Star Delta Jerry

[Flux]

I have re done the tests with the 6 coil stator now that I have potted it and made it mechanically stable.

At Jerry's request these tests are into a 24v battery held between 24.5 and 26.1 volts and are the same for all three tests.

Cut in for all three cases was 405 rpm into 24.5v.

I have to stress that these results may not apply to other types of machine but I am fairly confident that they apply to a normal 12/9 set up if you make it with sensible dimensions.

I make absolutely no claim as to what happens with serious variation from this or with any form of iron core.

Results are   speed  watts out   watts in

Star

1.     450   12.8   15.3
   
2.     487.5  32.25  41.45
   
3.      525   58.5   78.56
   
4.     562.5  84.5  120.5
   
5.     600   118.8   173.5
   
6.     637.5   161.8   238.5
   
7.      675    195    298.4
   
   

Delta

1.    450    8.92    18.36
   
2.    487.5  28.38   43.1
   
3.    525    61.1    89.28
   
4.    562.5  94.6    142
   
5.    600    143.5   214.2
   
6.    637.5  184     277.5
   
   

Jerry

1.    450   9.56    12.24
   
2.    487.5  29     36.5
   
3.    525   53.5    71.4
   
4.    562.5  84.8   114.8
   
5.    600    120   169.4
   
6.    637.5  156.6   227.6
   
7.    675    193    287
   
   

Flux

[Flux]

In both cases where I mentioned 24.5v I meant 25.5V for cut in and voltage variation.

Flux

[DanB]

Thankyou Flux!  Its all very interesting.  Delta is surprisingly inefficient in that setup..

[Flux]

Yes Dan I have always found delta to be worse than star. I don't think it is as bad as it can be in this case because the coils are not that close to the magnets.

When you close the delta the circulating current is obvious as a vibration but doesn't show on the torque reading until the rectifier conducts.

I have never been in a position to directly compare star with delta before but the tests I have done with the same machine have always shown the low efficiency at the bottom end in delta compared with star.

I believe you used delta for early machines, did you notice any difference. But to be honest you would never see it with wind power without sophisticated wind measuring equipment.

Jerry comes out well in these tests but not worth the extra 3 leads in my opinion.

One thing is certain, I would never do a star delta change now, star Jerry can be done with a simple relay and seems more efficient.

Flux

[windstuffnow]

  Great work flux!  I just did a quick comparison on the 600 rpm range, 68% star - 66% delta - and 71% jerry rigged.  Quite interesting indeed!   Could lead to other configurations that would actually be a little better the old standards....

.

[kitno455]

Flux, i am a little confused about the winding used here. can you give us turns per coil?

allan

[Flux]

Yes it's the original from the first post.

100 turns of 0.9mm for star and 170 turns of 0.71mm for delta Jerry.

Same weight of wire, same volume and same cut in speed.

Flux

[Gary D]

Just a thought, on an actual flying unit, wouldn't the blades go to somewhere between 2.5 and 3 possibly 4 times the cut in speed? The results of a higher top end might show more of difference in effeciencies? Tests can only be done at what the operator feels is a safe speed though.... Thanks Flux for the testing! And yes if you are like many here, it would indeed take 2 three wire drop leads to keep the rectifiers at a convienient ground level....  Gary D.

[zubbly]

hi Flux!

my hat is off to you. that is some excellent testing.

you have a 6 coil stator. 2 coils per phase (the 2 coils in series i think).

just hoping and wondering if that series connection bewteen the 2 coils is accessable.

i would just love to see what would happen if the same test was repeated, but using that "extended delta" connection i have mentioned a few times.

i know the values would be off considerably compared to the other connections, but am just curious how the eff would work out.

if the connection is not accessible because of the potting, we will find out if it is feasible to use in another test sometime.

have fun!

zubbly

[Jerry]

Hi Flux.

Thanks for taking the time to do these tests. Was the test alt conected directly to the battery or was it prossed through a charge controler of some type. If it were through a charge controler would it be posable to do the test with batter only.

If this is to much troble I understand. My concern is not having the alt directly connected to the battery post, there may be some influence that restrics the results I'm getteing.

