Power Factor Correction and Harmonic Resonance

[timm33333]

I have a question about power factor and resonance. When all motors are without VFD’s (variable frequency drives) then the power factor is lagging, and we use capacitors to improve power factor. When some motors are without VFD’s and some motors are with VFD’s, the VFD’s would improve some power factor but it would not be enough if the requirement of utility is that the power factor has to be close to 100%.

Also when VFD’s are present, there would be harmonics, and so filters should be used (instead of capacitors) for power factor correction. Because if we use capacitors, then there might be the problem of resonance because the capacitance of the copastors might resonate with the inductance of the source.

My question is that if we have a system which has some motors without VFD’s and some motors with VFD’s, then can we still use capacitors (instead of filters) to correct the power factor? Thanks.

 

[Gr8blu]

Timm33333: The answer to your question is yes, you can use PF correction capacitors (presumably in a switched bank format) to bring the system power factor back toward unity. However, since some equipment uses a drive and some does not, the SAME bank format cannot be used. Essentially, the bank design has to ensure the system does not over-correct at any given load point.

The other thing to note is to be very sure there are no “local” capacitive elements (either PF caps or some types of lightning arrestor) at the terminations of the motor being operated on the drive. This is because most drives have trouble with overly-capacitive “loads”.

Converting energy to motion for more than half a century

 

[waross]

Induction motors, = displacement power factor. Use capacitors.
VFD drives = distortion power factor (with many drives). Use filters.
Motors may have capacitors switched with the motor.
The capacitors may be connected at the motor, on the load side of the overload relay or at the line side of the overload relay.
In some plants, when a production line is comprised of one or two large motors and a number of smaller motors, you may over correct a large motor on the assumption that when it is running there will also be a number of smaller motors running.
If the penalties start at 90% PF, often correcting the large motors to 100% PF will bring the plant up to the 90% cut-off.
If the penalty tariffs are more stringent, the large motors may be over corrected. When motors are over corrected, the capacitors should be connected on the line side of the overload relay so that the capacitor current does not flow through the overload elements. This is a code requirement in more than one country.
This approach requires experience and judgment and may not be for every plant.


--------------------
Ohm's law
Not just a good idea;
It's the LAW!

 

[timm33333]

Thanks. In our case the capacitor-banks are at the main incoming utility switchgear (not at the motor terminals), so the capacitor current flowing through overload elements is not an issue. It makes sense to use the capacitors when some motors are without VFD and some motors are with VFD.

So can resonance problem still occur if we have only capacitors (no filters), or do we not have to worry about resonance in this case?

 

[LionelHutz]

You generally need filtered capacitors when the system has a mix of VFD and non-VFD motors.

Then, the VFD's need harmonic filtering to correct their true power factor.

 

[waross]

It was before my time in the industry when power factor first became an issue, but I have read the warnings in old literature.
The first mitigation was large banks of capacitors directly connected to the plant system. (Power factor controllers were still in the future.)
When the plant was shut down at night or during the weekend, the capacitors would become a series series inductive-capacative circuit.
The voltage across the capacitors would rise to a level that would cause rapid burnout of lamps.
The next step in PF correction as a work in progress was switching the capacitors with the motors.
How much correction was needed?
Early tariffs calculated PF penalties from the monthly ratio of KVARHrs/KVAHrs.
When faced with a power correction assignment, the first step would be to look at at least a year's worth of old power bills and calculate the number of leading KVAR hours needed to raise the PF to 90%.
The most effective source of KVARHrs is capacitors connected 24/7.
The second step was to estimate the number of KVARHrs that could be safely connected 24/7. A safe starting point was enough KVARs to offset the exciting current of any transformers. (Actually, we would cheat and use the no-load current in our calculations. That was close enough.)
It was a judgment call how much more capacity we could add.
Then the largest motor would be corrected, or over corrected. The number of running hours per month and the number of expected KVARHrs per month were noted.
Motors that ran more than shift hours were favoured, motors that ran infrequently may be passed over.
These steps were repeated until enough KVARHrs per month were accumulated to correct the PF to 90%, 11 or, more capacitors would be added to correct for 12 months out of 12 months.
Today, just install a PF controller.
However you may find it useful to follow some of these steps when sizing the PF controller and determining the number of steps.
Before you start, check the local tariffs.
Some tariffs may not use a monthly average but base the penalties on KVARHrs used past a fairly stringent limit.
You may find that you must correct the pF to above 98% or 99% at all times. You must have a PF controller to eliminate all penalties under these tariffs, however you may balance the cost of full correction against the cost of modest penalties and go for less steps on the PF controller.
That still leaves distortion PF. I will leave that discussion to others.


--------------------
Ohm's law
Not just a good idea;
It's the LAW!

 

[timm33333]

6-pulse drives produce harmonics of order 5, 7, etc. Filters should be used (instead of capacitors) to improve power factor when harmonics are present, but still sometimes capacitors have to be used for cost considerations. If the system has 6-pulse drives and the resonance occurs at 7th order harmonic, the THD and TDD might exceed the acceptable limits. However if resonance occurs at 7th harmonic but still THD and TDD are within acceptable limits, will resonance still be a concern in this case or not?