Littel alt lathe test.

[Jerry]

Well I'm happy to say I don't have to set in the corner with the dunce cap on.

My first test was with the star stator. 116 turns of 21 gage wire. Test rpm 500. first thing I noticed. the voltage was realy low.

After a gave that a littel thought it hit me that this little alt has only 4 very small magnets and just 3 coils, a gap of about 1/8" inch between the magnets and the coils and a diameter of only 4 and 3/16".

So I jumped the rpm up to 1600. This coil group is wired star. I used a 4 ohm 400 watt none indutive resistor as load.

I plan on doing a battery load test and look at the rippel and wave form on the osciliscope but that will be in a fewdays.

I recorded dc amps and volts. Then I removed the star stator and replaced it with the jerry riged delta stator.

Here are my test results.

  Jerry riged volts=4.484                          Star volts=3.890

  Jerry riged amps= 1.11                            Star amps= .96

  Jerry riged watts=4.977                           Star watts= 3.734

My point is still. I you take 1 once of copper and I take 1 once of copper and we both use the same magnets same discs same everthing. You use a wire gage twice the size of mine I use a wire gage half the size of yours.

Your out of phase sires conected coils will rob you of the power that exists within the same amount of matirial and the same money spent, the same size and the same weight.

There has been some concern that this jerry riged alt will reqier more blade power.

Yes it will, I know that. But at the same time it will give you the power of an alt that was built star with larger and more expencive magnets and larger and more expencive copper and every thing else that is required for a larger alt.

This is scalable. So imagin the power you could acctuly get from a 2 KW alt if you wernt wasting its power on out of phase coils that are not phase compatible with each other.

And for those that like to speed there alt up buy puting resistance in the line so the blades can reach a more produtive rpm, you can still do that.

Personaly I don't think we should build ineficient alts with star phases just to make our blades happy.

I think we should get all the power are magnets and copper can produce and build an apropriate balde to mach that alt. Controls such as dump loads and resistance can be added down stream from the genny to make electrical addjustments on the ground.

I don't want to be standing on the ground and look up at the genny on the tower and say that genny could be making more power right now if I had used a diferant phase configuration.

If scailed up its the diferance of 124 watts. If you build a 373 watt genny it could be produceing 497 watts at the same rpm. And again I relize there will be rpm diferances with blade diameter. A higher wattage alt requiers a larger diameter blade that spins at a lower rpm.

The diferance is the alt will cut in much sooner without the nessesity to build a much lager alt.

And Dan B I'm sorry buddy but you said it yourself. The voltage is lower cause the phases are out of phase ( you are so right on that one).

As I said earlyer if your happy with your alts phases out of phase at 120 degrees you should be real happy with your speaker out of phase at 180 degrees. Thats even more noticable then 120 degrees out. But 120 out of sinc, out of wack, on a diferant page on an on is still out of phase.

At one time disc alts were built with a single disc and laminations. We got past that. Now its time for the next step.

Some one has mentioned. The comercial manufactures wouldn't make 3 phase generators if they were not eficient.

They are eficient. There 3 phase power goes to a 3 phase load. Each phase is in sinc with each phase. And as far as the car alt.

These alts are spining at thousands of rpm with coils of 12 gage wire that ony have 3 or 4 turns each.

Extreamly low DCR and extream rpm and a 200 hp motor to power them. Save the expence of a few million diodes and this alt for this aplication is quite accepable.

As far as a 3 phase generator is concerned it supplies power phase for phase. We are combining 3 difearant phases to become one dc output.

The combining prosses of star looses power through coil wired in sires  that are not in phase with each other. There for the peak output of these out of phase coils can nerver be realized it lost to the 1.73 thing.

I won't design my alts that way. Magnets and copper ain't free. I want them doing all the power there capable of. Some of my tax money goes for wellfair. I refuse to but my magnets and copper on the wellfair roles.

I urge any one to do the same test of these 2 diferant phase configurations and make your own conclusions.

I've had my convictions from the start on this subject and for me they have been confirmed.

In the compairison I even gave the star configuation some help a handicap and it still lost.

If need be I can go ever my reasoning again. I would have to show more examples or try to explain why these diferances. To me its very simple and should be easy to understand.

I'm just not sure why folks are haveing trubles with very simple and obvius diferances.

Heres a couple pix of the test alt.

 

                    JK TAS Jerry

[Flux]

Jerry

Any chance of comparing delta with your connection.

Flux

[oztules]

Me thinks that is a bit rude don't you flux?  Surely you don't wish jerry to wire the star stator up as delta?. We know that outcome......v x .58.......oohhh, you mean wire the jerryrig up as delta?. That would not be rude, that would be interesting if that is what you were alluding to.

