Power Factor Capacitors and Reactive Current

[JoshL]

First, for convincing your boss, I would just point out that wattmeters take into acount power factor. If you have a portable wattmeter, you could quickly wire up a 1hp motor and show the difference between what you calculate with your amps and volts (VA) and what the wattmeter shows.

Second, is the cost of downtime less than the extra power cost? I sure doubt it. It would seem that your boss should be comparing the extra cost of power, the cost of keeping spare parts on hand, and the cost of downtime. I would think that the end result is to fix the cap bank and keep some spare parts on hand for the future. That is where the discussion would be in my plant. Keep running to keep the customers happy, get it fixed quickly, and then prevent the problem in the future.

 

[weressl]

or:
(3*2400V*27A)/1000*0.1$/kWhr*8000hrs

Where 2400V is the line-to-ground nominal voltage.

No sir, that is NOT a valid formula, because it does not follow the mathematical explanation how the power is calculated. (Even though numerically equivalent.)

 

[mivey]

No sir, that is NOT a valid formula, because it does not follow the mathematical explanation how the power is calculated. (Even though numerically equivalent.)

In what world does a per-phase analysis become invalid? Perhaps you would care to elaborate?

 

[mull982]

FYI We did order new fuses and are going to replace the capacitor. Now it is just a matter of me trying to prove my point and defend my origonal statement to my boss which we have all agreed is correct.

Thanks for all of the input.

 

[mivey]

FYI We did order new fuses and are going to replace the capacitor. Now it is just a matter of me trying to prove my point and defend my origonal statement to my boss which we have all agreed is correct.

Thanks for all of the input.

Now you have plenty of formulas to show him how wrong the production manager was and how right you were...maybe you will get a raise. The manager was probably already calculating his cut of the savings he was going to get in his next bonus check. :grin:

 

[Cold Fusion]

Part 1) No can do. The % is wrong as well as the formula. What is the basis for the 3-5% loss assumption?http://forums.mikeholt.com/showpost.php?p=827136&postcount=6

You're right, should have been ---- x 1.732 (just wasn't thinking )

Basis for the 3% - 5% loss is: 3% to 5% vd from service to end use is about as much as I have seen in industrial ckts. Vd X A = IR x A = I^2R. This is very similar to what you are advocating.

Since I used V(line-line), should have multiplied by sqr(3). (sloppy me: I mixed my I's and A's - but I suspect you got it.:smile: )

For a first approximation, I'm thinking this is a reasonable assumption.

so, changing to 27A reactive curent, puts the copper losses (first approximation) at:

for 5%losses and $0.10/kwh (4160 x 27 x 1.723 x .05 x 8736 x .10)/1000 = $8500/year

for 3% and $0.05/kwh ~ $2600/year

cf

 

[mivey]

You're right, should have been ---- x 1.732 (just wasn't thinking )

Basis for the 3% - 5% loss is: 3% to 5% vd from service to end use is about as much as I have seen in industrial ckts. Vd X A = IR x A = I^2R. This is very similar to what you are advocating.

Since I used V(line-line), should have multiplied by sqr(3). (sloppy me: I mixed my I's and A's - but I suspect you got it.:smile: )

For a first approximation, I'm thinking this is a reasonable assumption.

so, changing to 27A reactive curent, puts the copper losses (first approximation) at:

for 5%losses and $0.10/kwh (4160 x 27 x 1.723 x .05 x 8736 x .10)/1000 = $8500/year

for 3% and $0.05/kwh ~ $2600/year

cf

Partner, I think you are all over the place.

The approximate line-to-neutral voltage drop is obtained by:
IRcos@ + IXsin@
with the exact volt drop given by:
Vsource + IRcos@ + IXsin@ - sqrt[Vsource^2 - (IXcos@ - IRsin@)^2]

Power is R*I^2 or G*V^2

You are mixing the two together and getting an incorrect result.

 

[Cold Fusion]

I appoligize for being all over the place. I'll attribute it to my lack of experience.

