Reactance in 3 phase motors running with pf correction capacitors connected and leading

[squeeky]

I have a problem explaining the problems of running three phase motors with capacitors on or near the connections and at a leading power factor. I have read it here but I lack the comprehension to explain it to someone else. Please can someone assist with the problems with this type of running phylosophy. It revolves around installing capacitors and not worrying when motors go low or no load and / or the installation goes low load. Thank you.

 

[electricpete]

One aspect is if that if the caps remain connected to motor when it is disconnected from the powersupply, overvoltage may occur at motor terminals. To my simple thinking it is because resonance frequency of the LC circuit is excited by the decaying voltage that the rotor induces into stator.

Simple circuit for modeling the motor is magnetizing reactance Xm in parallel with load branch Rload.

Now connect Xc in parallel and focus on the Xc and Xm elements.

If it is overexcited, that means Xc < Xm at power frequency Fp

1 / (2*pi*Fp * C) < 2*PI*Fp*L
1 / (2*pi*Fp)^2 < L*C
1 / (2*pi*Fp) < sqrt(L*C)
(2*pi*Fp ) > 1/ sqrt(L*C)
We notice the right hand side is the LC resonant frequency. So overexcited system implies that at power frequency the LC circuit is being excited at a frequency above its resonanct frequency. When power is disconnected and motor coasts, actual frequency drifts down below Fp toward resonant frequency. As it passes through resonant frequency the magnitude will increase. Saturation limits the magnitude of the overvoltage but I gather it can still be on the order of twice normal voltage.


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

 

[waross]

I have a number of times double corrected motors to put KVARHrs 'in the bank' to improve the monthly average PF.
That is a motor that will require 5KVAR to correct to unity at no load has 10 KVAR connected. I have never had a problem.
This was part of the 'Old School' art of power factor correction, when power factor controllers were too expensive for many customers and PF and PF penalties were calculated monthly based on the KWHr and KVARHr consumption.

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

 

[electricpete]

Bill - were those caps connected downstream of the motor switching device or upstream?

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

 

[squeeky]

Hi. Thanks for the responses so far.

The capacitor is connected downstream of the contactor and the overload device. I have convinced them to set the overload correctly but am now trying to convince them that if the motor goes to light load or no load there is a problem. On Waross's reply is what I advised them to do, and that is, to correct only to the no load power factor to prevent a leading power factor at low load. I have heard it explained that there can be pole slipping and the chance of the resonant frequency causing massive current at the motor / capacitors that is not seen by the supply if it goes leading.

What Electricpete describes I've never seen before and will have to digest it and write it down on paper. I'm quite stupid. Hard work sees me through.

There is a debate on the issue of startup current when the contactor is closed. Does it draw more, same or less? I have said it draws more. I am building an irrigation Motor Assembly and will get a chance to test the installation as described on a 45kW, 4 pole, 400v motor while I have a power quality analyser connected. I hope to save the results and make the results known to anyone who wants them.

 

[ScottyUK]

Transient currents which exist until the system settles into steady-state operation will be hard on the contactor. The reduction in steady-state load current which results from the cancellation of the inductive vars by the capacitor doesn't apply during the transient phase when the motor's magneitc field is not fully established. The larger the capacitor the worse things are from the contactor's perspective. IEC contactors for capacitor duty are categorised as AC-6, which is fairly severe.

The following document might be of interest:

http://www05.abb.com/global/scot/sc...1140C0203 ContactorsCapacitorSwitching_BR.pdf

 

[electricpete]

I have convinced them to set the overload correctly but am now trying to convince them that if the motor goes to light load or no load there is a problem.

To my thinking, no problem unless and until the motor switches off. The only place that “no-load” comes into the picture is calculating a limit for correction (90% of no-load vars). It’s conservative because no-load vars are lower than full load vars. What problem do you envision at light load?

On Waross's reply is what I advised them to do, and that is, to correct only to the no load power factor to prevent a leading power factor at low load.

That’s not how I read Bill’s comment. He said he has overcorrected without problem. But I may be misunderstanding something.

I have heard it explained that there can be pole slipping…

Pole slipping applies to sync motors, not to induction motors. I assume here we’re talking about induction motors unless stated otherwise.

… and the chance of the resonant frequency causing massive current at the motor / capacitors that is not seen by the supply if it goes leading.

Haven’t heard of that.

Fwiw, this link touches on a lot of the things we’ve been talking about

https://extranet.w3.siemens.com/us/...ocs_MV/TechTopics/ANSI_MV_TechTopics20_EN.pdf

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

 

[electricpete]

Sorry, last post was garbled. Let me try again. Also added one thought in bold at the end.