I doesn't matter what I think or why I think it, congratulations jerry on your results. If as flux suggests you do the test with the jerry coils as delta, i think it would be interesting. I assume that the results will be the same as jerryrig.

once again well done

...........oztules

I put my round triangular hat somewhere..... hhmmm ........

[Flux]

Yes I meant jerry versus delta.

It's a pity the volts are too low to see what happens into a battery.

It would also be nice to see figures for the 2 windings as star and delta into a 3 phase resistive load ( star or delta) to check that we are equal at the start.

If Jerry has a 2 volt cell it would be nice to compare the three options into a battery load.

Interesting stuff.

Flux

[craig110]

Hi Jerry,

Nice numbers and keep going with your tests! Questioning "rules of thumb" is the first step to big changes.

Craig

 

[spinner]

jerry

as always, the rest of us learn (benefit) each time one of us trys something outside the "rules of thumb"..............no matter the outcome of that try

nice goin'

spinner

[SmoggyTurnip]

Jerry - it would be nice to know what power is being used in the stator in both of these cases.  We know you are getting more out with the Jerry rigging, but do you need to put more in?

[Gary D]

Jerry, your work is always interesting! I'm very glad you had the equipment and took the time from your busy schedule to now place the ball in another's court! For someone that is going with Hugh's 5 phase design, with the rectifiers already on the tower, if funds allowed,

would be great to try the two designs...

 As for my mentioning of only x ammount of power in a given windspeed, it was NOT meant for you! Newcomers coming to this board from ebay and other sites seem to think you can get unlimited power when it just isn't there....

 Great to see your results, hope you go to a bigger real life air core to replace one of the downed units for your store if you see fit(as time permits). Congrats. Gary D.

[Jerry]

Hi Flux.

I acctully was thinking of this while testing the alt last night. I want to test this alt and the 2 stators under several diferant conditions.

I had to jump in and do this first test last night cause i was lick a kid christmas morning.

I have several large 2 volt cels from electric lift trucks (Hysters). So the next test will be into that battery load.

Also was thinking of the delta test VS jerry riged. I think they will be the same or delta may have a slight edge?

This test had simular results as the GM alternater test and as per my #s crunched.

I'm not sugjesting a fullwave brige per coil either. A fullwave  bridge per phase will have the same results.

In the 9 coil 12 magnet alts the 3 coils of one phase are in phase with each other so they can be wired in sires no problem.

The problem starts when we wire coils or coil groups the are out of phase with each other in sires (Star).

The simplest the here is the key words "out of phase".

In 3 phase ac generated and 3 phase ac used no out of phase condition exsists.

But when we are trying to combine these 3 phase together into a common DC how we commbine them is very important. We either do it so they work well together or we do it in a way they are not best compatable.

On to more testing.

                       JK TAS Jerry

[Jerry]

I was able to do the 2 volt battery load test at work this morning. The diferance is not quite as dramatatic but still quite segnificant.

Here are the #s quickly as I need to get back to work.

Star 4.82 watts, delta 5.7 watts , jerry riged 5.72 watts.

I'm thinking that with a 9 coil stator the #s may be even a little better then my 3 coil test alt. There will simply be more coils that could have more sutel diferances.

I havn't crunched the #s but it looks to be around a 20% increase in power.

I think thats worth persuing.

                        JK TAS Jerry

[coldspot]

Great posting- As usual from our man "Jerry" !!!!!!!!

Please keep doing the tests!

THANKS!

Have a GREAT DAY/NIGHT!!!!!!!!!!!!!

[MrBill]

  Jerry    For those of us that came in late ,can you post a link to the wireing diagram and maybe a discription of how the "jerry method" is done..I would like to get the best out of my alts too.

[oztules]

Look at the bright side jerry, as you crank the rpm higher, the difference between the jerryrig and the star will become wider............matching your grim perhaps

........oztules

[Jerry]

Hi Smoggy Turnip.

Yes it does take more power to make more power thats true. It would take more power for an upgrade star wired alt to make the same power as this jerry riged alt. The diferance is the star alt would requier larger magnets and more wire to duplicate the same preformance and there agian thats my point.

If I can make the same power with smaller magnets and less copper I save some money or If I spend the same amount on copper and magnets I can make more power. Pick witch ever one suits you. But don't waist resorces.

                  JK TAS Jerry

[Jerry]

Hi MrBill

If you go back in the sires of diaries there is a diagram of the jerry riged phase wireing.

Its equal to delta. Thats why I've refured to it as delta/jerry riged most of the time.

