Jerry's new dual rotor wind test.

[Jerry]

This has been a fun project. Allthough way more time consuming then a motor conversion.

The goal of this project was to gain knoledge on the dual rotor desing and find a alt desing that would have a symular power curve as the Garbogen.

The Garbogen does a couple amps at 5 to 7 mph and hits 650 watts at near 30 mph. It also has my 49" 3 blade Jerry blades.

This dual rotor disc alt is 8" with small wedge NEOs. Paterned some what after EDs kit 8".

I think Ed sugjest a 7ft blade and it has larger magnets. For this reason there was no reason to expect this genny to come close to Eds rated 500 watts.

There are a couple things in my 8" alt that is not done as standard practice. I didn't go totaly by the book.

1 this was to wind my coils of 22 ga. wire and at 250 turns per coil. Next I did not use either the accepted or standard delta or star 3 phase conection.

I used a fullwave bridge rectifier per coil. Each coil has its own (9 each 6 amp bridges). Then all the DC outputs of these bridges are wired in perelell.

With a spin of the hand I can get around 18 volts no load dc. On the lathe at 500 rpm 44 volts no load.

On to the wind test. I new from lathe testing that this genny is best suited for 24 volt use.

But I did a 12 volt wind test anyway. I had to go 30 mph to even get 10 amps at 12.6 volts. It seemed as though the brakes were on the blades just wouldn't spin up. Bad blade stall. This was disapointing, I thought it might do better with a blade then it did on the lathe?

However the 24 volt lathe test yeilded 283 watts at around 400 rpm and I knew my 49" blades had done 650 watts at 500 rpm.

So I rewire the 2 new Optima red to batteries in sires and went back out on the hiway for a 24 volt test of this genny.

What a diferance. At the turn around point I had stoped the truck, I was hearing a sizeling sound, I got worried, I didn't smell smoke? I came to find the Optimas were complaining about the 35 amps and 35 volts I just put to them for a short time.

Heres the mph and watt figures.

  5 mph....36.75 watts.

 10 mph...130  watts.

 15 mph...280  watts.

 20 mph...450  watts.

 30 mph...726  watts.

 35 mph..1020  watts.

 40 mph..1225  watts.

I couldn't belive what (watts) I was seeing. I never thought this littel genny with these littel blades would come any where close to these #s.

My thoughts were I build this littel 400 watt genny then biuld a larger version to equal the Garbogen specs but this littel unit follows it very well and even a littel better.

Its surprizing to see these #s from a 49" blade? But I'm luvin it.

I also think this confirms my feelings about the antiquated thinking about useing the very missfit ineficient delta or star conections.

Test equiptment used for the above test. Fluke true rms digi meter, very nice 50 amp analog amp meter and a Chevy S-10 speedomiter.

To make this genny right for 12 volt use fewer turns of larger wire will do the trick.

I'm now working on a 10" dual rotor with 1.5" round X 1/4" Neos. It will have the same 9-12 arangement and yes 9 bridges. I'll do blade extentions on this one to 55".

Here a few pix. Hope you don't mind. I'm just rather excited.

                          JK TAS Jerry

[Jerry]

You can see from the pix the blade is shaft mounted about 3" in front of the rotor.

That close proximety worried me some what. The root of the blades are only 3" apart. I thought that closeness to a 10" stator just a few inches behind the blade might have a bad effect on the root portion of the blade. Gues I was wrong. The blades are 23 degrees at the root and just a few degrees at the tip.

Another thing surprized me. At 1 point I thought the blades went missing. I stuck my head out the window while driving just to see if they were still there. They were extreamly quiet. Even more then normal for these very quite blades.

Also after returning from a 1200 watt test with the Garbogen(motor conversion) it is quite hot. On my return from testing this littel dual rotor 8" at 1225 watts the stator and coils were stone could. However after doing the math each coil was doing less then 4 amps (3.8888a) each and the same for each 6 amp bridge.

