G'day Guys,
Well a mate has been saying he tried using non polarised caps on the phases of the F&P for a greater output and I did see the results...
Very impressive So not to be out down I had some 400 volt 470uf polarised caps here so by putting a pair back to back I made non-polarised caps. Tonight when I was up in the shed working on my cnc project I heard the genny take off so I went and turned on the cheap dmm set on 10 amps and instantly saw 8 amps . I then turned turned on my fluke 865 on scope view to see the waveform and cutin voltage and YES the caps are working perfect. Before it took a wild gale just to get 5 amps and mid winds were pultry current readings. Now with the caps as soon as cutin happens the current basically goes straight up to 2amps them if the blades speed up 6-8 amps bursts. I't will be good see the output in a full storm and I'm getting a cheap 30 amp analogue current guage tomorrow to hookup as I like my dmm's.

Cheers Bryan

__________________
" The only way to avoid human error is to avoid the use of humans"

 

I think you already know what capacitors do in a PSU?, and they don't produce more power - they just store the energy and release it as the input drops below their charge level.

If you're using it to charge a battery it should make no difference whatsoever, the battery already does exactly the same function.

__________________
PIC programmer software, and PIC Tutorials at:
http://www.winpicprog.co.uk

 

Sounds like your current readings were/are distorted by the frequency. As Nigel says you're not generating more power, your meter is just reading it differently now as the capacitors are smoothing things out. The analogue current meter will likely provide a more reliable reading weather or not the capacitors are connected. Even true RMS digital multi meters are only good up to a few hundred hertz. If you were feeding it raw generator power with minimal filtering it was definitely not reading the correct current draw as the generator will produce frequencies way past your meters usable range.
Don't think adding more capacitance will increase your power output because once you get past basic ripple filtering the battery doesn't care and you're just wasting power through the ESR of the capacitors you're using.

__________________

 

To get a "measure" of power you could connect a filament lamp(s) of suitable wattage (to obtain similar currents that you observed) to the genny output, connect/disconnect the caps & see if there is a change in brightness.
or
Place a decent resistor as the load & measure the temperature rise of the resistor relative to the resistors ambient. (shield the ambient probe from resistor IR)

Both of the above should indicate power produced regardless of frequency & the load characteristics are resistive, the time constant however of the second method will be much longer.

Would be good idea to bench test the F&P gen being driven of a motor.

I havent given this much thaught, but if there is more power being generated with the same wind speed it may be that the capacitors are helping to reduce magnetic drag in the rotor.

I have observed similar improvements in some DC brushed motor power when fitting a capacitor near the motor with a long / lean wire feed from the power shed about 30m.

 

hi Bryan,

'Tonigau' has the correct method, measure the power output, rather than just the voltage. The lamp brightness idea will give you a better indication of the rms value of the current/voltage.

__________________
Eric
"Good enough is Perfect"

PIC tutorials:
Gramo's: www.digital-diy.net/
Bill's: www.blueroomelectronics.com/

 

Guy's the caps are placed on each wild AC phase before the rectifer. My mate has done extensive testbench testing on this method using motor run caps and has proven it does work. I suppose with the F&P having 42 poles the caps are just doing some PFC and thats why I'm seeing a better result. We have complete data on all the different F&P motors on a graph ie: speed of the prop against current. Now using the caps those graphs are being re-written. Before I put the caps on the waveform was a distorted sinewave at best after the caps the waveform is damm close to a sinewave.

Now i do agree you don't get any extra power out of using caps but by putting them on each wild AC phase it's making a BIG difference. I have looked extensively on PFC to try and explain this but the only thing I could find was using a fixed frequency and the frequency varies with rpm so I haven't been able to find an equation to solve this on paper.

Hopefully this weekend we'll have some gale force winds and if so I'll take some pic's of the results on using caps and without caps.

Regards Bryan

__________________
" The only way to avoid human error is to avoid the use of humans"

 

Bryan, you are correct about the Power Factor Correction that the capacitors provide. The rectifier/battery load is nonlinear which causes the distortion and power loss. The capacitors simply store the energy from the unused portion of the cycle (when the rectifiers are off) and help supply current near the peak of the waveform when the rectifiers are conducting.
A better method would be to use active PFC (A boost converter [followed by a buck if needed]) so that the current is in phase with the voltage. This circuit will also inherently regulate the voltage to the batteries. This will better eliminate the harmonics than the capacitor solution.

__________________
--- The days of the digital watch are numbered. ---

 

Like any electrical device batteries will last longer with a regulated charge as opposed to a lot of ripple.

__________________

 

Ripple kills batteries as they're being charged and discharged at 100Hz. They aren't like capacitors which don't have a limited number of cycles.

__________________
I also post at the following sites:
http://www.stop-microsoft.org http://www.heated-debates.com
Screen name: Aloone_Jonez
And http://www.silicontronics.com, same screen name as here.

 

Most alternators have a lot of inductance in the windings. The open-circuit voltage of an alternator is often many times the running voltage. That is why load-dump on a car alternator is so severe and why voltage regulators on AC generators are so important.

In constant current, variable speed applications, the large inductance is used as a current limit. That applies in car alternators and in directly driven AC lighting like on bicycles and some small motorbikes. The current is largely independent of voltage load at higher speeds. I was able to change my motorbike lighting (1983 Honda CG125) from 30W at 6V to 60W at 12V with no change to the alternator, but it was dim at tickover.

On the wind generator, the capacitors in series will form a tuned circuit with the inductance of the windings. The impedance of a series tuned circuit is less than the scalar impedances of the inductor or the capacitor. The lower impedance may increase the current a lot. It isn't breaking any thermodynamic laws, it's just bypassing a current limit.

It is quite possible that the generator takes more shaft power and slows the turbine more, so takes more power from the wind. The shaft power is difficult to measure, and it isn't fixed for any particular wind speed. At zero speed, or at zero load, there will be zero power. Maximum power will be at somewhere around half the zero load speed, but it is unlikely that a small wind generator manages to hold the turbine close to maximum power speed all the time, and that is where some improvement can be had.

The downsides could be:-
The current limit is compromised, so there could be damage to the alternator if it is designed around the natural current limit.
The batteries could be charged too fast.
The effect will only be at its best at one wind speed, as there is only one frequency that gives best tuning. However, there could be some improvement at all useable wind speeds.

I agree that AC currents from wind generators are difficult to measure.