Okay guys here one for you.
Dont ask why BUT if you wanted to install a capacitor for power factor correction and the capacitors are a different voltage then the supplied voltage, would the capacitors have the same effect? Example
480volt capacitor and a 600volt system
Installing a 600volt pri and a 480sec then
installing the capacitors on the secondary, would that in your option re-correct the power factor?
I am thinking not-the winding are electrically isolated from each other.....any one esle?

__________________
MIKE
SHAKE THE TREE

That would depend greatly on what quality correction is needed, capacity of the capacitor, quality of the transformer, etc.

I have never installed less than 6 capacitors for corrections, but they were usually engineered for the application by a licensed engineer.

__________________

N.T.K.B.O.

In theory what you propose would work because the transformer will step the voltage down as the capacitor stores energy and step the voltage up as the capacitor returns energy to the system. In reality, it's more complicated because the inductive, resistive, and phase shift characteristics of the transformer must be factored into the power factor solution.

Just remember, if you're trying to correct for excessive non-linear loading (multitudes of AC to DC power supplies), a capacitor may ADD to the "flat topping" problem associated with this.

(Capacitor charges from the peak of the sine wave, and discharges during the "RMS" span, that is why a capacitor 'bank' is used, to help correct "flat topping")

__________________

N.T.K.B.O.

Sorry guys.

I read this when you replied, but didnt post. Thanks for the scoop.

__________________
MIKE
SHAKE THE TREE

The reactive power supplied to shunt-connected capacitors is given by Q = (V^2)/Xc

Thus, reactive compensation from a capacitor varies with the square of the voltage across it. This is why cap banks are typically delta-connected. A capacitor connected in delta will provide 3 times the reactive power that a similar cap would, if connected line-neutral.

Edit: I'd like to stress what NTBKO said about having a qualified engineer involved in the selection/installation of cap banks. You don't want the system resonant point to coincide with one of the pfc stages!

Hey Rooster
That is funny that you posted on this. I just saw this type of installation at a factory, this past week... The unit was installed in the ceilling, I hope to get a better look at it but it was just was I was asking. The supply voltage from the disconnect to the trany then to the bank. Still not sure if it would work or is still working but its out there.

__________________
MIKE
SHAKE THE TREE

You know, I just noticed that I did not actually answer your question.

The fact is that the effect of the capacitors will be transferred through the transformer and onto the rest of the system. The only reason I could think of, for such an installation, is that the 480V caps were cheaper, or more available (or some other reason to make them more desirable) than 600V rated caps. Or, possibly the leakage impedance of the transformer would reduce switching transients........in any case I'd have to see the installation data to begin to guess.

What size was the PFC bank? was there any other filtering/any other "interesting" equipment installed there?

I just thought of something else; that likely was not a transformer! PFC systems are typically "de-tuned" to shift the resonant point of the system "away" from the typical harmonics: 3rd, 5th, 7th, etc. That likely was a three-phase reactor -- and not a transformer. The inductance of the reactor and the capacitance of the bank will determine the system's resonant point, and fix it at a non-problematic frequency.

Cool, I forgot about that, I didn't know you could do that until I ran across a unit for sale that said they tie a reactor and a capacitor into one unit. I will tey and check it out this week, but it was big to be a reactor but you never know. Rooster-got anything else about capacitors?

__________________
MIKE
SHAKE THE TREE