Cap Bank Protection

[HiSet]

Is over load protection more effective and meaning full than over voltage protection for protecting the capacitor cans from over voltages as any over voltage will necessariliy manifest into over load. Your comments will be highy appreciated.

 

[Skogsgurra]

If there is a risk for high harmonics contents, a thermal protection is needed. The reason being that the current produced by a 11th or 13th harmonic is 11 and 13 times higher than the corresponding fundamental voltage. An overvoltage protection alone will not catch that.

 

[busbar]

HiSet, could you give us an idea of the bank rating and voltage, as well as for individual cans? How are they fused?

In ANSI regions, IEEE C37.99 seems to be the prevailing standard.

 

[Bung]

An overvoltage on a capacitor doesn't necessarily amount to an overcurrent for the bank or a can as a whole, even if it does for individual elements.

Bung
Life is non-linear...

 

[HiSet]

Thanks for your comments.The bank rating is as follows
110kV,30 MVAR,Double Star unearthed,48 cans (2x8x3)each of 7.23 UF, Detuned via 70.4 mH reactor to (4.5- 3.6 )harmonic number.

 

[Bung]

Internally or externally fused cans? Best would be neutral unbalance protection - look for current between the two star points, and set to trip when that current corresponds with loss of sufficeint elements to lead to either a can O/V or an element O/V. The O/V is the danger - it will either push to much I throught the affected element, or just puncture it, resulting in cascading failures of elements. The calcs are easy in principle, if somewhat messy in practice. You also need to know (or make some assumptions) about element and can voltage withstand levels.


Bung
Life is non-linear...

 

[busbar]

Seems like IEC may apply for 110kV {50Hz?} Consider industry accepted practices published by your peers as the basis for a decision. It’s like a big jar of mosquitoes. Breaking the bottle will make them mad, but shaking the bottle will too. Ungrounded double-wye racks can make good use of voltage-unbalance protection.

 

[RalphChristie]

Busbar:

Maybe a stupid question, but why would you leave the bank ungrounded? Is it just for the purpouse of voltage protection? We have a 13.5MVAr Bank in our sub, (double star arrangement, 66kV) but it is solidly grounded through a neurtal CT. This CT have a ratio of 2/1. Have to say I couldn't believe my eyes when I first saw that ratio.






Failure seldom stops us, it is the fear for failure that stops us - Jack Lemmon

Regards
Ralph

 

[busbar]

 
Ralph, I don’t have a pat answer for selection of an ungrounded double-wye bank. It could be a matter of practicality, protection sensitivity or economics. For ANSI regions, IEEE C37.99 and 1036 cover a lot of territory with 100+ and 50+ pages, respectively. In North America below 121kV, shunt-capacitor banks in this configuration seem to be more commonplace than there grounded-wye equivalents.

 

[hmchi]

http://www.sandc.com/edocs_pdfs/EDOC_008444.pdf

While the predomiant practice in the world seems to have us design the capacitor banks in double Wye configuration so that we could use the 'bridge' between the 2 neutrals to sense the differential signal to afford capacitor can protection, there is a significant advantage in using a single wye [ungrounded] construction and neutral voltage as the input signal.

Generally speaking, the signal 'strength' produced in single wye bank Vng is much more pronounced and useful than the differential voltage or current in the double wye configuration, hence the protection can be more discriminating, more sensitive and more accurate.

This particular implementation also has the advantage of alarming before tripping so that damaged can(s) can be replaced at night without having to shut down the cap bank when it is needed most.

The capacitor bank inherent 3 phase unbalance as well as unbalance in the 3 phase supply, if any, can also be easily compensated in this configuration. This cap bank inherent unbalance is 'amplified' in the double wye configuration, especially if the true differential signal is 'drowned out' by this inherent unbalance signal.

The cost of building a single wye bank should be less than a dobule wye bank, and requiring less substation space.