Leading Power Factor Issues?

[Robert789]

Aside from the associated voltage rise, are there any problems that would likely result from a leading power factor on distribution lines during light load periods?

I am planning to add some fixed capacitor banks to a rapidly growing area of our electric system, and would like to add slightly more than the base KVAR requirement on a few circuits.

I would look at switched banks, but equipment problems in the past have caused me to lose faith in their ability to operate as desired in an unsupervised mode.

Any input on this topic would be appreciated.

 

[redtrumpet]

thread238-25981

 

[Robert789]

I did a search for 'leading power factor' before posting, with no hits returned. I should have seen the thread from just two weeks ago regardless. Thanks

 

[electricpete]

One area that I didn't see covered in the thread.... some devices may have lower interrupting capacity for capacitive currents. Something to do with restrike and recovery voltages and all that jazz. A more precise explanation of this aspect at:

http://www.squared.com/us/products/switchgr.nsf/DocumentsByCategory/6C04856EAABF9DAF852565D400704AD6

 

[Robert789]

I wouldn't expect line protective devices to see a load that is substantially leading. I think that I have seen a discussion in the past about switches on the capacitor rack itself. I believe that the issue there is the current peaking at the same time that the voltage is crossing through zero.

Also on the capacitor issue... Every so often, a local telephone company that rides our poles will ask about capacitor banks that we might have in a particular area where their customers are complaining about telephone line noise. We will typically send a serviceman to meet one of their technicians and remove any capacitor banks in the area from service to help them troubleshoot. As far as I know, taking our capacitor banks offline has never resulted in any improvement in their condition.

 

[electricpete]

You're right rgm... it's a pretty unlikely that current would still be leading after you added the fault current which is likely limited by series inductance/resistance. I guess I didn't think too carefully before I posted.
Thanks.

 

[jghrist]

A small amount of overcorrection should not cause a problem. I suggest adding fixed capacitors to each circuit with a single bank that is the next standard size above the minimum kvar load.

There may be a penalty for leading power factor from transmission suppliers. One of the transmission providers in the southeast US has the following provision in their Service Agreement for Network Integrated Transmission Service:

"Valley Periods - The Transmission Customer must operate its electrical system in a manner resulting in a 100% power factor or lagging power factor as measured at the primary (high) side of the transformer at the hour of transmission system valley on a monthly basis."

The transformer will add reactive vars, so this allows a somewhat leading power factor on the secondary side.

 

[Robert789]

We will have to meet new power factor requirements from our transmission supplier beginning next year. Penalties will be implemented for power factor below 96.5% on peak during the summer months. Penalties will also be implemented for leading power factor during valley periods in the spring and fall months. There will be no leading penalties during summer or lagging penalties during spring and fall.

I expect to try and install only fixed capacitors on distribution lines. Then add some switched capacitors as needed inside substations where I can monitor them through SCADA. Every manufacturer claims to have great capacitor controls and switches, but each of the ones that I have tried in the past have developed severe operational problems within a year or so.

 

[SooryaShrestha]

One of the possible problem with capacitors in distribution lines may be ferroresonance, if the transformers used are of three phase, low loading with single phase fuses as protection. Please check

 

[infomaniac]

A slight leading power factor is no problem. Your main concern as you already have mentioned is voltage rise during lightly loaded periods. A rule of thumb is to limit the voltage rise to 2-3% but it is not set in stone. You can calculate the approximate voltage rise on a transformer with just the capacitors connected (i.e., no load) as follows:

V rise (in %) = (Transformer %Z * KVAR)/(Transformer KVA)

Note: Keep the %Z as a percentage (i.e., do not convert it by dividing by 100)

 

[jghrist]

Is the 96.5% pf requirement at the transmission voltage? If so, this means a considerably higher power factor at the distribution level. The supplier I referred to in my earlier post has (or will have if it is ever implemented) a 96.5% requirement on the high side or 98% on the low side at the time of the transmission system peak.

It will be interesting trying to comply with this. You won't be able to switch in steps small enough to ensure a 98% pf at all times without going leading sometimes. If you can predict when the transmission supplier peak will occur, SCADA control seems like the way to go.

Make sure that you do at least an approximate calculation of resonant frequency to make sure that it is not near the 5th or 7th harmonic. Example:

5200A fault available at 12.5 kV (112.6 MVA)
4.5 Mvar capacitor bank

resonant h = sqrt(112.6/4.5) = 5 (see IEEE std 519 section 8.2.1.)

 

[Robert789]

The 96.5% requirement is at the transmission level. We may be talking about the same power supplier (Duke). I don't plan to control the switched banks using SCADA, but just monitor them and see that they appear to be switched in when needed. I expect that we will pay some penalties rather than try to fully meet the requirements that the power supplier has requested.

 

[jghrist]

We are talking about the same supplier. I don't see anything in the Service Agreement (for the City of Seneca, SC) about differences between summer and other months. This agreement also does not take effect until Duke implements the requirements for itself. No telling when that will be. I also don't see how Duke can require a particular pf during their peak hour without providing a load signal or something for the customer to predict when that peak hour might be. I don't think they will provide such a signal for competitive reasons.

How do you plan on controlling the capacitors? Time, vars, volts, amps, or some combination?

 

[Robert789]

I will try controling based on temperature. Most of our load is residential, with air conditioners being the largest motor load. My thinking is that since the air conditioners are controlled by a thermostat, we would essentially be controlling our switched banks on a thermostat as well.

We do have a few circuits that have more commercial and small industrial load. On these circuits, I may try to switch based on amps or vars.

Duke used to provide us with a load signal (through NCEMC), but I believe that they stopped doing this a couple of years ago.

 

[reactive]

In South Africa we tend to use bulk PFC as opposed to distributed PFC on MV systems i.e. at the MV intake point (although in your case remote banks on the lines may be preferable depending on your voltage problems). These banks are typically controlled by a PLC controller and have been found to be very reliable provided the capacitor/reactor(if reqd.) equipment ratings are correct.

Depending on local harmonic generation and the system resonance frequency, the banks may be required to be in the form of a detuned filter (probably tuned at say 4,8 p.u.). As your target PF is only about 0,96 (as opposed to close to unity)you may resonate above crical frequencies (5th, 7th) and get away without a filter.

Leading VARs are not a problem and voltage rise and potential VAr overloading of transformers are generally all that has to be considered.