I figure I'd have to see around 1620 watts befor things start to heat up. That 6 amps per coil and bridge at 30 volts.

This is all very much fun.

                       JK TAS Jerry

[hiker]

up up and away-----

sounds like a super alt. their JERRY...........

might have to rewire my alt--and give it a try...

[johnlm]

Jerry

I guess Im puzzled.  Not sure how you are measuring the output but you are reporting outputs that seem to be impossible for a 49 inch prop.  See my comments beside your numbers.

Heres the mph and watt figures.

  5 mph....36.75 watts.    Betz limit: 5.0 W, you get 7X higher?

 10 mph...130  watts.      Betz limit: 40W, you get 3.25X higher

 15 mph...280  watts.      Betz limit: 136 W, you get 2 X higher

 20 mph...450  watts.      Betz limit 323 W, you get 1.4 X

 30 mph...726  watts.      Betz limit 1089 W, your number is high but possible if

                           your alternator is running 95% effecient which Im sure

                           its not.

 35 mph..1020  watts.      Betz limit 1730 W your number still high but possible.

 40 mph..1225  watts.      Betz limit 2582 W, your number possible

As you can see there is something wrong especially below 35 mph.  Not only do you exceed Betz substantially but Im sure your alternator is not more than 100% effecient.  Maybe the speedometer on your pickup is way off (reading 5 to 6 mph lower than real speed makes the numbers make more sense) or something else is wrong.  And by the way in theory rectifying each output individually then adding in parallel cannot be as effecient as a Wye connected 3 phase unit with the same number of magents and same coils used.

Nice looking unit, but you might want to check out your speedometer, you might get a speeding ticket in a school zone.

Regards,

Johnlm

[kamikaze]

Wow! What a result! I'm going to go back over your diary for this project to try and understand how you did it. I think you're on to something big.

Inspiring stuff!

Regards, Kamikaze

[electrondady1]

this is very interesting jerry, congratulations. please do more testing if possible.exceding the betz limit on your output figures is kind of puzzeling. as john has pointed out it might be the speedometer.   the thing is , is this the same truck you did your other testing on . i'm just a novice builder and not enough experience to know what would be  the normal output from those mags using a more typical coil setup. what degree of improvement do you feel you have made by using your multi rectifyer/individual system?

[hiker]

sence its not 3 phase--3 coils hooked togeather[times three]-- would it work any better using 1 mag per coil like single phase--but still rectife each coil?

[Flux]

Nice machine Jerry, I have been watching the progress, you make a nice job of these thing.

I am a little relieved because your garbogen figures have always puzzled me and I have been dreading the day when you confirmed that this little axial was going to be a disappointment to you. Now with the same testing method you have come up with these figures I am happy.

I expected the air gap machine to considerably out perform the machine with iron core and core losses in lower winds. I was happy to think the garbogen could have won in high winds.

Measuring performance with wind speed is incredibly difficult and full of problems.

I am surprised your figures are so high, but air speed increase over the cab and errors in measuring speed are inevitable.

The figures are undeniably suspect but I still think you have a very good comparison between the two types of machine.

Flux

[Warrior]

Hi Fellas,

Great job Jerry!!! Your work is always cool and I love breaking the "rules" just as you.

I can't remember which one of Jerry's machines it was, but I too found the numbers to be above the benz limit, giving above 100% efficiency.

The general consensus was that the car speedo was off, making the numbers possible.

A GPS might be the best method for car testing.

Jerry, we're not trying to through down your work, just trying to find the cause of thje strange numbers.

Keep up the good work!

Warrior

[Gary D]

Jerry, great to see you like the duel rotor type! Wonder if you could open the air gap just a bit to make it a better 12 volt machine? Two in hand #22 guage 125 turns might work better than a bigger gap? Is this a downwind unit or is the tail just not on in the pictures? Love seeing different methods of getting power from the wind! Great post, hope it has been worth the effort, still wonder if you could get 10 coils in there, but the blades probably would need more swept area....  Gary D.

