Power factor correction for 3 phase motors.

PFC came up in another thread to do with drones, i havnt found much info on how to do this. Same kind of thing but this time a different power source, anyway what i am looking for is a start point to research how to do this with 3 phase BLDC motors as generators, they are non hall switch types so the back emf technique is used for zero cross point.

Its actually for a mock up size methane powered steam engine power station, the steam engine drives the BLC which acts as the power station. I might even use the 3 pole microwave oven table motors, simply because they knock out 240V at low revs. But whatever is used i still have to find a solution to pfc. Pommie mentioned taking the power at the top overheats the generator, so am i right in thinking the main technique is working out when to switch in the take off feed? according to most info the commutation when acting as a motor is around 30 degrees after zero cross, so in my simplified universe 30 degrees should work when generating and taking the power off the generator?

Yeah i made that sound all jumbled sorry.

There is no power factor correction with a _generator_ (or rather an alternator, for AC), regardless of how its constructed.

With no load, the alternator produces voltage - but there is no current. No current means no current phase relationship to measure.

When you connect loads to it, the phasing of the load current is the sum of all the power factors of the loads; some may lead and some may lag.
It's the loads that needs correction (if not resistive), not the alternator.


If you are meaning you want to rectify the alternator output efficiently, then a switch mode PSU with power factor correction would presumably be suitable.
However at the scale of the microwave oven motor, it's efficiency gains is likely zero or negligible.

Another approach would be in effect use a three phase PWM inverter circuit, as you would use to drive the motor, but continuously running in peak "regenerative braking" mode.

If I remember right, that's the point where the back EMF of the motor would be half that of the speed you are "braking" from; ie. if you have 240V back emf (output) you set the pwm to the percentage it would be at when driving the motor in the same direction with a back EMF of 120V.
That gives you the maximum possible braking force, which is the point the mechanical to electrical power conversion is at its peak.


The regenerated power is returned to the DC bus that is nominally feeding the output bridge, if the inverter was being used to drive the motor.

rjenkinsgb said:

There is no power factor correction with a _generator_ (or rather an alternator, for AC), regardless of how its constructed.

With no load, the alternator produces voltage - but there is no current. No current means no current phase relationship to measure.

When you connect loads to it, the phasing of the load current is the sum of all the power factors of the loads; some may lead and some may lag.
It's the loads that needs correction (if not resistive), not the alternator.


If you are meaning you want to rectify the alternator output efficiently, then a switch mode PSU with power factor correction would presumably be suitable.
However at the scale of the microwave oven motor, it's efficiency gains is likely zero or negligible.

Another approach would be in effect use a three phase PWM inverter circuit, as you would use to drive the motor, but continuously running in peak "regenerative braking" mode.

If I remember right, that's the point where the back EMF of the motor would be half that of the speed you are "braking" from; ie. if you have 240V back emf (output) you set the pwm to the percentage it would be at when driving the motor in the same direction with a back EMF of 120V.
That gives you the maximum possible braking force, which is the point the mechanical to electrical power conversion is at its peak.


The regenerated power is returned to the DC bus that is nominally feeding the output bridge, if the inverter was being used to drive the motor.

Click to expand...

Ok i ned to think about that, the loads are 3 wire BLDC motors as is the generator, the motor powering the bigger BLDC as a generator id a liquid fuel motor. I will give you more details.

Its a bit to get y head around, as i understand it the main motors i am using are not sensor ed, so one pole is always used to measure back emf to work out firing of the pulse etc, i think you set it to fir at 30 degrees past zero cross.

This should of been simple........but isnt, i tried it with straight diode rectified connection, and pommie is spot on, the motors heat real quick

Did you try a full wave three phase bridge rectifier, ie. six diodes? That should be reasonably effective; it's what is used in car alternators & most basic three phase PSUs.

If you tried single diodes the DC can mess up the magnetic properties of an alternator or transformer and limit the output compared to symmetrical current flow.

eg.
http://hiend-audio.com/wp-content/uploads/2014/04/3-phase-rectifier.jpg

rjenkinsgb said:

Did you try a full wave three phase bridge rectifier, ie. six diodes? That should be reasonably effective; it's what is used in car alternators & most basic three phase PSUs.

If you tried single diodes the DC can mess up the magnetic properties of an alternator or transformer and limit the output compared to symmetrical current flow.

eg.
http://hiend-audio.com/wp-content/uploads/2014/04/3-phase-rectifier.jpg

Click to expand...

Thank you, no i havnt yet. I have some other issue i need to attend to first with the design. I am not sure why but for some reason this is a topic i find counter intuitive. But its a project i am getting behind on, i feel i am 'dithering ' a bit with it. It will be cutting it fine,but in around 2 weeks i can take a 3-4 day block of time and try and get my head around this. Its for a special kind of drone power plant, not easy over all as realistically only one or maybe 2 motors on the drone will have power at any one time, so fast accurate switching to each motor is also something i need to consider carefully.

In hindsight i took on the project as drone projects interest me, but looking back i think taking this one on at this time was maybe a mistake. First time i have doubts i will get a project completed.

There is quite a serious article on alternator efficiency here:
https://cdn.intechopen.com/pdfs/381...ormance_of_lundell_automotive_alternators.pdf

Under most conditions the difference in efficiency with synchronous or PWM rectification looks to be fairly minimal and not worth the complexity.

The fact that current is only drawn for a small part of each cycle is not ideal, but in real world applications it's largely compensated for by the fact that the winding is unloaded and can cool for the rest of each half cycle.

If using a drone type motor as a generator, remember you need the same cooling air volume as if it were being driven at the same current or power in motor mode.


I was playing about with a motor-generator type idea a few decades ago for a kind of remote ATV using "diesel electric" power - a glow engine driving a big DC flight motor as a generator & that charging a battery to run the wheel or track motors, each with its own PWM controller.
It also allowed for it to self-start when the battery started to run down, simply switching the battery & motor connection on with a miniature power relay. Battery voltage sensing then controlled the engine throttle.


[I was heavily in to RC models and electronics at one time, producing all sorts of oddball devices - one of them is on page 54 of this magazine: https://www.americanradiohistory.co...tronics/70s/Practical-Electronics-1978-12.pdf ]



Edit - I've only just found your other thread mentioning the engine-powered drone..

How about contacting Bladon and see if you can get one of their micro turbine generators?? The smallest one is only about 3Kg from what I remember..

These are a bit bigger one of theirs - but 70KW each output at 35Kg weight: **broken link removed**

I posted a vid from a guy who has built a 2KW generator for a drone, very interesting channel and its got alot of information. I didnt realize the size! Its in the actual drone thread and the channel the vid is from is well worth a look, not your normal you tube rubbish

https://www.electro-tech-online.com/threads/photo-drone-mk1.153542/

He explains why they switched from 4 stroke over to a couple of two strokes. His drone is for very heavy lifting, but the generator side is very interesting especially as he addressed the vibration problem.

Crossover - I'd just found it before seeing your last post..