Power factor correction 1

[freefallingbody]

Hi,

This is a question related to improving power factor.

I have a 22 kV supply from the electricity supply utility grid and five transformers ( 2 x 1000 kVA, 3 x 500 kVA) step teh voltage down to 433 Volts. The utility billing is based on a meter connected at the 22 kV line.

The five transformers supply power to various dedicated sections of the plant. The load on these transformers are not steady due to ON/OFF operation of Motors. 20% to 100% is the order of change in the transformerload pattern.

I have a few LT capacitors installed at the main LT panels at transformer output. No HT side capacitors due to high cost of HT switchgear. I also have an automatic power factor controller on the LT side of one 1000 kVA transformer.

I am not able to maintain unity power factor at the utility meter point; its only 0.96. If I maintain unity, I get a discount of 4% on energy bill. My question would be:

Is there any way I can use the existing capacitors and auto p.f. controllers to maintain unity p.f? One way I found is to sense the power factor at 22 kV meter side ( using a p.f. sensing relay) and feed signal to the auto p.f. controller at the LT side.

Is this likely to work? Any other schemes?

Thanks in advance

Dinesh

 

[waross]

There may be a problem with over-voltage. As I understand your question, you want to do the correction on the secondary of one transformer to correct for five transformers. There may be times when the power factor is low on the other transformers and the transformer with the automatic correction is over correcting to compensate. This can lead to over-voltages being generated by the overcorrected transformer.
A 4% correction doesn't sound as if it would be a problem, but you must consider how much overcorrection one transformer system will need to correct four other transformer systems.
yours.

 

[RalphChristie]

If I maintain unity, I get a discount of 4% on energy bill.

Are you sure about this? I have never seen a requirement to maintain unity - normally the utility's requirement will just be not to go below say 0.96 or 0.97. Below that you'll be penalized. Also check for resonance (resonant condition) at pf of 1.

[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

 

[freefallingbody]

Waross,

You have understood my requirement. Now, the capacitors connected to the underloaded transformers are kept off manually when the load is less. And when I measured reactive power requirement at the LT side loads on all the transformers, it has shown positive. ( my meter will show negative p.f. and reactive power if overcompensating). hence I may rule out overcompensation. What I have observed is that the acapcitors in Auto p.f. controllers are not fully ON, even when the p.f. at 22 kV supply side is 0.96.

Ralphchristie,
This system is in India, where may state owned utilities are giving 1% rebate on energy bill for every 1 % increase in p.f above 0.95. Penalty starts if teh p.f. drops below 0.9.

 

[waross]

freefallingbody
I would check the settings and the documentation for the auto P.F. controller. If the input signal is showing .96 and the capacitors are not all on, it is probably set to correct to .96.
A couple of points that you probably know; If the correction is connected on the secondary and the metering is on the primary you must consider the KAVARs of the transformer. If the secondary is at 100% p.F. the primary will be less by the amount of the magnetizing current of the transformer.
In your case there will probably be times when the system fed from the transformer with the auto P.F. correction will be over corrected.
This may be acceptable. The question is "How much over correction and how much over voltage."
I would suggest watching the secondary voltage closely when the P.F on the system with the auto p.F. controller is over correcting.
You may also try connecting enough capacitors to each of the other systems to supply the magnetizing current of the transformers without auto correction.
Measure the no load primary current of the transformers. Multiply by the transformer ratio and select a capacitor that will draw this much current. If you have lighting transformers connected you can add capacity to correct them also. Continue to switch the main capacitors depending on the load.
Correcting the power factor with permanently connected capacitors is a balancing act. So many things depend on site conditions that almost any general advice can be contradicted in specific instances.
What is the responce of the utility if your power factor is leading?
Do they let you "bank" KVARHrs for credit when your P.F. is low?
Do they use a metering scheme that ignores leading P.F?
Do they apply the rebates and penalties the same regardless if the P.F. is leading or lagging?
respectfully

 

[freefallingbody]

Waross,

Thanks for ur suggestions. Thats what I am going to do. Keep the capacitors needed for magnetising kVARs ON. Switching off all capacitors at no load ( what my client practiced till now) may not be a good idea.

In the auto p.f controller, the sensing point now is at the secondary of one 1000 kVA transformer. If i keep base ON capacitors for taking care of magnetising kVARs and keep the auto p.f controller at 1.0, i think the problem wil be solved. May be I have to go on doing that for all five transformers.

The utility people are smarter than they appear. No banking of kVARh possible, leading or lagging makes no difference to them. i.e. my rebate will be zero if i do social service by maintaining p.f. at 0.95 lead.

 

[waross]

Is your incoming voltage stable?
One problem with fixed capacitors for transformer correction is varying voltage. The KVARs vary with the square of the voltage. This is usually not a problem but your situation is probably the most stringent that most of us have seen.
Sensing the primary with the the auto P.F. controller will compensate for this. Consider a primary current transformer to feed the current input of the auto P.F. controller. Feed the voltage input from the secondary of the 1000 KVA transformer. How are your transformers connected? Star-Star is easy. for other connections we will have to correct the phase angle between current and voltage or possibly a potential transformer.
respectfully