PF correction. 1

[itsmoked]

Hello, Walking to my office today I came to a lineman in a bucket messing with a three capacitor bank and a control box that was halfway down the pole.

I asked him if this was a power factor correction bank. He responded in the affirmative.

I asked what the small cans were(coffee can size) near the big rectangular caps. He stated that they were "wet switches" used to disconnect the capacitor bank. He went on to say since it was 21kV that it was a little hazardous to do manually.

I asked if they could be remotely operated. He said yes, but not in this case. He said they where, in fact, hooked to the control box. He went on to say, since this was a housing area the inductive reactance of the neighborhood drops significantly each night, so the controller disconnects the capacitor bank each night at 11pm and reconnects it at 5am.

I asked him, "Why? Don't you just get closer to a unity power factor late at night, which is the whole object, why disconnect it?" His response was "we found we saved 36.4A by disconnecting it at night".

Questions:
1) Is he saying that these three capacitors are consuming 36.4A at 21kV? That would be about 750kW why isn't the pole glowing incandescent?

2) Isn't the utility struggling to reach a PF of 1 all its days, while never being able to actually get there? Why would you do anything to reduce available capacitive reactance?

3) What is the typical low voltage distribution system PF in a non-dense residential neighborhood in Moderate(size) Town, USA?

Thanks for the forthcoming enlightenment!

 

[dpc]

Because the daily load cycles can be extreme on residential and commerical feeders, it is easy to have leading power factor during low load periods if the cap banks are not switched off. Automatic switching via oil, vacuum, or SF6 switches is really common.

Power wholesalers in my area actually penalize utilties for both lagging and leading power factors, so overcorrection is a concern.

But typical power factors for residential feeders are generally pretty good - maybe 0.85 to 0.95 uncorrected.

Industrial facilities are another matter. Around here, we often see sawmills with power factors of 0.2 to 0.4.

HTH

 

[PowerfulStuff]

A couple of other points-
As voltage level is effected by P.F., it is possible for system overvoltages where toom much P.F. correction is in place.

Also most generators are designed for lagging P.F. so going into leading P.F. also heads closer to instability.

Finally the 36.4 A would be reactive (i.e. 90 degrees out of phase with voltage - no real power transfer) although it may have a few percent losses associated with it. The pole doesn't get hot and there is only minimal savings switching it on or off.

 

[alehman]

Capacitors serve two purposes in distribution systems -

(1) power factor correction, which is used to reduce energy losses in the system and reduce the reactive power load on the generation, and

(2) voltage control. Adding capacitance to a system with inductive load will increase the voltage. In this case, at night when the load may be low, the voltage may go too high if the capacitor is left on.

 

[itsmoked]

Utility power factor *CAN* get to leading!

Leading PF can bring on generator instability.

Leading at a facility can cost you money just like lagging!

Industrial can actually drop to dismal levels, 0.2!!

So the 36A in this case was reactive current and not real current. The reduction of 36.4A being a monetary savings is null and void.


alehman:

I have seen a capacitor bank on a pole and looked into the control box. Inside it said "VOLTAGE CONTROL UNIT".

There was a pointer that looked like it set the voltage. This particular pole/system was supplying a large quarry. Is this what you are referring too?


How can a few capacitors control the voltage delivered to a customer/neighborhood?
My guess is; if you reduce the lagging reactance you would suffer less voltage drop thru all the transformers feeding the area, hence raising the voltage?? (can't be - too weird)

Can this voltage be controlled in a linear manner?? Or is it just a step correction; 1 set of capacitors => 1 correction step?

 

[electricpete]

Adding shunt capacitance will almost always boost voltage at that point and adding shunt inductance will almost always reduce voltage.

As a first approximation the real and reactive power flow are decoupled. Real power flow determines phase angle and reactive power flow determines voltage distribution. As reactive power flow through series inductance (for instance the trnasformers you mentioned, also lines), voltage decreases in the direction of reactive power flow. So all other things being equal if you decrease reactive flow toward a point or increase reactive flow away from a point you will increase voltage at that point.


=====================================
Eng-tips forums: The best place on the web for engineering discussions.

