0982
-0.0570
-0.0356

The correct values are actually:
-0.0281
-0.0164
-0.0117

No wonder they were giving me fits.

 

[waross]

Thanks for the update, David.
Yours
Bill

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

 

[QBplanner]

Good to know that in WECC system there are still excellent engineers like David doing studies themselves.

Here sometimes we have been forced to contract studies out to consultants.