The diferance is in delta each phase leg end is tied to another phase leg end. Then a conventional 3 phase fullwave bridge rectifier is wire to the 3 tie points made by these 3 phase conection. In jerry riged each phase is treated as its own and seperate single phase. To each phase is conected its own and seperate fullwave bridge rectifier (4 terminals). Then the dc outputs of these bridges is perelelled. 3 + together form the positive and thev 3 - form the negative. This is electricaly equal to having those 3 phases wired in perelell. The rectifiers remove any phase conflicts that exsists with coils out of phase and the small diferences that exist between phases. As so many times stated these phases are out of phase with each other by 120 degrees. They can be at the same place at the same time.

So the only way we can combine the power the out of phase phases make without conflict is to first convert there power to dc then there is no conflict.

                   JK TAS Jerry

[Jerry]

Should have writen. They can't be at the same place at the same time. Thats what my mind ment but not what my finger wrote.

                         Jk TAS Jerry

[Jerry]

A few more thoughts. My goal in this test was to compair what would happen if I had disc alt that was wired star and used the exact same amount of copper wire that is the same weight but smaller gage.

The reason for the smaller gage is because delta has less voltage then star but more amperage. If I wind more turns in delta its voltage can rize to meet that of star however at the sacrifice of amperage.

Again delta is equal to perelell coils and star is equal to sires coils.

If we were talking a single phase compairison to make tings equal

if 2 10 volt coils are wired in sires we get 20 volts if each of these coils makes 10 amps we get 20 volts and 10 amps.

To equal this with 2 perelell coils (delta) we could have 2 20 volt coils at 5 amps each and the result will be as above 20 volts and 10 amps the same.

Now comes 3 phase. 3 coils or sets of coils. For my test I did for simplisity 3 coils 1 per phase.

In the star configuration coils are efectively wire in sires. The problem is the coils are never at 10 volts at the same time the end result of that is 17.3 volts not 20. Were also deriveing power from 2 coils at a time at any give time while 2 coils are making partial power the 3rd is at zero potential. It is an inactive dead beat for that fraction in time when its neighbor is trying to pass current though its lifless body to the recifiers.

In delta/jerry riged the coils have no kinship to there neighboring phases other that there on the same stator getting power from the same rotating magnets but not interfearing with each others work. Contributing to the dc output un impaired by there half dead budy phase next to them. Not passing there hard earned power through a lasy coil that is some what oposed to that current passing through it.

I wanted to use the same amount of wire in delta/jerry riged. Not gage but weight.

It was sugjested to me to be fair to the star configuration Id have to use a few more turns of wire to compensate for star voltage loss The idea was to use a slighly small gage wire then star to get the extra turns but have the same weight but still be of a larger gage than delta/jerry riged.

I did that I gave star the handy cap I even went one better and squeezed the original larger gage wire in. This has a higher amperage potential then the comprimized wire gage for star would have had. Even with the 2 cruches more wire and the larger gage star still did not preform as well. And If I would have use the star comprize wire to get the same weight stars preformance would have even been less.

and if I had used the larger wire at the same weight (100 turns not 116 turns) as delta/jerry riged there again the power would have been even less.

So If I had made the exact pound for pound copairison on wire weight star would be even lower then the #s I've posted. I need to make the 100 turn coil as I sugjested in the begining.

100 turns of 21 gage should be the exactly the same weight of 200 turns of 24 gage. With this exact same pound for pound compairison. There will be an even greater diferance between these 2 phase aragements.

That was my goal. to test these 2 phase aragements on a pound for poud even playing feild without giving either one a cruch or handicap.

See witch one can produce the most power with 7.5 onces of copper wire.

In this case delta/jerry riged had 7.5 onces and star had 8.5 onces and still lost. Star would have lost even worse if I had to used the same amount of copper as delta/jerry riged.

So now I'm still confused why star is refured to as more eficient and I still havn't heared where that 2.7 volts went. Personally I'm convinced it went to never never land never to be heared from again. It was a one way trip.

                    JK TAS Jerry

     

[Jerry]

I tryed to crunch some #s. If I had used the same weight in wire but half the wire size in delta/jerry riged VS star I think the diferance would be as  an example

delta/jerry riged 572 star 468 dif of 104. A test must be done pound for pound to confirm this. But if It works out that way I'd take an extra 104 watts for the same money spent any time.

                    JK TAS Jerry

[Jerry]

I would like to see someone else do the same test I've done here. To make sure and confirm what I'm seeing in these results or hay maybe just show I'm kinda goofey?

Guid lines pound for pound.

                        JK TAS Jerry

[willib]

Jerry i must say , that is one of the coolest alts i have seen.

i really like the see-through stator.

in the pic below , is that part a standard flange ? if so what is it from?

