Tourque test balanced.

[Jerry]

I've added a balanced 12" beam. 24" centered and same size drive pully and pma pully.

Also added postal scales. I reversed the drive motor direction so the balanced beam pushes down on the postal scales. I did this test at 5 amps only from both PMAs.

Here are the new test #s.

                       Hughs 4 ft PMA

  5 amps from pma

  RPM  426

  Drive motor volts 60.4v

  Drive motor amps  2.4a

  Drive motor watts 144.96

  Weight on Postal scales 1 LB 13.5 oz.

                      Garbogen PMA

  5 amps from pma

  RPM  491

  Drive motor volts 66v

  Drive motor amps 2

  Drive motor watts  132

  Weight on Postal scales 1 LB 9 oz.

Flux I hope this helps with your calculations?

                           JK TAS Jerry

[Flux]

Thanks Jerry that is fine.

5A out into a nominal 12v battery lets say at 13v gives 65 Watts out.

Hugh alternator in is 111 Watts in. Efficiency 58%

Garbogen  input is  109 W   Efficiency  60%

Those figures look quite realistic. The drive motor current lines up with the torque as it should do so I think this is perfectly believable.

Flux

[ghurd]

Thanks Jerry.  Very Nice!

BUT,

Let us see the fish scale at the same 5 amps.

Let us see the same test at 10A, with the fish scale and postal scale.

Then they can compare apples, oranges, and bannanas.  :-)

G-

[gizmo]

58%, is that the sort of efficiency we could expect from most axial flux alternators, 60% give or take a bit?

Jerrys method is very accurate as far as I can tell, has anyone else tested the axials to get true efficiency figures? I know efficiency isnt everything but its often something the axial is praised with.

[Jerry]

Hi G.

I'm not sure what your saying here. I went to the postal scale for better accuracy.

Unfortunetly it was close to being pegged on the Hugh test.

The fish scale was to hard to pin point a # exactly.

I'd have to find a scale that reads to maybe 10 to 20 LBs full scale with some accuracy.

I thought this test was a good enough indicator?

I don't think the diferances will change no matter what scale I use.

Its just that some will harder to read then others?

                      JK TAS Jerry

[oztules]

Gizmo,

I don't think this is quite indicative of axial flux windings generally. It has 300 turns of spiderweb wire and so fairly high resistance. It is proper Jerry-rig, each coil is rectified independantly.

It has ceramic magnets which caused the winding number to be very high. This high resistance I think is what has caused the results to be not as I would expect for an axial. More watts are being lost in the windings than it would have had it been in Neo's.

Essentially, the efficiency in a PMA driving into batteries is  (in an axial) directly related to the winding resistance and the current drawn. By the use of the ceramics, this cruels this axial considerably as the current has to get through a pretty high resistance winding.

Eg, in my machine @5A @52v, we have 260 watts. My stator is .7R So the voltage loss over the stator at this power level is e=IxR = 5x.7=3.5v  w=ExI =3.5x5 = 17.5 watts lost in the stator. 260 watts into the batteries = about 93% efficiency... Also bearing drag, wiring run and diode loss and not much else.

At 2kw we would have 35A into the batts. @ probably 57v Stator loss would be 35x.7=24v..... Watts lost in stator will be 24x35=857Watts ..closing in on the 1kw losses but still nearly 70% efficient.... and starting to get toasty.

Being less then 3m off the ground at the moment, I can run it hard and stop it and feel the stator straight away, it would appear to bare these figures out. At around a consistent 1kw, it is still quite touchable....

.....................................................

I just went outside and stopped it, it has been running at 500-1000w all afternoon. (wind front coming through today .. reasonably blowy)  The stator was barely skin temp.

If I used more magnet and less turns, then it would be higher efficiency still.

If I messed this up we will soon know it...

..........oztules

[Ungrounded Lightning Rod]

Another way to look at it:

A given cross-section of copper with a given strength of field, pole and coil count and geometry, HP, and RPM, will produce essentially the same power and same percentage of loss whether it's divided coarsely (few turns, thick wire, low voltage, high current) or finely (many turns, thin wire, high voltage, low current).

But increase either the field or the RPM and you increase the voltage.  Then your power at a given current rises but the losses in the stator do not.

So weaker magnets make for a more lossy genny at a given HP - or at a given output - unless compensated for by higher RPM (which isn't available in this case).