I have convinced them to set the overload correctly but am now trying to convince them that if the motor goes to light load or no load there is a problem.

To my thinking, no problem unless and until the motor switches off. The only place that “no-load” comes into the picture is calculating a limit for correction (90% of no-load vars). It’s conservative because no-load vars are lower than full load vars. What problem do you envision at light load?

On Waross's reply is what I advised them to do, and that is, to correct only to the no load power factor to prevent a leading power factor at low load.

That’s not how I read Bill’s comment. He said he has overcorrected without problem. But I may be misunderstanding something.

I have heard it explained that there can be pole slipping…

Pole slipping applies to sync motors, not to induction motors. I assume here we’re talking about induction motors unless stated otherwise.

… and the chance of the resonant frequency causing massive current at the motor / capacitors that is not seen by the supply if it goes leading.

Haven’t heard of that. But it makes some sense that we do not preclude resonant condition if we don’t limit capacitance to keep the system resonant frequency above operating frequency

Fwiw, this link touches on a lot of the things we’ve been talking about

https://extranet.w3.siemens.com/us/...ocs_MV/TechTopics/ANSI_MV_TechTopics20_EN.pdf

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

 

[squeeky]

Thanks for the continued input.

Induction motor. The corrected power factor at the motor will cause the current being supplied from the starter (Contactor and overload)to be lower. As the motor goes into overload condition it is not seen by the overload as the current has been corrected. Motor burns out.

The transients and settling is where Electropete is coming with the formulas. Need to study them. Resonance and Q factor are a serious concern for me as specifics with everything going on are too complex for my tiny brain. In Dunlop, Bulawayo, Zimbabwe we had a 3.3kV, 2.0MW machine that destroyed the contactors and made the motor vibrate so much it ripped bolts on the foundation. This was stated as being a Q factor problem. The cap bank was right next to the motor as was the switchgear.

I agree with Scotty about the inrush current. On starting the motor draws more. When you energize a capacitor, there is an inrush current. Putting the inductance with the capacitor it should not cancel out, but should increase the current. Happy with the AC6 specification.

 

[squeeky]

Sorry. I took Warross's comment to mean that he had applied full correction to a motor and that it survived. It only becomes a problem on light load when the kVAr vector drops. In a normal installation, equipment runs flat out and light loads are uncommon. If you are running a light load and wish to correct at the motor, you should only correct for the no load to prevent a leading power factor. I'm trying to establish why. As stated pole slipping is quoted to be for synchronous machines only but the term is used when an induction motor is getting to the point of stalling. If you lose all the VArs there is no inductance. No inductance, no torque thus the falling back of the rotor speed compared to the staor speed thus the term pole slipping for an induction motor. I'm looking for confirmation and explaination on these factors. Thnaks again.

 

[squeeky]

me again. The Siemens document is perfect. Thank you electricpete. I quote.

In the preferred situation, the power factor correction capacitors are sized at or below 90 percent of the no-load kVAR
requirement of the motor. If the capacitors are too large, the motor can be subjected to self-excitation, which will result in
excessive voltages applied to the capacitors and motor. The capacitors are sized based on 90 percent of the no-load kVARrequirement because the manufacturing tolerance of the capacitors is –0 percent, +15 percent.

This is to prevent self excitation.

 

[waross]

My capacitors were connected at the motor terminals in one plant and on the contactor load terminals ahead of the O/L relay in other plants.
The old school method of sizing capacitors with no sophisticated testing equipment was with a clamp type ammeter.
At no load, there is very little difference between the reactive current and the load current.
We calculated the KVARs based on the no load current and then used the closest sized capacitor.
If we were "Banking" VARHrs for power factor correction, we would select capacitors based on double the no load current.
A favorite spot for over correcting in saw mills was the dry kiln fan motors. They typically ran 24/7 most of the time.
The motors typically stopped and reversed direction every 4 to 6 hours.
As for starting current with capacitors, it has been proposed that large capacitor banks be connected to the motor terminals to supply the reactive current required during starting. The capacitors are cut out in several steps as the motor accelerates. I have read papers on this starting method but I have never seen it first hand.
The peak currents through the inductance and the capacitance may not be at the same point on the wave and may be subtractive rather than additive.

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

 

[ScottyUK]

ePete,

That's a good reference library you've linked to. I hadn't seen it before, but for anyone who wants to take a look here is the link to the index to the Siemens MV Tech Topics library.