[RobC]

This is one of the most interesting ideas to come across the board in a long time Jerry. I think I could build one of these, I always drew a complete blank on that 3 Phase stuff. Way to go keep it up. RobC  

[Flux]

Jerry

Good stuff, thanks for taking so much trouble. Any chance that you could check a couple of things to try and shed a bit more light on things.

With no load, for star could you measure ac volts between 2 lines and from one line to star point.

For the other winding can you measure volts from one coil on its own and volts between lines when connected in delta.

Also it would be nice if you could get a waveform of line volts open circuit in star and delta.

This is interesting and and it would be good to get all the information while you have it set up.

You have been very fair in the tests and there is some issue at work here that needs investigating.

If may be that the characteristics are different over a speed range but I doubt that you could check that with the little machine.

Flux

[spinner]

jerry:

ws the diagram in your diary that you referred to #hpm 1614?

thanx

spinner

[Jerry]

Hi Willib.

Your looking at the back of a speaker magnet end plate. Being in the audio biz since the 70s and never throwing anything away (may use that some day?).

The standard flange part your refuring to (pipe flange) holds the plexi stator support. The stator is made of 1/4" plexi also. The plexi is 1/4". Just some scrap left of from a custom install at my car stereo store. This littel alt was thrown together out of quike easy stuff.

These come in very small to up to (in my collection) 8" with a vent hole in the center that is easly drill out to shat size. These are very straight in alignment and also the very best steel for magnetic flux cause thats what there made for.

They also make very nice hubs to mount larger diameter discs to and best of all there free.

Check with your local car stereo shop, they just through them out any way. You can collect same size and shape as microwave magnets but in much larger quanities without the bulk of the ovens to get ride of.

The real large discs ceramic magnets can easly be cut into magnet wedges. Not as good as NEOs but free helps make up for that if you want to just learn or make a low powered alt.

And the pipe and bearing stuff is cheap and easy from the big box home improvement stores.

This all allows someone with very limited money, tools, and skills to play with wind power in the home garage.

Lets let every play.

                      JK TAS Jerry

[Jerry]

Hi Flux.

I did the 2 volt battery test yesterday at work. We use 2 volt left truck cells there added to our 2850 LB lift truck battery for our 12 volt RE system and 14 volt amp testing bench system. In our amp test system we have none inductive resistors at 1/2 ohm,1 ohm,2 ohm and 4 ohm These resistors are actuly large banks able to handel thousands of watts. We also use sine wave audio generators and 2 osciliscopes true rms meters and large inductive loads as well.

I took the opertunity to conect the scope to these 2 stators. the surprize was that the wave formslooked the same. It was not a pure sine wave. It looked more like a sine wave that the peak had caved in.

I'll try to get a picture of both stator wave forms. Both in signle coil each and in full phase conection.

                          JK TAS Jerry

[Jerry]

Hi Spinner.

#1614 is a partial veiw of the standard 3 phase diode bridge conection.

#1607 is the diagram of the delta/jerry riged 3 phase 3 diode bridge conection I was refuring to.

A seperate bridge for each phase. Combining the power after the diodes not befor.

Trying to combine the power befor with ac phases that are not compatable only looses power. Combining power after the phases have been changed to dc eliminates the conflict.

                        JK TAS Jerry

[Flux]

Thanks Jerry I think we have got to the root of the issue. Could you please get me those ac voltages with your rms meter.

Simple theory breaks down when the waveform is not a sine wave or at least close to it. Your circuit will manage to make better use of the power in the harmonics than star.

In terms of power out for a given amount of material it may well be that star does come off worse. If you consider power in then the star will be more electrically efficient. Much depends on what you call efficiency. Bang for your bucks may not be the same as power out for mechanical power in.

The machine I did my test with produced virtually a pure sine wave so in the end physical construction may have a serious effect.

Flux

[DanB]

Yes - that's an important test - perhaps I missed it  and its here but if not... it would be very important to know open voltage (With no load) AC at the same rpm for each stator.  'Jerry rigged' and delta should be the same - and if the same load is being used for testing Star as it is for the other two, then it should also be the same.

Very fun stuff Jerry thanks!

[oztules]

"Simple theory breaks down when the waveform is not a sine wave or at least close to it. Your circuit will manage to make better use of the power in the harmonics than star."

The only problem here is that the delta config worked almost the same as jerryrig in his tests (jerry vs delta test).

The slight difference between these two systems, should reasonably reflect the harmonic/circulating losses inherant in the deltaconfig compared to the jerryrig.

If  as the discussion has reflected so far, that star is less prone to these loses (than delta), then this is not an excuse that should be reasonably proffered in defence of stars poor showing. If it were true, then delta should have fared worse than what it did.

I think it also showed that the losses delta is rumoured to catastrophically suffer as a result of the circulating currents, is nowhere near as high as has previously been presupposed.

