Power factor - getting full power from a generator and maximizing efficiency

[jzeeff]

This is a summary of various inputs on power factor.

Low power factor is caused by current flowing back and forth from source to load but not being consumed
Power factor is easily measured with a "Kill-A-Watt".
A low PF load increases copper losses proportionally  - ie, reduces efficiency
A low PF caused by inductive motors can be corrected with a capacitor
A non-linear or harmonic low PF caused by electronic loads (for example, things with rectifiers) needs a tuned inductor/capacitor circuit to improve it
Recent electronics may have a PF very close to 1
kVA accounts for power factor but kW doesn't.
Generator heads should be sized for kVA but  engines should be sized for kW

It is not clear to me how to buy or build the tuned circuit (called passive power factor correction).  Perhaps a properly sized inductor in series and capacitor in parallel with the load.  Or just use a ferroresonant transformer.

It is not clear how to measure power factor for very short durations - for example, during the startup period of a motor.

[Doug]

First you need to understand reactive aparent and real power.

Best find a good book and do some research. If you are generating your own power I would look at doing some correction where needed ( like electric motors for the was machine ect ) but most things are less critical and you need to deciided how much if any amount of correction is worth the trouble.

[oliver90owner]

ie, a PF of .7 means your 5kw generator is only good for 3.5kw
 
Generator outputs will always include the power factor with the kVA rating.  kW rating will be at unity power factor.

Most generators quote kVA at 0.8 power factor.

The odd motor run intermittently is not worth worrying about - only worry about it if it is continuous.  Starting it is the biggest problem.

The leccy supply companies were a bit picky, while running a couple of 4 MW and several 1 or 2MW machines (together or separately).  They were always run at 0.95 lag IIRC.

Regards, RAB

[mobile_bob]

you find a good, definitive, and correct answer to that last one (power factor with the use of rectifier loads)
and i will pay you 10 bucks!

seems there are lots of folks that understand powerfactor to various degree's, even electrical engineers argue
about power factor in its most common appications, but

try to find two electrical engineers that agree on the merits of power factor with the use of a rectifier...

good luck, i ain't found it yet,, and i have been looking high and low.

seems i will be forced to get it illustrated on a scope, and then work with various methods to alter the powerfactor
from what i think i understand it is a very difficult beast to get a grasp on and make any meaningful correction.

even after all that is done, the gains in efficiency are very low, you basically are increasing output only by the amount
of heat reduction in the stator,,, that likely is very little (maybe a percent or two at full capacity)

good question

bob g

[jzeeff]

Ok, I got it on the kW vs kVA issue wrt generators - the engine is sized for your expected kW and the gen head is sized for the expected kVA.  If you expect to run reactive loads, it might make sense to run a 3kVA gen head on a 4 HP engine.  If you expect only resistive loads,  then a 2kVA head on a 4HP engine would be a closer match.

[Doug]

One thing I would like to add to Bob's statement ( well 2 ) :

The slight lagbetween when a diode is biased and the voltage across it creates the  PF issue ( even with a reisitive load ) now if you add a DC motor to the load that will draw reactive power from the genertor unless there is a free wheeling diode there to deal with the inductive nature of the motor.

Next cosnideration is gating with SCRs the effect of  gating lags the current behind the voltage and creates harmonics. Thats on of the reasosn cycloconverters ( and to a lesser extent current fed inverters  with dc links ) fell out of favour and. The modern way of doing this is full rectification ( for high PF with a large capacitor bank to filter and hold the charge in the link ) and PWM to re invert the dc to AC.

[oliver90owner]

jzeeff ,

kwh vs kva

Sorry to be picky but kWh is an amount of energy9measured in Joules).  kW is a rate of doing work, or power, transferring energy from one form to another (measured in Joules per second).

In the SI system of units a small 'k' means 1000 or 10³.  Don't know why, as it is one of the few multiples which happens to be lower case.  Most multiples are upper case with the reciprocal values (smaller than one) being lower case.  Most things make sense in the SI system but there are a few bits that seem to buck an otherwise neat system)

Anyway back to your post, 3kVA at 0.7 pf would be equivalent to 2.1kW (so about 2HP per kW) and for resistive loads it is our general rule of 2HP per kW (so 2kW head on a 4HP).  I reckon you knew what you were posting but wrote it wrong.  Some reading the post might be baffled and possibly misled.

You need to always quote the power factor with kVA or it means nothing.  What it boils down to is kVA rating is kW rating at unity power factor, so most quote resistive power (kW) and leave it to the motor users to work out their kVA's.

Hope this helps to clarify kVA and kW.

Regards, RAB

[jzeeff]

Fixed, thanks.

My understanding is that it is not uncommon for motors to have a very low PF (.3 for example)  when they are starting.

[Ian]

Hopefully, this is on topic, and if any of you guys can offer a reply - may help me and others too....

