Sparks Electrical News August 2016

CONTRACTORS’ CORNER

17

POWER FACTOR NOT THE ONLY FACTOR B ob Freeman writes: I have just received my copy of Sparks Electrical News (June 2016). Here are my comments on your column: However, this calculation is easily done and then just to specify that the capacitor should run at 20% overvoltage safely. No problem.

Years ago, I commissioned a synchronous ventilation fan at a steel plant. In operation it was going to generate VARs. Of course there was a danger that the machine would pole slip so there was an out-of-step relay, which should trip had this happened. Gingerly, I reduced the excitation of the fan motor … down and down … until there was no excitation. Ha! The real power load on the machine was so little that nothing happened!

I have often told colleagues and equipment suppliers that I do not care about power factor! What matters is the contribution made by the reactive component to the maximum demand (MD) charges. But that’s the same thing, isn’t it? No, it is not! It’s a different mind-set! If the predetermined MD limit is not threatened, the prevailing power factor does not matter. Under those circumstances, switching capacitors in and out is unnecessary and just increases wear and tear on the equipment. Also, think of a thyristor rectifier conducting for half of each half cycle into a resistive load. The voltage and current are not in phase. What is the power factor? What is the reactive component? At the various projects in which I have been involved, the supply authority had installed Enermax meters. It was then possible to download from these meters several months of consumption history, kWh and kVAh, in half- hour intervals. A bit of Pythagoras then gives the kVARh. Having put these figures into a spreadsheet we would do an analysis – much more revealing than the billing history. For example, we could ask the plant operators, ‘What is it that you do each Thursday at 10 am?’ That operation could possibly be re-scheduled to a low tariff period or to a lesser demand period. The plant or factory may well run 24/7, but there will be non- continuous operations that happen at scheduled or random times within the billing period. The Enermax meters also provided pulses for kVAh consumed. This pulse was used to accumulate a counter in a PLC. Another counter, triggered at the required intervals, represented the target MD. Comparing the values of these two counters would show when the MD limit for that half hour was threatened. Then, action would be taken to reduce the reactive component or institute load shedding whichever was appropriate. Note: VSDs and controlled rectifiers generate harmonics and these may not significantly affect the MD, but they do increase the current in any installed capacitors. The capacitors and their fuses must allow for this unless harmonic filtering is used. If there are large machines such as air compressors driven by synchronous motors, which run continuously, tweaking the excitation can put them into lead. Terry Mackenzie-Hoy replies: Thank you for your email. Naturally we don’t want to switch in capacitors unnecessarily. I tell my engineers that if we had a person who could see into the future then all we would have to do is to wait until the person visualises an increase in demand, which will exceed the previous maximum demand in a month and switch in a capacitor. If you have a situation where there is a notified maximum demand with a penalty for going over it, then you have a fixed target. But if you really want to limit the demand to the best financial limit possible you have to have a fairly clever thing. But, I do disagree with you; many tariffs now are not based on maximum demand but rather on kVARh consumption. Under these conditions a power factor of unity is an obvious requirement. There are about 10 different types of thyristor rectifier all of which have different VAR generation. Again the pulse number of the rectifier also makes a difference. I used to work for a company that designed and supplied harmonics filters but always regarded them as something of a ‘fraud’: symmetrical and asymmetrical waveforms supply odd and even and odd harmonics. Obviously, if the source impedance of the supply and the capacitor resonated at the same place as one of the harmonics then ‘bang’.

NEW

Like us on

@brothersa

Visit www.brother.co.za to view our full product range

SPARKS ELECTRICAL NEWS

AUGUST 2016