If what you are getting at is that the non-sinusoidal waveform is detrimental to stars performance, then most alt's built in the bush will suffer from this shortcoming, as gaps are altered, wire packing is inconsistant with perfect flux interference, disk thickness, saturation problems... and the list goes on.

I,m inclined to think that jerry has solved more problems for variable-build alternators than he has created.............this could well be seen as a net gain?

The more bang for the buck is in all probability the best way to go, because of course, if we lose efficiency, we can add a few more inches to the blade ie make it 4'6'' instead of 4'...this is pretty cheap compensation, and may solve the stiffer alt's presentation to the blades.

In Danb's new alternator, his losses were rumoured to be in the 4kw range, almost soley due to the internal resistance. A jerrywound in this case would lower the copper cost, and decrease these losses by a substancial factor. This would also decrease the heat buildup in the stator, allowing a slightly higher furl speed (although what one would do with all those kilowatts i dont know, heating i guess.)

possibilities have been re-opened thanks jerry

[DanB]

Hi Jerry - this is good stuff, it's great you're doing this.  Hopefully something will come of it!

"So I jumped the rpm up to 1600. This coil group is wired star. I used a 4 ohm 400 watt none indutive resistor as load."

I expect testing everything with the same load is good althuogh it might be interesting and show some small differences if you try a couple different loads.  Of course all this gets time consuming..

"I plan on doing a battery load test and look at the rippel and wave form on the osciliscope but that will be in a fewdays."

I expect the wave forms and the amount of ripple in teh rectified output will be about the same.

"I recorded dc amps and volts. Then I removed the star stator and replaced it with the jerry riged delta stator.

Here are my test results.

  Jerry riged volts=4.484                          Star volts=3.890

  Jerry riged amps= 1.11                            Star amps= .96

  Jerry riged watts=4.977                           Star watts= 3.734"

This is interesting.  Im assuming that you measured volts while it was under load - or is that open voltage with no load?  A very important test would be to run all the configurations at the same rpm, and find the open voltage.  They should all be the same - if your testing them with the same load.  If open voltage is different between the star stator and the delta (or jerry rigged) stator then you'd need to adjust your load to make it a fair test.  My guess is you should be close because you compensated for this with the winding but it would be good to double check that.

If your voltage measurments above are for 'open voltage' (and not under load) then there's not enough windings in teh star stator and the load needs be be different to make a fair comparison.  And - when measuring things under load, I think you ought to be measuring voltage and current at the same time.

"Your out of phase sires conected coils will rob you of the power that exists within the same amount of matirial and the same money spent, the same size and the same weight."

no - I dont see how we're llosing power anywhere with a star connection, in fact I think current is flowing more often through the coils (even though overall voltage is a bit lower).  Name our 3 phases 1, 2 and 3, and imagine that phase #1 is at peak voltage.  During this moment, even though our best voltage (between phases 1 and two is only 1.73x that of 1 by itself - the voltage of phase one is also being added to phase 3~!  And phase three might be so low that by itself voltage is below battery (or useful) voltage but when combined with phase 1 its also useful!  So we have current flowing through all the coils, all the time during this moment when in 'jerry rigged' some of your coils have 0 current flowing through them.

"There has been some concern that this jerry riged alt will reqier more blade power."

That wouldn't be my guess...  if its a more powerful alternator then yes.  My guess is its really about the same but probably still a bit less efficient than a star connection but =- perhaps more testing or clarification of the tests youve done would be good.  I should also test this for myself sometime it would be fun.

"Yes it will, I know that. But at the same time it will give you the power of an alt that was built star with larger and more expencive magnets and larger and more expencive copper and every thing else that is required for a larger alt."

I still think its probably the other way around... by rectifying each phase seperately you've un-done one of the main advantages of a 3 phase system which is...  more power from a smaller space.  Other than smoother operation here (less vibration) I don't think you have any advantage over a single phase machine at least from teh point of veiw of your coils.

"This is scalable. So imagin the power you could acctuly get from a 2 KW alt if you wernt wasting its power on out of phase coils that are not phase compatible with each other."

Thats the thing - I dont think any power is wasted.  There's certainly not more heat happening I dont think (unless overall the resistance of a star machine is higher).  Each coil in a Star machine is doing just as much work as each coil in a single phase machine except that it's also able to use conductors from 2 other phases to help carry the current.

"Personaly I don't think we should build ineficient alts with star phases just to make our blades happy."

We don't do it to make the blades happy...

"And Dan B I'm sorry buddy but you said it yourself. The voltage is lower cause the phases are out of phase ( you are so right on that one)."