My ST 7.5 kW 3 phase generator feeds to a full wave rectifier and the resulting DC is fed to a grid tie inverter. Just to give the full picture, the ST rotates at about 1500 RPM and outputs around 380 volts RMS phase to phase and is held purposely this low by an AVR. The DC measured under load at the inverter is about 450 - 480 volts DC when the inverter is outputting around 3.5 kW single phase into the grid.

I was wondering if the generator was seeing a resistive load (from the full wave rectification to DC) or a capacitive load (capicitors in the inverter), or an inductive load (inverter transformer) but could not find anyone who could give a definitive answer. So I installed a 3 phase kW hour meter between the generator and the full wave rectifier which has pulse indication of "real" power and pulse indication of "reactive" power. The meter shows one thousand pulses per kW hour on each of the "real" and "reactive" LEDs.

In operation, the "real" power LED pulses at about one pulse per second (as you would expect for a grid output of about 3.5 kW) and the "reactive" LED pulses about once per 6 seconds. In other words, the "real" LED pulses six times for each of the "reactive" LED pulses.

I have assumed this means I have a power factor of about 0.8 (5/6ths of the power is "real" and 1/6th or the power is "reactive". Is this a correct assumption ?

Can I tell from this information if the load seen by the generator is capacitive or inductive ?

Many thanks for any help in shedding some light !

Regards,
Ian

[jzeeff]

I don't know how your meter works, but it is almost certain that the rectifiers/inverter creates a non-linear harmonic load.

[mobile_bob]

then just when you think you have it all figured out...

then you get a reality check in the difference between "diodes" and "diodes"

not all diodes are created equal, even if the amp and voltage ratings are similar

horribly expensive lessons are learned, when dealing with alternator fed rectifiers.
not all 3 phase rectifier bridges work the same, you got avalanche, non avalanche, you got frequency concerns
you got distortion concerns, you got standard recovery, fast recovery, ultrafast recovery,,, and other stuff
and they all look like they should do the job???  ya  right!!

don't ask me how i know!
cuz i ain't tellin!

bob g

[Ian]

jzeeff  - I assume the meter analyses the phase angle between the voltage and current and integrates the results (but I do not KNOW this for sure). The meter is non-moving-parts digital.

I agree that the load will not be such that the generator produces a perfect sine wave but I do not know about harmonics. As my goal is essentially to produce DC, the generator AC output is almost of superficial interest as long as the generator is happy to handle the load it sees.
 Having said that, I sure wish I understood more about what the generator really sees. I am so in the dark, I do not know if the load is essentially resistive, leading, or lagging!

Is a non-linear harmonic load anything I should seriously worry about ? (Your post would lead me think that I SHOULD be worried...)

mobile_bob - Yes, agreed. The bridge rectifiers I am using are rated at 1000 volts reverse voltage and 16 amps continuous. they are bolted to a 0.3 degree C/W heatsink.

I think most rectifiers fail open circuit but in the event of a closed circuit failure, AC will pass through and then a reverse polarity diode will allow this current to effectively short circuit - which should trip the 16 amp three phase breaker (hopefully before any catastrophic damge is done to the capacitors of the inlet stages of the inverter due to reverse polarity).

I ran this setup for the whole of the last heating season and added around 1000 hours without any obvious issues that I can see. The previous heating season, I was operating on single phase 240 volts without any obvious issues either. But, of course,  it does not mean that there ARE NO issues just because I am ignorant and cannot see them....

Regards,
Ian

[Wizard]

Diodes always fail shorted.  Sometime after it shorts out, body of diode smokes up or explosively if current is high if there is not a current limiting devices like fuse is present.

Cheers, Wizard

[jzeeff]

I wouldn't worry about a non-linear harmonic  load if things are working well.      Maybe if I were trying to get the last little bit of efficiency or output from a system.

> I do not know if the load is essentially resistive, leading, or lagging!

Probably none of these.  So the fix isn't easy.

[mobile_bob]

not all diodes fail to short, and those that do cause all sorts of other problems in my experience.

when a 3 phase alternator is humping out high currents at moderate voltages and the rect bridge goes
into avalanche,, the alternator makes an excellent brake and the exhaust turns very black in a hurry.

for those that are working on 50/60hz luckily this is not an issue

for those of use working on 500/600hz, it is a huge issue.

as is the harmonic content, and distortion power factor, both of which can push a bridge into the avalanche
region which effectively makes for a short passing high AC currents onto your DC buss.

ugly things happen when that happens!

ugly=costly=embarrassingly expensive=mad as hell=oh man!!

the cost of an education i suppose, still cheaper than one college credit and likely something that one will remember
much longer and have broader application as well.

i know this is the case with me!

so i want to know about pf of an alternator driving a rectifier, specifically at 500-600hz
also want to know about harmonics, distortion power factor and how to eliminate them!

bob g