[Flux]

gizmo

For a windmill alternator to perform well over a reasonable speed range it can't have a high full load efficiency or it will stall. That one is the single rotor design that Hugh uses with his little 4ft machine.

I have run many tests on various alternators and at cut in the axials are very efficient and you can use this efficiency in low winds before the blade tsr starts to drop. Normally the efficiency falls to about 50% at full rating if you want to match blades direct. You can hold the alternator efficiency higher and introduce the loss in the circuit rather than heat the stator and that has a lot going for it.

At cut in the efficiency is mainly determined by the bearings and I find that machines with ball bearings achieve a higher efficiency than those using taper rollers ( I still prefer the taper roller). You would never be able to see these factors with an iron cored machine with losses in the core.

When you get down to these lower losses you find that delta is a poor choice, the circulating currents within the delta give losses comparable with an iron cored machine.

Jerry's method is very good and will give perfectly good results in the working range but there is too much losses from the belt and the various bearings to get a true efficiency at a few Watts. Even at 60W I expect both alternators are doing a bit better but it is near enough.

The alternators all show a higher efficiency when loaded as normal ac machines into resistors than they do into rectifiers. The messy waveform and lower power factor of the rectifier holds low load efficiency down to about 80%. Winding resistance alone would predict values over 90% and you probably can get this with 3 phase resistive loading at low load.

Flux

[ghurd]

Pegged scale.  OK.

I was thinking if the proportions stayed the same at the lower powers in 2 spots,

then the higher power proportions should stay the same too.

"Reverse math" to get some ball park numbers at the higher outputs.

G-

[Darren73]

Jerry, if there is a way you can double the length of the arm then the force would halve allowing a greater range of tests.

[Jerry]

oztules.

Sorry to disapoint you budy but this PMA is a 100% by the BOOK Hugh Piggot 4 ft wind generator.

Without the blades and tail ofcource.

It Has a 9" X.25" steel disc, 8 each of the 1"X 2" X 1/2" NEOs. It is wound with the gage of wire and turns count that Hugh instructs in his book on the 4FT.

I baught the book and went by the  book.

It is not wound with 300 turns of spider web wire. It is not connected "Jerry Rigged" it is conected star as per Hughs instructions.

Now critisize this PMA and your critisizing Hugh.

My whole point from the beging was to compair the 2 diferant aproaches to these 2 PMAs.

A 100% by the book Hugh Piggot 4ft and the Garbogen. And right now I think our old freind Zubbly (rip) is looking down with a smile on his face.

I stated in the very begining of this compairison what 2 machines were being compaired.

However speaking of the ceramic magnet, 300 turn "Jerry Rigged" machine. The 14-11, when its done I will put it through the same test.

Wouldn't that just be a shocker if it out preformed both of these machines at 5 and ten amps.

When I test that one I'll tell you and discribe the machine I'm testing and under what conditions.

                                JK TAS Jerry

[Jerry]

Hi oztules.

I looked one of my diaries on the build of this Hugh 4ft machine.

The story has the alt specs.

      http://www.fieldlines.com/story/2008/4/28/45732/6107

                     JK TAS Jerry

[oztules]

Jerry,

My sincere apologies. I went back through your diaries, but didn't think to go back to April. I figured it was amongst your latest diaries..

I would have to agree that the king of motor conversions would be well pleased with your results.

Your test rig seems to reflect a good comparison now, so Hugh looks to be beaten fair and square by a garbage device!! Well done.

I hope to see the jerry rig ceramic do a fair effort too. I like the jerry rig for each coil independently, and may do the next stator in that configuration... if I can find enough wire of the right size for my project (2kw).

The numbers on mine seem to add up ok, so at least I got that bit pretty right.

Dinges (on the backshed site) has nearly completed a large motor conversion, and I think will test it to death for us. I will be most interested to see the results of that one too.

Well done as always

.........oztules

(village idiot)

[Darren73]

Jerry, out of interest is there a way you can jerry rig the hugh's 4ft or is the star point buried, I appreciate it will take more rpm to drive however I am interested in seeing how it affects the efficiency more than the comparable speeds.

keep up the great work

Darren

[Flux]

You can't really get a direct comparison with the same stator. Jerry connecting it will raise the speed and that will be bound to raise the efficiency when loaded into a given voltage battery as the effective winding resistance is reduced.

Flux