Power Factor Load Balancing

[gatormc9]

I am designing a system and I need to balance the load across a 208 3 phase input. I am balancing a bunch of equipment, each having a different power factor. When trying to balance the power factors on each phase, should I just average the PF or is it more complicated than that? Balancing just the loads is easy, but I think the PF needs to be close on each leg also. The answer is probably simple and I will most likely kick myself when I hear it, but I am still new to all this (recently graduated). Thanks for the help ahead of time.

 

[jghrist]

I'd set up a spreadsheet with the resistive and reactive component of each load calculated. Then try to shift loads around to balance both the resistive and reactive components on each phase. If you use Excel with the Analysis Toolpack add-in, you might be able to use Solver to reach an optimum solution.

 

[peebee]

Unless you are dealing with some really bad PF loads, I wouldn't worry too much about it. What kind of loads do you have (resistive, motor, fluorescent lighting, computer, etc).

 

[DanDel]

Warthog302, to expand a little on the explanation by jghrist, you need to balance the reactive loads, not the PF. The PF is just a percent, and tells you nothing about the actual size of the load connected. In other words, if you have two loads, one has 0.8PF and the other has 0.9PF, you can't simply add them together and divide by two and get a PF of 0.85. The first PF could represent a load of 1000kVA and the second could be 10kVA. I'm sure you can do the math from here.

 

[busbar]

A goal of load-balance calculations is to make the system efficient, and to “red flag” potential overload problems. If there us more than one load device and a degree of cycling during normal operation, per-phase power factors will likely vary but have little undesirable effect.

 

[electricuwe]

If you consider to take the PF into consideration for laod balancing you should be aware of the fact that there is a displacement power factor cosphi and the true power factor lamba. The way suggested by jghrist will only work if the currents are sinusoidal (no harmonic currents).

Give more details about your application and the kind of loads. With the information already provided no one can decide if it is useful to consider PF or not and how to do it.

 

[gatormc9]

I have determined that I don't need to take the PF into account when balancing the load. The load will be mostly electronic equipment and RF stuff. It turns out that I probably won't be getting accurate power factor numbers to work with anyway. Thanks for your help.

 

[jbartos]

Suggestion: The power factor is a dynamic variable, in fact the more dynamic variable than a load variable. Therefore, an automatic power factor compensator will be needed, which is not exactly a simple device.