We have a 1000kVAR de-tuned 11kV PFC bank and want to de-rate it to 400kVAr. It consists of air cored reactors in series with five capacitors. It was not designed to correct for specific harmonics. Is there any reason why we cannot simply disconnect three of the capacitors?
Eng-Tips is the largest engineering community on the Internet
Intelligent Work Forums for Engineering Professionals
Reducing Cpacity of De-tuned PFC Bank
- Thread starter GerH
- Start date
The reactor is specifically designed with a certain kVAR to tune at a point away from any harmonics (De-tune). If you change the kVAr without changing the reactor you might end up tuning on one of the significant harmonics and amplifying it.
- Thread starter
- #3
Thanks KJLRSA. Have calculated that LC resonant frequency is 177Hz, 197 Hz with 4 caps, 228 Hz with 3 caps, 279Hz with 2 caps. Do you think it will be OK if after reducing the number of capacitors the resonant frequency is not within 10Hz of a harmonic of 50Hz e.g. if it is not close to 200, 250Hz etc. If this is true 2, 3 or 5 caps would be OK, 4 would not. Also was planning tests using 400V - would this be representative of results at 11kV?
The LC series resonant frequency causes a low impedance path to current at that frequency. You want the LC resonant frequency to coincide with a harmonic source frequency to provide a path for harmonic currents other than through the utility system. This is the purpose of the filter.
You need to worry about a high impedance of the filter in parallel with the utility system. There will be a parallel resonant point with the system at a frequency somewhat below the tuning frequency of the filter. You need to make sure that this resonant point does not coincide with a large harmonic current source. Currents at the parallel resonant point can cause a high harmonic voltage.
You need to worry about a high impedance of the filter in parallel with the utility system. There will be a parallel resonant point with the system at a frequency somewhat below the tuning frequency of the filter. You need to make sure that this resonant point does not coincide with a large harmonic current source. Currents at the parallel resonant point can cause a high harmonic voltage.
rcw retired EE
Electrical
Also check for parallel resonance of the system with the capacitor.
h= Square root (MVAsc / MVARcap)
h is the harmonic order for the parallel resonance of your caps and your system.
MVAsc is the short circuit MVA at the bus where the caps are connected and MVARcap is the capacitor MVAR rating at the operating voltage. If h is near a frequency that might be present in your system, you could have a problem.
By reducing the capacitors, you are raising the harmonic frequency. If it happens to land on the 11th harmonic and you have some VFD's producing 11th harmonic currents, you need to investigate further.
h= Square root (MVAsc / MVARcap)
h is the harmonic order for the parallel resonance of your caps and your system.
MVAsc is the short circuit MVA at the bus where the caps are connected and MVARcap is the capacitor MVAR rating at the operating voltage. If h is near a frequency that might be present in your system, you could have a problem.
By reducing the capacitors, you are raising the harmonic frequency. If it happens to land on the 11th harmonic and you have some VFD's producing 11th harmonic currents, you need to investigate further.
After re-reading the OP, I see that the device is not a tuned filter, but a detuned capacitor. In that case, the purpose of the reactor is to avoid harmonic frequencies that are present, thereby preventing excessive currents in the capacitor. You still have to worry about parallel resonance with the system as pointed out by rcwilson.
KJLRSA,
If the reactor is bypassed, you will know the "tuning point". The natural frequency of the capacitor bank will be determined by the reactance (inductance) up to the capacitor bank and the capacitance of the capacitor bank itself. I don't think bypassing the reactor is a big deal because we would normally not have one installed on banks of this size unless there was an identifiable source of harmonics present and none is mentioned, but you seem to think it will cause the capacitors to smoke. Why do you think it will smoke?
GerH,
Seems like an unbalanced capacitor bank to start with. 1000 kvar or 400 kvar is not a balanced rating. What capacitor can size are you using? 200 kvar?
If the reactor is bypassed, you will know the "tuning point". The natural frequency of the capacitor bank will be determined by the reactance (inductance) up to the capacitor bank and the capacitance of the capacitor bank itself. I don't think bypassing the reactor is a big deal because we would normally not have one installed on banks of this size unless there was an identifiable source of harmonics present and none is mentioned, but you seem to think it will cause the capacitors to smoke. Why do you think it will smoke?
GerH,
Seems like an unbalanced capacitor bank to start with. 1000 kvar or 400 kvar is not a balanced rating. What capacitor can size are you using? 200 kvar?
- Status
Similar threads
- Locked
- Question
- Locked
- Question
- Locked
- Question
- Locked
- Question