Flat plate Axial Mill Coil Test Results

[brianc4]

I didn't think I would get to it today but I got the test coil mounted om some plywood and placed between the rotors for a test spin.

I wound up leaving my digital tachometer out at my cabin so I am only guessing at RPM. I will bring it with Me tomorrow I post the exact RPM.

I wound the coil with 36 turns of 11 ga. square magnet wire & at what I am guessing is 200 rpm I am getting 3.16 Volts.

If I am following the math right 3.16 X 3 Coils = 9.48 Volts per Phase X 1.73 = 16.4 Volts unrectified output. Am I following the math right & if so is the 16.4 Volts going to be about right for a 12 Volt machine once rectfied?

Brian Clark

[tecker]

Star. Think about bringing each coil to an edge connection so you have max flexabilty.

As your system grows you'll want a 24 or 48 volt system .

[Flux]

"If I am following the math right 3.16 X 3 Coils = 9.48 Volts per Phase X 1.73 = 16.4 Volts unrectified output."

16.4 x 1.4 = 23v dc.

This is effectively cut in at 24v at 200 rpm.  I seem to remember that this is a 10ft machine. Unless the blade tsr is very low this is going to stall like hell at 12v. If you are in a good wind area and your blade tsr is 6 or over I would think it is going to work far better on 24v. With a boost converter for the very low winds this would be ideal for 24v.

For direct connection cut in about 160 rpm would be about ideal so you would aim for 45 turns for 24v or about 24 for 12v.

If you are prepared to deal with 6 leads ( and 6 slip rings if you go that route )or mount the rectifiers at the top, you would hit it pretty close for 12v with it Jerry connected ( each phase rectified with a single phase bridge).

Failing that I would try to get a few more turns in by reducing the hole size if necessary and aim for 24v.

Flux

[brianc4]

Flux,

I hated to post incomplete information yesterday but I was just so glad & excited to finally see the mill put out something!

I brought my digital tach with me this morning & can now post accurate info.

At 160 RPM I was getting 2.17 Volts At 250 Rpm I was getting the 3.17 Volt reading.

I was planning on bringing only 3 leads down the tower & rectifiying at the battery.

Also plan on hooking up the stator in star wound configuration.

Now what I have evidently completley misunderstood is the rectification to DC.

I thought there was a voltage loss when you rectfied! How & where does the X 1.4 figure come from in the rectification formula?

I have a set of 10' 6" fiberglass blades I am planning on using on the mill. I got them from Applied Magnetics the same place I got my magnets. They advertise a TSR of 6 under heavy load & a startup speed of 6 MPH. If they are not satisfactory I will make a set of blades but with time constraints at the present I am going to try them.  

I also need to mention Battery bank size I have several big 12 Volt steel case forklift batteries at 750 amp hours each. I plan on putting in several more. I plan on always having room to store energy when the wind is blowing. I also have a Xantrec C-60 charge controller for diversion if the batteries do wind up at full charge. I will hook it up to a water heater. The batteries are individual 2.2 Volt cells & can be hooked up for any voltage I choose but I was planing on staying at 12 volt for now if possible.

My usage at this time is very low limited to lights only. But as soon as the system is up & going a deep freeze will be added & a refrigerator.I will also have a computer running full time to log data from the mill & my weather station.

We don't use many small appliances, a mixer once in a while & a blender sometimes.

We dont own a TV or VCR as there is always something more entertaining & usefull to occupy our time.

In the future I do plan on putting up a shop building so I can work a few days out of the week at the cabin & avoid the 20 mile round trip in to My main shop in town. This will change demand considerably.

Sorry it has taken Me so long to post complete information about my system as I know it is necessary information but I have been so busy with customers & the fab work on the mill I barely have time to sleep.

I am not opposed to a 24 Volt system but I was planing on 12 Volt. If I need to decrease number of turns in the individual coils to 24 instead of 36 not a problem.

Thanks for the help!!

Brian Clark

[Flux]

If you think about normal loads such as lamps run from a single phase rectifier then the mean dc voltage is lower than the rms ac.

For charging batteries current flows when the dc input volts exceed the battery volts. For a single phase system current will start to flow as soon as the peak ac value exceeds battery volts so nominal cut in occurs when the ac peak reaches battery volts. this peak is 1.414 x rms for a sine wave. The current pulsed are tiny at the start and you will probably need to get the ac up to near battery volts before there is enough current to say you are really cut in. The current pulses get longer as the voltage rises but there will still be breaks in the current at voltage zero.

For 3 phase the same true cut in occurs at 1.414V rms, but there is not much ripple on a 3 phase rectifier and the voltage never falls to zero as for single phase. For a lamp or similar load the 3 phase dc mean of a 3 phase rectifier is 1.35 x vrms so conduction becomes continuous just beyond true cut in at peak. It is reasonable to use a value of about 1.4 v rms as a sensible cut in point.

You have slight discrepancies between your two readings so let's take the higher voltage as your meter is more likely to be better at that point.

For 3 coils you have 3.17 x 3 x 1.73 x 1.4 = 23v, so we now have the 24v cut in at just over 250 rpm. 12v will be over 125 rpm and we have some diode drop and the battery will generally be about 14v. This looks to be in the right region for 12v, a little on the slow side so you could reduce turns a bit or widen the air gap. I think I would drop a few turns and then do fine adjustment on the air gap to get cut in about 160 rpm. if you stall with that cut in you will need a bit of line resistance but that is unusual at 12v where line losses are generally high.

My guess is to go in at 30 or 31 turns per coil.

Not sure what they mean tsr under heavy load, it's not load dependent. Probably the things are about tsr7 and will run down to tsr6 without significant stall so it should be fine for 160 rpm cut in.

Flux

[brianc4]

Flux,

Thanks greatly for the explanation & the help. I will drop a few turns & adust air gap accordingly.

I have never ownwed a good test meter! Never have had the need for great accuracy till now!

I know that Fluke brand is top of the line & priced accordingly. Any suggestions on a accurate meter that isn't so expensive.

I do relize you get what you pay for most of the time but I don't want to waste money on more meter than I need.

Thanks Again:

Brian Clark

[Flux]

I am not familiar with the meters available in N.America.

Most of the common multimeters other than the very cheap ones at give away prices are near enough for this sort of work. The ones with only 2 ranges for ac volts are best avoided.

Fluke are excellent but there may be little point in paying lots of money as the very expensive ones are true rms and for this sort of work with strange waveforms they may be no better or in some cases worse than the basic mean reading scaled rms types.

For measuring many things involving power, true rms is necessary but there is no particular virtue in reading true rms with waveforms chopped up by rectifiers, batteries work on average volts and current.

At the very best this measurement of ac on a test coil and predicting the dc out is not going to be accurate to better than a few % and in extreme cases where people choose strange magnet spacings or coil shapes errors of 30% can be common.

It is usually in the region below about 5v ac and at frequencies lower than about 40 Hz where things can go astray and that is the region where people tend to base their measurements especially if hand cranking to find speed. If you can motor the thing at a fair speed it makes the voltage and frequency higher. It is perfectly ok to take speeds in the region of 1000 rpm and scale the volts to the value expected at low cut in speeds as long as you can measure that speed accurately and drive it at a constant speed.

With this type of alternator the open circuit voltage is linear with speed. If your voltage/speed readings show non linearity then there is a measurement error, not a funny alternator curve. Only one point is needed but two are useful to confirm things are reasonably correct.

Flux