Jump to content


Photo

Power Factor Correction For Unbalanced Load


  • Please log in to reply
12 replies to this topic

#1 kana

kana

    Intermediate Member

  • Full Member
  • PipPipPip
  • 44 posts
  • Gender:Male
  • Location:Kuala Lumpur

Posted 28 September 2008 - 04:56 AM

Hi,

How can we do power factor correction for an unbalanced load in a low voltage distribution system?

Say,
Line Voltage: 415V
Phase A, I: 80A
Phase B, I: 60A
Phase C, I: 50A

Pf: 0.75 lagging

OR, is it a must to do load balancing before PFC can be applied?

Kana

#2 kana

kana

    Intermediate Member

  • Full Member
  • PipPipPip
  • 44 posts
  • Gender:Male
  • Location:Kuala Lumpur

Posted 27 August 2009 - 09:56 AM

Hi,

Sorry for refreshing this topic again. I'm dealing with a similar problem, anyone got an idea on power factor correction for unbalanced system?

Can we reduce the total kWH if we do load balancing for unbalanced system?

Regards,
kana

#3 marke

marke

    Posting Freak

  • Moderator
  • PipPipPipPipPipPip
  • 2,602 posts
  • Gender:Male
  • Location:Christchurch, New Zealand

Posted 27 August 2009 - 06:56 PM

QUOTE
Can we reduce the total kWH if we do load balancing for unbalanced system?

Adding power factor correction will not reduce the KWHr drawn by the load. It will only reduce the distribution losses between the point where the supply is metered and where the capacitors are connected.

To comment further on correcting the unbalanced load, we would need more details about the connected load.
If you have a three phase motor and some single phase load as well, the VARs for each phase may be the same and a standard three phase capacitor may surfice.

Best regards
Mark.

#4 kana

kana

    Intermediate Member

  • Full Member
  • PipPipPip
  • 44 posts
  • Gender:Male
  • Location:Kuala Lumpur

Posted 28 August 2009 - 02:37 AM

Hi Mr. Mark,

Thank you for the reply. There were a number of three phase loads like A/C compressors and single phase loads like fluorescent lights, split A/C units and computers. I have measured the ampere reading in each phase and it was about 70A/49A/31A and the neutral current was about 30A.

There was a penalty charged on this installation since the power factor drop to 0.8. I recommended a power factor correction unit to be installed and I sized it to correct the overall power factor to 0.95.

When computing the size of the capacitor bank I choose to use the 70A reading since it's the highest reading of all there phases. Can I assume the installation to be balanced when computing the reactive power?

The maximum demand shown by the kWHr meter was 40kW.

Can load balancing for this system reduce electricity bill besides the use of capacitors to eliminate the penalty?

Thank you.

Regards,
kana

#5 yuri

yuri

    Member

  • Full Member
  • PipPipPipPip
  • 93 posts

Posted 28 August 2009 - 05:45 PM

Sorry for interfering.
I consider a net with 31A on one ph an 70A on an other is asking for balancing.
40kW's consumer in any case is not a very big deal on my view, will costs connected with installing the unit justify the penalties paid now for the bad p.f.? What is the penalty for 1 kVAR (returning to mains)?

Best regards.

#6 yuri

yuri

    Member

  • Full Member
  • PipPipPipPip
  • 93 posts

Posted 28 August 2009 - 05:59 PM

QUOTE (kana @ Aug 28 2009, 03:37 AM) <{POST_SNAPBACK}>
Can I assume the installation to be balanced when computing the reactive power?


Can load balancing for this system reduce electricity bill besides the use of capacitors to eliminate the penalty?


I think a standard 3-phase capacitor will not suffice untill all 3 phases are load balanced.
Load balancing might not affect the bills at all.
Sorry again for interfering.