[electrondady1]

jerry, like i said in my first post this is very interesting. in fact it might be very important. like, change the way everyone builds generators important. i think there is only one way to be confident about the method you are using . on your next project, the one with round mags, wire the stator up in both traditional three phase star with the normal rectifyer lay out and be able to use a switch to shift to a "jerry riged" system. do this test on your laith at various rpm. be realy scientific about it. do it so that there can be no question about the results.   i think your on to something .

[DanB]

It could be a speedometer issue - it could also be the fact that the machine is mounted in the back of the truck, maybe the air over the top of the truck is moving a bit faster - or it's denser (or both)...

I shouldnt try to think about this stuff before coffee though.

[Gary D]

Another thing Jerry, the compressed air that your truck puts off is similar to a roofline... you could raise the apparent wind a few mph with a jenny positioned at the right height above the roof. It's possible to get a rise in output similar to a higher real wind without a higher tower? Your blades might handle this air better than wood blades due to their flexability? Bentz doesn't have to be broken, just bent a bit... Gary D.

[johnlm]

Jerry,

I guess I should say that Im not trying to be difficult.  I really enjoy your work and your enthusiasim and think you have done a great service to this AE community with all the novel ideas and approaches you have brought forth over the years.  Im just a stickler for accurate information as many on this board seem to be when some outside company advertises performance like this, it is usually widely questioned and at times strongly condemned. In not without error myself, and it is difficult to accurately measure performance of wind turbines; but Im only trying to keep the integrity of the information presented by contributors of this board at a level above reproach.

Johnlm

[finnsawyer]

Part of these discrepancies may be due to the effect of the truck cab on the airflow.  Most people don't realize it, but the air disturbance that an object creates actually extends a considerable distance (many multiples of its basic dimension) from the object.  The air passing over the windshield and cab is sped up.  This effect can extend quite a distance above the truck as this speed up drops off relatively slowly with distance.  Because of the truck's complex shape it also could vary with the speed of the truck.  Those wishing to use a vehicle to test their wind mills should mount the mill in the front of the vehicle so as to guarantee a uniform air flow at the vehicles speed past the prop.

One effect of the fast moving air coming off of the cab of the truck is to cause a vortex to appear over the box of the truck.  This will then affect the air flow over the wind mill.

[DanB]

I'm with you 100% - I really enjoy watching Jerry's work, it's always well done - well presented and very interesting.  If nothing else - truck testing is quite valuable just for comparing different machines, even if the numbers come out somewhat skewed.

I think accurate testing is really tricky stuff... I think it takes either lots of data  aquisition and time on a good tower, or a wind tunnel (or both).  Most of this stuff is tricky/expensive for homebrewers.  

Very nice job on this one Jerry - thankyou for sharing it.  Do you notice much difference in startup between your conversions, and this dual rotor machine?  Do you think it takes noticably less torque to get things turning?

[para t8]

lookin good jerry,the#too.do your gennys power your store?you toke the idea out of my head,i was thinking the samething about lesser turns thicker wire.      may the wind be in your blades and the sun in your panels,  para t8              

[zubbly]

hi Jerry!

very happy to see that you got some excellent results with your new genny!  :=>

whether the air movement speeds up or not over the roof of the truck is not really important. the fact that your testing procedure is the same is what really counts. i wish i had a set up on my vehicle to check performance also.

as Danb put it, a combination of truck testing and wind tunnel testing over a period of time may be more accurate, and i am sure it would be, but i think what you do is just fine for answering the question of checking performance. to be very scientific, you would also have to take into account air temperature and also humidity.

just my opinion, i really don't think comparing the performance of a conversion against a dual rotor can lead to any decision as to which is a better machine. they really are quite different and would be liking comparing a diesel driven boat to a steam powered boat. they both get you there.  one machine uses a core, the other does not. one takes more copper, the other does not. one uses considerably more magnet material than the other and there is a huge weight difference between the two given the same size prop and output.