Yes - that's true - but there is nothing lost.  The 'peak' power of every coil is basicly the same(amount of power generated in each coil as the poles change).  We are not getting twice the voltage when we hook two phases together... but were hooking 3 phases together here.  So a single voltage measurment across 2 leads does not tell the whole story.

"As I said earlyer if your happy with your alts phases out of phase at 120 degrees you should be real happy with your speaker out of phase at 180 degrees."

No - I don't think thats a fair comparison its a completely different thing...  plus - at 180 deg apart they cancel completely.

 "Thats even more noticable then 120 degrees out. But 120 out of sinc, out of wack, on a diferant page on an on is still out of phase."

You should tell Nicola Tesla that he was out of whack ;~)

(actually - I think he was...  but not about this)

"Some one has mentioned. The comercial manufactures wouldn't make 3 phase generators if they were not eficient.

They are eficient. There 3 phase power goes to a 3 phase load. Each phase is in sinc with each phase. And as far as the car alt."

No - a car alternator does things exactly the way we do - a star connection to a bridge rectifier charging a battery and each phase is 120 deg out of phase with its neighbors.

"Extreamly low DCR and extream rpm and a 200 hp motor to power them. Save the expence of a few million diodes and this alt for this aplication is quite accepable."

Perhaps... I dont think so though.  

"As far as a 3 phase generator is concerned it supplies power phase for phase. We are combining 3 difearant phases to become one dc output."

I don't see the difference.

"I urge any one to do the same test of these 2 diferant phase configurations and make your own conclusions."

I may play with this...

Flux suggested wave form could be a factor - I expect thats a possibility but more importantly I think we need to be very sure of this:

If you use the same load to test all three configurations then you must be sure that the open voltage (no load) is the same for all three at teh same rpm.  If you supplied that information allready under 'voltage' and then tested current under load you may have allready disproven your theory.

If Your readings were open voltage, then you didn't have enough turns in star or the airgap was wider or somethiing... because a good test here requires that both have the same open voltage at teh same speed.

If at 1600 rpm you were getting 4.48V open voltage in Jerry rigged, and 1.11 amps under load

And in Star you had 3.8 Volts open and .96 amps under load.

Then you need to compensate for the voltage difference.  And I think you could safely say in this case that if .96 is to 3.8V in star, then you should probably expect about 1.22amps in star @ 4.48V with the same load (which is more powerful slightly).  Of course... you'd have to add a bit more winding so resistance would be a touch higher so hard to say...  

The test is very interesting but I think your conclusion is skewed if open voltage is not the same for both stators at the same rpm - and wave form may well be part of the reason for that because you made the best bet at the number of windings.

I've had my convictions from the start on this subject and for me they have been confirmed.

In the compairison I even gave the star configuation some help a handicap and it still lost.

If need be I can go ever my reasoning again. I would have to show more examples or try to explain why these diferances. To me its very simple and should be easy to understand.

I'm just not sure why folks are haveing trubles with very simple and obvius dif"erances.

[oztules]

Hi Danb.

We can clarify a few things here

"Here are my test results.

  Jerry riged volts=4.484                          Star volts=3.890

  Jerry riged amps= 1.11                            Star amps= .96

  Jerry riged watts=4.977                           Star watts= 3.734"

This is interesting.  Im assuming that you measured volts while it was under load - or is that open voltage with no load?  A very important test would be to run all the configurations at the same rpm, and find the open voltage.  They should all be the same - if your testing them with the same load.  If open voltage is different between the star stator and the delta (or jerry rigged) stator then you'd need to adjust your load to make it a fair test.  My guess is you should be close because you compensated for this with the winding but it would be good to double check that.

If your voltage measurments above are for 'open voltage' (and not under load) then there's not enough windings in teh star stator and the load needs be be different to make a fair comparison.     "" And - when measuring things under load, I think you ought to be measuring voltage and current at the same time.""

Danb, jerry must be measuring the voltage and current under load conditions.

If W = ExR and E = IxR then   W = I squared x r  

for jerryrig W=4.977 I=1.11 and R=4 ohms.

so W= 1.11 x 1.11 x 4.00 = 4.92 close enought to jerry's watts out

So that is the load voltage not the open voltage

for the star

W= IxIxR = .96x.96x4 = 3.686 close enough to 3.734w as jerry said. So I think that shows what was measured.

So this part becomes a bit ifffy

""If Your readings were open voltage, then you didn't have enough turns in star or the airgap was wider or somethiing... because a good test here requires that both have the same open voltage at teh same speed.

If at 1600 rpm you were getting 4.48V open voltage in Jerry rigged, and 1.11 amps under load

And in Star you had 3.8 Volts open and .96 amps under load.