The approximate line-to-neutral voltage drop is obtained by:
IRcos@ + IXsin@.

Excellent interpretation of Chapter 9 - meets what I think.

Power is R*I^2 or G*V^2

G must be "conductance"? Isn't that 1/R? My goodness you are confusing to a new poster like me.

You are mixing the two together and getting an incorrect result.

Huh? What two am I mixing? "R*I^2" and "G*V^2" Assuming G = 1/R and V = IR still hold in the Immutable Rules of God and Physics, those two are the same thing

Considering copper losses are resistive, wouldn't V x I = P, (where V is the voltage drop across the copper losses caused by the reactive current) also be true.

cf

 

[mivey]

I appoligize for being all over the place. I'll attribute it to my lack of experience.


Excellent interpretation of Chapter 9 - meets what I think.


G must be "conductance"? Isn't that 1/R? My goodness you are confusing to a new poster like me.


Huh? What two am I mixing? "R*I^2" and "G*V^2" Assuming G = 1/R and V = IR still hold in the Immutable Rules of God and Physics, those two are the same thing

Considering copper losses are resistive, wouldn't V x I = P, (where V is the voltage drop across the copper losses caused by the reactive current) also be true.

cf

I=V/R so P=R*I^2=R*(V/R)^2=R/R^2*V^2=1/R*V^2=G*V^2

Have you graduated yet?

 

[mull982]

Another question regarding these capacitors:

If I have a 450kvar pf capacitor at 4.16kV I am calculating that is will supply 62.5A of reactive power. What is the rule for sizing fuses for these capacitors? The cable coming off of the capacitor is #4 so it should be more than adequate for carrying this load.

I have another 150kVar capacitor which I calculate to supply 20A or reactive current. The fuse for this capacitor is only 18A but the cable is #8 which is plenty big. Could undersized fuses be what is causing some of my cap fuses to blow?

Also what determines what how much current is drawn from these caps? Do they always supply the their full calculated current, or is only a portion of this current used depending on other circuit paramenters?

 

[mull982]

After I posted I did a little investigation. I looked in the NEC handbook and found a couple of things. First off I noticed that I believed they had the equation incorrect for determining the current corrosponding to a capacitor.

It had I= (kVar * 1000)/1.73

I believe it should be (KVar/kV/1.73)

Also the exception in 460.8(B) states that you do not have to have overcurrent protection if the capacitor is located on the load side of the motor overloads which is the case in my application. Do I even need these fuses that keep blowing? With out fuses how does the motor overloads protect these capacitor cables from and overload or fault?

Lastly I noticed that these capacitors are usually manufactured with a 15% tolerence and and on top of that line voltage fluculation can cause an increase in current. With these two variables it can casue the current from a capacitor to sometimes be as high as 135% of it rated value. If I have fuses which are less than 135% of the Cap current could this be another reason why I am blowing fuses?

 

[weressl]

After I posted I did a little investigation. I looked in the NEC handbook and found a couple of things. First off I noticed that I believed they had the equation incorrect for determining the current corrosponding to a capacitor.

It had I= (kVar * 1000)/1.73

I believe it should be (KVar/kV/1.73)

Also the exception in 460.8(B) states that you do not have to have overcurrent protection if the capacitor is located on the load side of the motor overloads which is the case in my application. Do I even need these fuses that keep blowing? With out fuses how does the motor overloads protect these capacitor cables from and overload or fault?

Lastly I noticed that these capacitors are usually manufactured with a 15% tolerence and and on top of that line voltage fluculation can cause an increase in current. With these two variables it can casue the current from a capacitor to sometimes be as high as 135% of it rated value. If I have fuses which are less than 135% of the Cap current could this be another reason why I am blowing fuses?

It does not require overload protection because it is a fixed parameter device. The fuses are there for short circuit protection so when the capacitors fail they can be isolated from the motor and the motor can keep running. So oversize the connecting cable to the cap to 250%, fuse it to that level and fawgeteboutit.......