#7 marke

marke

    Posting Freak

  • Moderator
  • PipPipPipPipPipPip
  • 2,602 posts
  • Gender:Male
  • Location:Christchurch, New Zealand

Posted 28 August 2009 - 08:46 PM

My suggestion is to add static correction to each of the three phase motors.
To do this, add enough correction to neutralise 80% of the magnetising current to each starter. This will make a significant improvement.
It may then be necessary to add further correction to the inductive single phase loads, but I would wait until you have done the three phase loads first.
Computers can not be corrected using capacitors as the displacement power factor is better than 0.95. Fluorescent lights come in a number of formats and modern lights do not usually require correction.

Best regards,
Mark.

#8 kana

kana

    Intermediate Member

  • Full Member
  • PipPipPip
  • 44 posts
  • Gender:Male
  • Location:Kuala Lumpur

Posted 29 August 2009 - 11:43 AM

Thanks for the reply Mr.Mark and Yuri.

The penalty is about 8% of the total payable amount. The cost of installing capacitor bank is not really expensive solution after all considering the penalty paid.

Static power factor correction is a bit difficult solution. Itís hard to get the parameters of the three phase motors because the A/C compressors are hard to reach. Static power factor correction would cost more as there is a lot of work need to be carried out considering the number of loads.

The bulk power factor correction is only effective with balanced load.

This means I have to balance the loads in the distribution boards before I could apply the bulk PFC.

Regards,
kana



#9 marke

marke

    Posting Freak

  • Moderator
  • PipPipPipPipPipPip
  • 2,602 posts
  • Gender:Male
  • Location:Christchurch, New Zealand

Posted 29 August 2009 - 08:18 PM

QUOTE
The bulk power factor correction is only effective with balanced load.


I disagree. There is a combination of single phase load and three phase load. If you correct the three phase load, then the overall result will be much improved.
You may then need to add some single phase correction, but I would suggest that the single phase load may not need much if any correction.

If you wish to use a bank correction installation, then size it for the lowest phase current, and connect the feedback CT on that phase. This will effectively correct the three phase load. Once you have done that, then look at the power factor on each of the other two phases and see if it is a problem.

Best regards,
Mark.

#10 yuri

yuri

    Member

  • Full Member
  • PipPipPipPip
  • 93 posts

Posted 30 August 2009 - 04:26 PM

QUOTE (marke @ Aug 29 2009, 08:18 PM) <{POST_SNAPBACK}>
If you wish to use a bank correction installation, then size it for the lowest phase current, and connect the feedback CT on that phase. This will effectively correct the three phase load.
Best regards,
Mark.


Sorry but only the lowest loaded phase p.f.
would be properly corrected.

#11 marke

marke

    Posting Freak

  • Moderator
  • PipPipPipPipPipPip
  • 2,602 posts
  • Gender:Male
  • Location:Christchurch, New Zealand

Posted 30 August 2009 - 07:05 PM

QUOTE
Sorry but only the lowest loaded phase p.f.

So what will happen to the pf on the other phases?

If you consider the total load is composed of a mixture of three phase motor load (connected to all three phases) and single phase load which may not need correction, then correcting the three phase load will improve the power factor on all three phases even iff the current is unbalanced due to the single phase loads.

Take the case where there is a three phase motor on three phases and heating (resistive) load on two phases. For power factor correction, we are only concerned about the magnitude of the inductive current and this is still equal on all three phases. A three phase correction will correct all three phases even though the current is imbalanced.

Best regards,
Mark.


#12 yuri

yuri

    Member

  • Full Member
  • PipPipPipPip
  • 93 posts

Posted 30 August 2009 - 07:27 PM

Ubiquites nowadays in almost any building air conditioners (splits) and heat pumps are mostly 1ph.
Best regards.

#13 kana

kana

    Intermediate Member

  • Full Member
  • PipPipPip
  • 44 posts
  • Gender:Male
  • Location:Kuala Lumpur

Posted 31 August 2009 - 01:47 AM

Thanks again for the reply.

I will go with Mr.Mark's suggestion.

Regards,
kana




0 user(s) are reading this topic

0 members, 0 guests, 0 anonymous users