Then you need to compensate for the voltage difference.  And I think you could safely say in this case that if .96 is to 3.8V in star, then you should probably expect about 1.22amps in star @ 4.48V with the same load (which is more powerful slightly).  Of course... you'd have to add a bit more winding so resistance would be a touch higher so hard to say...  

The test is very interesting but I think your conclusion is skewed if open voltage is not the same for both stators at the same rpm - and wave form may well be part of the reason for that because you made the best bet at the number of windings.""

This part is now redundant but remains true.

"If you use the same load to test all three configurations then you must be sure that the open voltage (no load) is the same for all three at teh same rpm.  If you supplied that information allready under 'voltage' and then tested current under load you may have allready disproven your theory"

Your Observations:

"no - I dont see how we're llosing power anywhere with a star connection, in fact I think current is flowing more often through the coils (even though overall voltage is a bit lower).  Name our 3 phases 1, 2 and 3, and imagine that phase #1 is at peak voltage.  During this moment, even though our best voltage (between phases 1 and two is only 1.73x that of 1 by itself - the voltage of phase one is also being added to phase 3~!  And phase three might be so low that by itself voltage is below battery (or useful) voltage but when combined with phase 1 its also useful!  So we have current flowing through all the coils, all the time during this moment when in 'jerry rigged' some of your coils have 0 current flowing through them'

I think Flux would be the person to better explain what is going on than me, however, current flowing in all parts of the circuit all the time does not necessarily mean more power, or more efficiency. In a previous post, Flux indicated to me that the delta and star are the same in efficiency terms in a three phase environment. It's when the results pop out of that environment into our dc world, that some efficiency in delta is lost. Jerry has somehow helped to reduce this, but at the coil level, their efficiency is equivelent. Current flowing in a circuit does NOT mean power......Sacrilidge you say?  Not really, Watts are volts X current but ONLY.. when "in phase". Current without in phase voltage is wattless current It still flows and it is real, but if no accompanying in phase voltage exists in that frame of reference....no power.

Sounds silly I know, but if you read about Power Factor, the dimensions of watless power become very real.

An example of this is if I run a 100vac induction motor , the line sees an inductor. This will create a phase change and we call this it's power factor.

This may mean for example that for  10 amps supplied, this motor may use 7 amps at 100 volts The line remains the same, the motor uses (7x100) watts, the other 3 amps supplied is not lost.

It does not turn up as heat in the motor, the motor only uses the 700 watts, but the distribution system still has to supply the 10 amps so that the motor can use 7. This is a vector "thinggy" Because work can only be done with in phase volts and current, the vector of the line current at 10 amps 100v at the new power factor angle can only supply seven in phase amps to tha motor.

Had we only supplied seven amps, the motor could only have taken 4.9 amps at that power factor times the line voltage. So here we have real current gone missing in action, we can't measure it, the power company can't measure it, but they still have to generate the 10 amps, just to sell you the 7 amps. Thats why they require 100% power factor if they can get it, other wise they must generate excess amps for no monetry return. (The other three will turn up as heat in the distribution lines and fuses) They will also have to supply heavier gauge grid to get this power to us....bit like stars thicker wire? yes Star has to generate higher current in order to get a vector sum that will produce current into the out of phase legs. So at no time shen star is generating, can the voltages or currents generated in star legs get anywhere near the unvectored sum of the two coils that make up the leg.

Danb, yes there will be other legs that can share these currents, but always the vector sum will be 1/1.73 time whatever the available voltage or current is. Just because current is flowing, does not mean it is useable. You need 1.73 times more current flowing in any other leg to vectorally add what our current leg may be.

In star, one in phase coil plus an out of phase coil, can only generate the inphase voltage plus the vector of the out of phase voltage. That remainder voltage that is out of phase cannot be utilised, even though it exists. (jerry that's where your other voltage has "gone". It hasn't gone at all, it is just unusable in the other phases ie. the vector sum of whats left is to low to use inanother phase.

Even if we dont understand how i've poorly explained this, the universe does the vector addition for us, and the amp meter gives us all the available in phase power in the system averaged out over time and reflects this as amps. It does not matter what we think is happening in any single frame of reference, its the output that adds it up for us.

So having current flowing in all legs all the time or any part of the time is no substitute for amps out.

This post is getting a bit long, but I think you get the gist of it. Some simple math I have done for unloaded volts would say that there is a slight discrepancy in the unloaded volts. will wait for jerry's figures.

hope you can make some sense of this and hopefully someone more worty will explain it better

.............oztules

[oztules]

"Hopefully someone more worty will explain it better "  yes folks I meant worthy

....and yes I can't get the hang of these laptop keyboards....my excuse and I'm sticking to it..............oztules

[Jerry]

 Hi Dan B.

Thanks for taking the time to give such an indepth responce. Its clear that we are both pashionet about are point of veiw.