 

[tinfoil]

To put it in simple terms, if you bring the power factor towards unity, you are reducing the kVA for a given kW.

Well, this is reducing CURRENT for a given 'real' load.

Ohms Law: Reducing current reduces voltage drop (and incidentally line losses)

Reduced voltage drop means higher delivery voltages.

Utilities have a quite narrow band of acceptable voltages near nominal, and switched capacitor banks provide an inexpensive and efficient way to make a 'rough' adjustment on the voltage, and then voltage regulators can make 'fine' adjustments as necessary.

 

[SierraJ]

IEEE Standard 141-1993 (Red Book) provides the following aproximate formula to "show the importance of reducing the reactive component of current in order to reduce voltage drop":

/\V = R I Cos(Phi) +/- X I Sin(Phi)

/\V = (+) for voltage drop and (-) for voltage rise

Phi = Power factor angle (from 0 to 90 degrees)

The "PLUS" sign is used for lagging PF and the "MINUS" sign is used for leading PF.


Since the system reactance (X) is always larger than system resistance (R), this equation clearly shows how for lagging power factors, voltage differential will be positive (voltage drop) but for leading power factors, /\V will be negative (voltage rise).

Hope this helps.

 

[electricpete]

I apologize in advance for being disagreeable, but I believe the explanation by Sierra misses the mark. The impact of capacitance on voltage distribution goes far beyond the small effect of reducing current magnitude.

I believe the equation from SierraJ illuminates the important aspects.

First remember the series impedance is primarily inductive. Therefore the major contribution of voltage drop comes from the 2nd term +/- X I Sin(Phi)

Furthermore you see that this term is approximately zero for real power flow (Phi=0) and is maximum for reactive power flow (Phi = 90).

Once again this illustrates the importance of reactive power flow through series inductance in establishing the voltage profile.

=====================================
Eng-tips forums: The best place on the web for engineering discussions.

 

[electricpete]

Sorry, I meant to day that the post prior to Sierra missed the mark, Sierra hit the mark.

=====================================
Eng-tips forums: The best place on the web for engineering discussions.

 

[electricpete]

"meant to day" should have been "meant to say".

Now my corrections need corrections. I'm going to quite while I'm behind ;-)

=====================================
Eng-tips forums: The best place on the web for engineering discussions.

 

[itsmoked]

electricpete: Step away from the bottle... :0

Yes, that equation is quite illuminating.

Wow guys. I have learned much here! Quite a change in voltage control methodology from the computer systems I design. We don't add capacitors to vary voltage. Guess that's a neat/useful aspect of operating in a steadystate AC realm verse the switched DC. Also the fact that the systems are primarily inductive is also an eye opener to me.

Now then, the control box I looked into had an adjustable pointer pointing at a voltage scale like a Simpson meter. I take it this was more like 'if the voltage drops below this pointer position turn on this capacitor bank'?

Thanks again everyone.

 

[mc5w]

If the power factor at a generator goes past about 90% of 95% capacitive there will not be enough field current to carry the torque from the prime mover.

That said, power factor correction capacitors need to be disconnected at night maybe even during lunch breaks. Capacitors put very severe stress on switching devices and the capacitor switching rating is a completely different animal from switching inductances.

The underground cables in residential neighborhoods also contribute capacitance. In a place like New York City where air conditioning and elevator motors run incessantly this is not a problem.

The reason why utilites do not mind to much is a residence has 1 or 2 low power factor security lights is that they do need inductive loads at night.

 

[alehman]

Most utilities don't penalize unless you are below a certain power factor. If your utility penalizes for anything < 1.0, theoretically you want to stay as close to 1.0 as possible.

If you don't have generators or significant harmonic producing loads, you can set your controller to maintain 1.0 (if you have enough capacitance) without fear of damage or disruption. As stated above, leading power factor can cause difficulties for generators. Also, capacitors create resonances that can cause problems if there are harmonics on your system. In that case you would want to avoid capacitance values that create resonances corresponding to the harmonics present.

 

[Skogsgurra]

Good to see misunderstandings cleared up and myths busted in these fora.

Gunnar Englund

http://www.gke.org