PF correction capacitor for VFD run motor 1

[edison123]

If a VFD has a sinusoidal output filter and if the V/Hz ratio is maintained, can we connect a PF correction capacitor across the motor terminal fed by this VFD? Will this reduce the VFD output current (since motor no load current is compensated by the capacitor current)?

Muthu

http://www.edison.co.in

 

[itsmoked]

since motor no load current is compensated by the capacitor current

Why is this true?

Keith Cress
kcress -

http://www.flaminsystems.com

 

[LionelHutz]

Why would you want to do that? It will have no effect on the power draw of the VFD.

 

[waross]

There will be an initial cost saving if you reduce the size of the motor conductors in a long run.
Or with the same size conductor and the capacitors connected near the motor there will be less I²R losses.

Bill
--------------------
"Why not the best?"
Jimmy Carter

 

[LionelHutz]

I'm pretty sure code doesn't allow you to reduce conductor size on a motor circuit. They must be matched to the motor FLA.

Reducing I^2R losses sounds great in theory but spending money to reduce I^2R losses is a fools errand.

I would rather hear the OP's actual reason instead of speculation.

 

[ScottyUK]

Keith -

At no-load the line current is almost all magnetising the core at a very low lagging power factor. An appropriately-chosen capacitor should be able to more-or-less compensate the current so only the in-phase component remains. With a VFD this plan may be complicated by the presence of any switching hash on the output which will see the cap as a virtual short-circuit, although a small series reactor or HF choke would calm that down.

Lionel - given Muthu's line of business I'm guessing that the intention is to allow no-load testing of motors much larger than the VFD is designed to handle. I think the biggest problem will be that the capacitive reactance rises as frequency falls, so the magnetising current at low frequency will still be high.

 

[itsmoked]

Thanks Scotty.

Keith Cress
kcress -

http://www.flaminsystems.com

 

[electricpete]

Interesting concept to correct a vfd.
If high frequency content is filtered out, the concern about overheating the caps goes away.

Let’s define Correction Fraction CF = capacitive correction current / inductive current drawn by motor
CF = [V * (j*w*C)] / [V / j*w*Lm ]

Let’s further assume V is varied in a constant volt/hz fashion V = Vbase*(w/wbase)
CF = [Vbase*(w/wbase) * (j*w*C)] / [Vbase*(w/wbase) / j*w*Lm ]
CF = w^2 * C *Lm
(above I assumed no-load operation.... inductive current drawn by the motor is higher under load, and things would be even trickier if load is changing with speed).


Prudency dictates CF < 1 at highest w (call it wmax). This is traditionally referred to as avoiding "overcorrection". I would say it is needed to avoid passing through an LC resonance at some w below the wmax.

If you decrease w to 50% of wmax, then CF goes below 25% (if CF was 80% at wmax, then CF would be 20% at w=0.5*wmax). In other words the caps have limited effectiveness in cancelling the inductive current at lower frequency w (unless some provisions are made for switching the capacitance as speed changes).

At least that's my initial take, please double check my algebra.

=====================================
(2B)+(2B)' ?

 

[jraef]

I think that the effect would be the destruction of the PFC caps, unless they are put in series with an inductor (as they are in a dV/dt filter), in which case you add a voltage drop and risk the loss of maximum torque capability in the motor. And for what? There is no real point in using PFC caps on the output of a VFD, the VFD will present a .95 displacement PF to the line anyway, so what's happening below the VFD is irrelevant. As to it POSSIBLY allowing for the reduction of conductor sizing, that is only theoretical because in most cases the minimum conductor size is dictated by installation codes anyway. In addition, the capacitor charging current would have to come FROM the VFD output, so the VFD would have to be over sized to allow for that, and any theoretical conductor cost savings would likely go to that. But again, it's a moot point, the capacitors alone would not survive.


" We are all here on earth to help others; what on earth the others are here for I don't know." -- W. H. Auden

 

[edison123]

Thanks, guys. As usual, Scotty nailed it. My 'plan' was to use the VFD for no-load testing of motors much larger than the VFD is designed to handle by balancing out the mostly inductive no-load current with almost equal capacitive current. Sorry I didn't make this clear in my OP.

But as scotty and epete pointed out, the caps current, being proportionate to the frequency (and voltage), is going to do jacksquat in canceling out the constant motor inductive no-load current right up to the rated speed/frequency.

jeff - If the VFD output has a sinusoidal filter, why would it destroy the PFC capacitor as long as the frequency does not exceed the rated frequency of the capacitor? I get that the front end displacement PF is close to 1. I wanted to see if I can get that near unity PF at the back end, which is now clear is not possible in the run up.

Muthu

http://www.edison.co.in

 

[jraef]

I missed the part about it having an output sine wave filter, sorry. It's there, my brain apparently didn't see it until just now when I went back and read it again...


" We are all here on earth to help others; what on earth the others are here for I don't know." -- W. H. Auden