Tuned capacitor bank/filter configuration

[Gauss2k]

I am designing a tuned capacitor bank (which will probably also be used to filter some of the harmonics of the network, mainly 5th) for a factory. It's low voltage, 600V.

I modeled the plant under PSAF (made by Cyme) to simulate the network and determine the resonance and harmonic currents/voltage distortion. But now I added the capacitor bank itself and I need to choose the configuration of the bank. I was wondering; what is the impact of the bank configuration (delta, wye, wye-grounded,?)?

Of course, it will change the filter capacitor and inductance value since it will have a voltage of 600V in delta and 347V in wye. Is there any advantage, disadvantage, standards, etc. about each configuration?

 

[waross]

There may be a cost difference.
The KVARs of a capacitor increase as the square of the voltage. The line to line voltage is root three times the line to neutral voltage. That means that 3 times as many farads are needed for the same KVAR at 347 as are needed at 600 volts.
I note with interest that the NEC and the Canadian Electrical Code (CEC) require 135% rated conductors for capacitors as compared to 125% rated conductors for motors.
I note also that where in the case of motors, the CEC requires feeders for motors to be the sum of the motor currents plus 25% of the largest motor, feeders for capacitors must be 135% of the sum of the currents.
The 1990 CEC Handbook explains that there is a tolerance of +/- 15% on capacitor manufacture. With the second power relationship to the applied voltage this could lead to a current of 1.15^2 = 132% of rated current.
yours

 

[RalphChristie]

Hi Gauss2k

I stumbled across this site when I was searching for some information - maybe worth a check:

http://gilbertelectrical.com/library/document_library.htm

check especially:

http://gilbertelectrical.com/library/technical.discussions/fusing.article.htm

Regards




[red]Failure seldom stops us, it is the fear for failure that stops us - Jack Lemmon[/red]

Make the best use of Eng-Tips.com
Read the Site Policies at FAQ731-376

 

[Gauss2k]

Thank you for your answers!

To continue about this project, we are doing this design in order to correct the actual capacitor bank (un-tuned) that causes a resonance when it reaches certain stages (11th and 13th resonance resulting in high voltage distortion)

There is about 20% 5th current harmonic present on the bus where the capacitor bank is connected. Ideally, I would like to filter that harmonic, but after trying several simulations with the bank tuned at the 4.7th, 4.8th, etc with several capacitor voltage rating, I always have a scenario where there is a parallel resonance on the 4th harmonic. There is about no 4th harmonic current present, but I think it would not be wise to risk having problems with that particular configuration.

I tryed tuning at 3.7 and 3.8, but again, parallel resonance near the 3rd. The only attempt that seems to be successful is a 2.7 (162 Hz) or 2.8 (168Hz) tuned capacitor.

Is there something I should be aware with a capacitor tuned at that frequency? I saw several cases where the tuning frenquency was 3.7-8 or 4.8, but never at 2.7 or 2.8.

Also, I was just taking a look at active filters on google. They seem to be great, but I always heard that they were very expensive. Just by curiosity, anyone has bought one of these filters? What's the price range? I could contact a distributer, but I just want an idea of the price compared to a classic tuned capacitor bank, I don't plan to buy it soon