I love motor conversions, but also ador dual rotor design and the unique look of them. both are excellent units and both end up giving us what we need.

i started a dual rotor unit about 2 years ago just for experimenting with. basically it is 2 1/2 inch steel plates with 14 inch diam. you've sparked my interest in them again and i really should finish it just to see what i can achieve with it.

damn!  here i go again.  up to project #07607078556   LOL!  what a habit this is. absolutely no cure!

have fun Jerry!

zubbly

[SparWeb]

Jerry,

Even a small head-wind would account for the discrepancy.  You could increase the confidence in your figures if, the next time you truck-test, you drive in multiple directions.  Take the average between, say, 20mph northbound and southbound.

[scottsAI]

Hello Jerry,

Awesome testing out ideas, when results are better than expected, it's time to sit back and find out why.

First thing, bridge rectifier on each coil.

If coils are laid out as in 3 phase then the output is going to be less than Delta.

Your using 12 rectifiers, vs the normal 6.

In Delta configuration one phase is at peak, the sum of the other two are on par.

Each voltage above the battery can and will be contributing.

With individual rectification you will not get the adder of the other two phases.

The average current will be lower.

Local wind speed needs to be considered.

Testing in both directions helps rule out local wind speed boost.

Wind turbine needs to be above the influence of the vehicle.

Measure the wind speed without turbine, find how high up to remove vehicle influence.

Measure wind speed just before going into wind turbine, many simple devices around. Any body that built that beautiful turbine can build one.

Keep up the good work, now time to dig in and find out what happened.

Have fun,

Scott.

[Jerry]

Hi Dan B.

Yes I noticed during installation of this genny on the mast in the truck that the blade started turning even though it was off axis to the wind. There is a lot of weight in a dual rotor but the startup is very easy.

                      JK TAS Jerry

[Gary D]

Scott, "each voltage above the battery can and will be contibuting" sums it up! As the blade rpm increases, each coil top end voltage wants to rise. It's clamped by the battery voltage tho. So the coils even not at "peak" will contribute, as long as they are above battery voltage (rectifier and line loss too). Just as a standard 3 phase machine, but different...  Jerry has definately tried and succeded with his first attempt to do something completely original. I think I get it Jerry....  Gary D.

[Jerry]

Thanks to every one who has responded to this story. I do appriciate all the input.

The purpuse of this test was ofcourse to learn something. All the great input also helps in the learning prosses.

Also the purpuse of this test was to see a compairison of the Garbogens wind/truck test results to the exact same test proceedure of a small dual rotor with the same blades.

I do agree that my test results do not reflect a real world or tower mounted preformance.

But I did what I set out to do. and that was to compair 2 diferant alternator designs with the same blade and same conditions.

I have room at my store for both of these machines to be tower mounted at the same hight and load.

This should show real world results and a good side by side compairison.

Even with the very flawed testing I did not expect the #s I was seeing from this littel dual rotor. I thought I would have to build at least a 10" dual rotor with larger magnets?

I concur that my location of the genny on the mast in the truck bed is subject to increased wind speed over the windshield and cab. Also the varying prevailing winds through things off conciderble.

However the Garbogen was tested under the same conditions.

I posted last year about building a wind tunnel. I still want to do this for more acurate testing but that is still some time out.

To respond directly to a few of you. Hiker yes it is still 3 phase. There are still 12 magnets per rotor and still 9 coils.

Johnlm thanks for the Betz calculations.

Electrondaddy1 yes this is the same test truck I've allways used. Not sure if theres an improvement with my coil diode scheem but I'm convinced its better than either delta or star. I intend to go into great detail on this subject on another post or diary.

Flux as usual you are right on. As you pointed out this was a compairison of 2 alt under the same conditions.

Warrior the GPS sounds good but I'll put the expence of the GPS into my wind tunnel.

I think I'll have better accuracy and not have to worry about trafic.

