HT Capacitors

I know nothing about the harmonics involved, but you say "Capacitors - 175 kvar - 7.5 KV * 2 nos. in series * 4 banks = 700kvar in each phase."

As I interpret this, you have two 7.5 kV capacitors in series, and it sounds like you have four of these pairs connected in parallel for each phase. Two 7.5 kV capacitors connected in series can handle only 15kV with no safety factor, so of course they blow when connected to 22kV. Adding more capacitors or sets of capacitors in parallel does NOT make them withstand more voltage. You need four 7.5 kV capacitors in series to reliably withstand 22kv. Additional series sets of four can be used in parallel to obtain the capacitance needed.

I also presume as DK has said two 7.5 KV capacitors in series i.e. 22 KV

applied to them. How can they withstand 22KV.Further if only one capacitor

fails in a series system other capacitors are likely to fail immediately as 7.5 KV

is to to withstand 22 KV. Further the capacitive requirement cannot be fulfilled

as the capacitance is inversely proportional to the voltage.

As per above presumptions the capacitors will not withstand and need

redesigning. Further the question is for how long these capacitors being used?

Is this correct ?

Two capacitor with working voltage 7.5 kV , placed in series can withstand 15 kV. A 22 kV system voltage will normally destroy the capacitors.

BUT the capacitors are connected in STAR (wye) so the system voltage over the 2 series connected capacitors is only 22 / SQRT 3 = 12.7 kV ……

Yes, and that would be 12,7kVeffective = 1,4 x 12,7 = 18,6kV peak. Thats why I asked the owner of the question in my reply, to sit down with pencil and paper, and add up all the INSTANTANEOUS voltages. Still waiting for his reply, while we try to guess!!!

I missed that detail on my first reading of the post; the caps are indeed connected in wye. Of course you are correct.

I can't help wondering if there is any possibility that the replacements were wired in delta. Surely the replacement capacitors weren't rated 7.5 kV DC...

I can only imagine what it must be like to hear/see one of those blow...

Unfortunately, we have no idea whether the OP is seeing any of this!

Yes, DK is right too. When you have a series/parallel arrangement of the capacitors, two added factors come into play.

Capacitors have tolerances. The plate on the cap has to say something like: 2 uF (micoFarad) +/-20%. If the lower parallel set happened to be at -20%, and the uppers at +20% by accident, the voltage is not divided evenly. Roughly speaking, the larger gets 20% less voltage, the smaller gets 20% overvoltage, and potentially blows. One gets around that by measuring each cap real capacitance, and mix and match before installation.

Capacitors, no matter how well made, have leakage current thru them, that allows extra voltage buildup over the better one. Not good. Standard solution for that is an extra parallel resistor carrying around 5 times of the highest expected leakage current, that usually occurs when everything is hot.

All of these leads back to the vendor's technical personnel. I would strongly hesitate to do any modification without them signing off on it. Much rather, let them do it, explain and demonstrate the necessary modifications and measurement.

connect inductor in star pt. calculation of inductor depends on fault level of bus & gnd resistance value. generally manufacterer provide this data

The three factors which will destroy capacitors are:

1. Excess voltage

2. Excess ripple current.

3. heat from an external source

If the environment has not changed, and if your replacement caps are rated the same as the originals they should not blow unless there is another fault.

Measure the voltage and current on a good system, then the faulty on ..... trace it from there.