Take for example your test #1. It shows quite a loss in either delta or jerry riged with the rpm remaining the same.

I know the input power would be more fore delta or jerry riged but your RPM has not changed. When I run the same test in 3 direant alt delta/jerry riged had conciderably more power than star and my rpm did not change either.

I know that the power in would be conciderably more in delta/jerry riged. I did not measure that but my RPM was constant just like yours.

If we are both testing the same way what accounts for my #s being diferant than yours. I know its not keeping the voltage to a narrow band. That would not explane our diferances.

I've started the small wind tunnel test. I have a large covered patio area. Under the patio area I have an 8ft wooden picnick table. I positiond it directly below a rafter and placed a 2"X6" between the table and the rafter and another at the other end of this 8ft long table.

I have a 2.5 hp tredmill motor mounted to the 2"X6" at one end and another 2"X6" at the other end. That end will hold the test alts.

I have a 4000 watt 120 volt Variac and a 50 amps 120 volt conection to power the variac.

The tred mill motor is 130 volt dc, 3050 rpm. I've mounted a 24" comercial air mover blade to the motor. I test fired the fan today and at about 33% on the variac the thing sounded like a Cesna getting ready to take off and the wind was quite strong.

I should be able to vary the wind from 0 to ? mph.

I lost my littel RS hand helt anamometer so I'm trying to repair my Oregon Scientific that went down when my tower went down.

If that dosent work I'll just use the referance #s on the variac dial to compair and use fore ?mph referance.

I'll try to test all 3 alts. The load will be a 2 volt battery cause these test alts have very low voltage outputs.

If things go right I should have some test #s tommorow?

                          JK TAS Jerry

[Flux]

Sorry Zubbly,

To meet Jerry's turns/wire size requirement, the windings only have one coil. Three are used for star and 3 for delta.

I have all leads available if you think the test can be done with dis-similar coils.

I will try to find the connection in your files but I may have to ask you for a sketch. I thought it was a closed delta with an open bit on each arm in phase with the rest.

Flux

[Flux]

Jerry

The thing was connected directly to the battery. The charge controller leads were taken from the battery terminals by sparate wires.

One thing I did find was that at a constant speed the battery volts have a very significant effect on the current( and everything else).

I tried to do the test at 28v ( battery full charge) but soon found that impossible as the battery volts fell quickly to about 25.5v and the low end part of the tests was being done at different volts for each point and more seriously for the same speed point for the different connections.

By using 25.5v the regulator held volts constant for the different type tests at the same speed but actual volts did climb with load current to 26.1 at 7.5A

If I had very short leads on the regulator the volts would have been constant but I wanted the batteries to be the load with the regulator holding things down because you had asked for a battery load.

I think the reason for your difference in star with the 3 coil alternator is the waveform. I have identical cut in speed, I suspect your star volts are low and the cut in speed is higher, this makes a drastic difference at any one speed.

I really can't comment on the other alternators with iron cores, I have no suitable machine to try those tests.

Keep up the good work , none of this is easy, my first rig was not up to the job, I have had to pot the stator to make it survive the dozens of tests needed to be reasonably exact. Even now it is only a representation of that machine at those voltages and any variation would show different curves, but shouldn't alter the trend for the three connections.

The fact that I can't duplicate your star results doesn't imply that your results are wrong, just that your alternator is different.

Flux

[Lennard]

hi all,  i really want to thank you guys for explaining this in a way that an average person can understand.  a lot of times it goes right over my head.  on this one i understand.  thanks,  Lennard

[SparWeb]

Flux,

The work so far has been very informative.  I'd like to correlate it with my computer models, if you'd agree to share the dimensions of your rotor and stator. (Magnet type & size, Stator size, and spacing between?)

Crunching the numbers a little, the tests illustrate a point that is often made about Delta.  For lack of a better name, I've plotted a "load factor" graph that shows a Delta connection requiring a lot of torque at low RPM.  It's not that your battery doesn't start charging, it's that the turbine driving the alternator has to work harder when the wind isn't strong.

Correct me if I'm wrong, but I infer that if your mill's blades are pitched for a high TSR, then Delta connection will prevent it from starting in low winds.  Am I on the right track there?