Gary D I'll keep the gap the same. In my next dual rotor alt a may just build several stators and do a laithe comparison. This will however be abit tricky to make the coils campairable in potential but with there diferant charicteristics.

paraT8 keep working at your genny don't let any thing I do stop you. I'm eager to see your alt and test #s.

Thanks again everyone. I'm going to do some home work (whoops dog eat it) and offer my opinion on the star-delta or Jerry rig it contraversy. I'll post that latter here or on rants. I'm to tierd tonight and my brain is out of phase.

                    JK TAS Jerry

[richhagen]

There is only about 8 watts or so of available power in the wind to a 48" rotor at 5 mph not counting Betz and other inefficiencies besides.   Your rotor is only  49", about 4.2% more swept ares, so the numbers on the low end are probably off a bit.  Since the power available is proportionate to the cube of the windspeed, it really wouldn't take much to create a huge error on that end of the testing, a small switch in headwinds of less than a couple of miles per hour, or a similar error in the speedometer could result in more than double the available power.  Combining  multiple margins of error could exagerate the low wind readings to a multiple of there actual value pretty easily.  The little mill turned out well though.  I am going to try a small dual rotor axial again soon as well.  Your results are further encouragement.  I have a small plasma cutter now, so I can do the metal work a little more easily.  Rich

[hvirtane]

Really nice work.

Thank you for publishing this.

I think that some further tests

might confirm Jerry's idea

that rectifying all the coils

separately will give the best power.

In fact Alfred Forbes also got the result

that rectifying all the coils separately

was the best (Alfred Forbes: 'Homebuilt Dynamo'.)

The same result was reported by

'WindstuffEd' some time ago.

I think that reason for this is that

you cannot get the wires accurately enough

for delta wiring (or star either). When

most of wires are a bit of off-phase,

there will quite much cancellation

of the power from different phases.

In theory a perfect delta wiring should

be as efficient as rectifying coils

separately, but in practice you cannot do

a perfect three phase system.

- Hannu

[DanB]

It would be fun to see some good tests.  I can see where it would be an improvement over Delta - I have a hard time understanding why it would improve over a Star connection.  Its been discussed here before and my understanding is that it comes up about the same as Star, though I cant quite see why - to me it seems that with a more conventional 3 phase arrangment we have current flowing in all the conductors, all the time, which seems to have a benifit over 'single phase' coils which spend a good bit of their time with no current flowing, especially near cutin when we're only using the very peak of the sine wave.

I would considering playing with that if it was a big improvement... if its the same, or even a very 'small' improvement I wouldn't simply because I don't care for rectifiers at the tower top.  If it is better though - or if I were going to do it for any reason I would consider doing something more like hughs 12 pole/10 coil arrangment or something - just to reduce vibration and have smoother operation.

[Flux]

Here are some results of a test I did several years ago.

Red is output, green is input for delta, blue is input with individual rectifiers.

Hope the gif works.

X axis is rpm  y axis is watts.

I didn't persue it at the time, seemed near enough to be within experimental error.

Both worse than star, but I dont't have a direct comparison with the 1.73: change in turns needed to be fair.

form your own opinion, I'm with Dan on this.

Flux

[Flux]

Sorry, my mistake, lets be fair to the thing, blue and green should be reversed from what I said.

[electrondady1]

the way i'm seeing jerrys latest effort is that there are actually two experiments takeing place. the first is runing a rectifyer to each coil and the second is he's able to form a coil which on its own can supply a usable voltage.(can't really do the second with out haveing done the first)   .i'm curious as to the total internal resistance of the geni,

[Flux]

From the results of my tests the effective internal resistance will be the same as any other generator of the same output and efficiency at the same speed.

How many volts you produce per coil is not an issue, Hugh's 5 phase design does much the same and produces the full voltage from one coil rather than a string of series coils, doesn't affect the  basic argument, but you are right that each coil has to produce the full volts.

Flux