The good thing about that is we're highly motivated to confirm our points of veiw.

This in the end will bring even more and solid knoledge to the table.

So let the discusion continue. Further testing will also help greatly.

In terms of wave form. Both stators seem to exhibit the same wave form.

I will try to get pitures of both and post them in a day or 2.

One of the reasons I've been reluctant to give star a helping hand has to do with its by nature lower voltage do to its out of phase contribution.

Flux has sugjested the 116 turns rather than 100. 116 turns was to give star the help it needed. If you can sugjest another turns count and wire gage I will try that.

I do know that if a single coil produces 10 volts and it has 200 turns you can wind another coil of 100 turns and expect 5 volts.

If you double the wire physical size (circular mills) of the 5 volt coil  it should use the same space as the smaller gage wire.

There for lowering voltage but increasing amperage. But obviusly we can't do this with star it does not have a direct relationship with delta. It behaves diferantly

In order for star to have the same voltage in the same given space we must penelize its amperage ability with smaller wire to compinsate it for its lower voltage.

If we could keep the lager wire then we could also keep the amperage upgrade. However the amperage upgrade is reduced with smaller wire and more turns to bring it to = voltage of delta.

Looks like star has to be nocked back abit in amperage just so it can catch up in voltage and at the same time increase its resistance. Lowering its resistance would be nice but the nessesarey turns woun't fit in the space provided.

So yes I can find a smaller wire gage for star to get its voltage exactly the same space as delta. But this will be at the exspence of less amperage and higher resistance.

Lets say that we do manage to get the voltage the same in star as delta. In star  the coils are wired sires so there amperage does not add but the voltage does (star math 10 + 10 = 17.30).  Now lets say delta coils are 20 volts at 5 amps (3 coils stators) The total output of delta/jerry riged since all coils are wired perelell after there doide bridges is 20 volts at 15 amps (not accounting for diode loss in either phase). Now in star we would have had 10 volt coils with twice the amperage of the delta coils but we had to back the amperage of abit so we could get the voltage up to equal  delta. This caused us a slight amperage loss but it did bring the volage up to deltas level. How much less amperage loss ? lets say 1 amp? That seems fair could be more dano?

Now since star  coils are wired sires ther  amperage does not add but the compensated voltage is 20 just like delta but there amperage is less. 20 volts 9 amps. Delta 20 volts 15 amps. The tests seem to indicat this also.

Now I know from my test the loss is not quite that bad but there was a loss both in the GM car alt and my littel test alt.

I wound the test alt delta/jerry riged with 24 gage. I wound the star with 21 gage. These are half sizes of each other. The only other wire I have in between is 22 gage.

Its bigger than 24 gage and smaller than 21 gage. Should be able to get the turns count up to apoint that star will make the same voltage as delta/jerry riged.

I've mesured volts for both stators loaded and unloaded at the same rpm. I've looked at the wave form of both stators on the scope.

"Then you need to compensate for voltage diferance for the voltage diferance" there that word compensate with more turns ( crutch, handicap ect' ect).

If I said to you I'll build an alt just like your star alt but I'll wire it delta/jerry riged and use the same weight in copper and same space and make more power I won't ask for any handicap. I'll simply do it on the merrits of a new coil configuration then what? Is that not fair or am I cheating? Or do I need to degrade my alt somehow so it won't make more power then yours?

"If your voltage measurements for above are for "open voltag" They are under load thats why there is a wattage figure there. Open  voltage would not have a wattage reading.

"you ot to be measure voltage and amperage at the same time" Yep thats how its done 2 meters one doin volts ones doin amps. I have pictures.

"Flux said wave form could be a factor" Good Idea cause I'm sure the GM alt has a completly diferant wave form and I got the same results.

You said peak power of each coil is the same (star or delt/jerry riged). This is true. How dose the peak power of the delta/jerry riged coil get to the batteries what path does it take (direcrt)

How does the peak power of the star coil get to the battery (not direct but through another coil that is not at its peak power point. Through a coil that is sometimes in the coconfiguration is only a coil of wire (a resistore) Part of the time thats what all coils are beit star or delta/jerry riged. The diferance is delta/jerry riged coils don't have to send there power through a non productive coil (resistor) There power gets to the battery direct no degridation. No 1.73 thing but 100% of what they produce.

And yes delta, jerry riged and star are all 3 phase. They all have coil goups that produce power at 120% intervals.

In my mind there are 2 things that degrads stars preformance. 2 coils at any given time in sires (amps don't doulble) voltage loss do to coils out of phase (1.73 thing).

Tell me wich one or both of those is wrong?

Ok my offerings for tonight. More testing and debate to follow.

Thanks Dan lets pick it apart.

                          JK TAS Jerry