 

[mull982]

It does not require overload protection because it is a fixed parameter device. The fuses are there for short circuit protection so when the capacitors fail they can be isolated from the motor and the motor can keep running. So oversize the connecting cable to the cap to 250%, fuse it to that level and fawgeteboutit.......

So what you are saying is that since the capacitor is a fixed paramenter device it will always supply its rated current to the motor (except for voltage variations) Therefore a 450kVAR cap will always supply 62.5A ideally at all times except with voltage flucuations as mentioned.

You say the fuses are for short circuit protection. Can you explain the sentence that says when the caps fail they can be isolated from the motor and the motor can keep running?

Do I even need these fuses according to code? Why do you choose a level of 250%? My cables are already existing cables and are about 175% of cap rating. For short circuit protection I dont believe you have to select the OCPD according to the cable size correct?

 

[mull982]

Also, if the voltatge a the capacitor terminations drops wont the cap supply more current thus the possible need for overcurrent protection?

 

[Cold Fusion]

Also, if the voltatge a the capacitor terminations drops wont the cap supply more current thus the possible need for overcurrent protection?

I think what you are asking is: If the voltage goes down, will the current go up, holding the kvars the same?

No, the caps have a kvar rating at a specific voltage. If the applied voltage is lower than the rated voltage, the kvars also go down. The capacitance is constant, not the kvars.

cf

 

[Cold Fusion]

I think what you are asking is: If the voltage goes down, will the current go up, holding the kvars the same?

If the applied voltage is lower than the rated voltage, the kvars also go down.

Missed the edit time.

The kvar change is porportional to the square of the voltage change.

KVAR(new)/KVAR(rated) = V(new)^2/V(rated)^2

cf

 

[mull982]

Ok I see. That equation makes sense. So if the voltage goes up the opposite would be true, and therefore the kVAr's would go up and the cap could possibly supply more current thus blowing my fuses. Could voltage spikes be attributing to an increased current blowing my fuses?

Also I am still trying to determine exactly why these fuses are necessary in my arrangment, and what they should be properly sized at.

 

[wirenut1980]

If the caps are always blowing the fuse right away upon energization that usually means a bad cap or the fuse size was replaced with the wrong size. If the fuses seem to be blowing intermittently without ryhme or reason, then it could be something to do with the voltage RMS levels or harmonic resonance.

 

[mull982]

I was able to get some fuse information from the manufacturer for selecting the proper size fuses for these capacitors (GE capacitors)

Currently we have only been fusing the A & C legs of the fuse and the capacitor is connected in an ungrounded delta fashion. The engineer I spoke with told me that since my system is Y with a resistance ground I needed to fuse all three phases. He gave me a small explanation but it wasn't very good. Can someone explain to me why I would only need two fuses if I had an ungrounded delta system but need all three with my Y grounded system?

Is it because with an ungrounded delta system there could be no potential for a ground fault, and with a grounded Y or Delta system there can be a ground fault thus the need to fuse all three phases?

 

[weressl]

I was able to get some fuse information from the manufacturer for selecting the proper size fuses for these capacitors (GE capacitors)

Currently we have only been fusing the A & C legs of the fuse and the capacitor is connected in an ungrounded delta fashion. The engineer I spoke with told me that since my system is Y with a resistance ground I needed to fuse all three phases. He gave me a small explanation but it wasn't very good. Can someone explain to me why I would only need two fuses if I had an ungrounded delta system but need all three with my Y grounded system?

Is it because with an ungrounded delta system there could be no potential for a ground fault, and with a grounded Y or Delta system there can be a ground fault thus the need to fuse all three phases?

You are only providing short circuit protection.

Even with the delta you do have a potential for ground fault, except the current of the ground fault would not be sufficient to open the fuse. In the case of phase-to-pase fault one of the faulty phase would be fused, so it would open. The remaining fused phase would still be ready to fault through the unfused phase. Delta systems should have a ground fault monitoring system where the ground fault can be located and isolated at the first oppurtunate time.

It also used to be common to provide overload protection in only two phases, but it was bask in the days.