Andy Williams Stator

[Capt Slog]

What I hear, I forget. What I see, I remember. What I do, I understand.

-- Confucius

So to get a little closer to understanding, I built this. I was then able to look at STAR vs DELTA

I'll leave you to look back into my previous diaries if you're interested enough for the details, but this is a mock up of a stator to go in my dual 12 pole gen.  I'd looked all round for a piece of plastic that I could afford to waste to mount the coils on, and then I suddenly thought of my mum's old records.

The coils are 55 turns and mounted around track #8.  The disc is seured using a couple of wood screws, each one had a spring at the back of the disc to make it easy to level.

Untill I did this little exercise, I didn't realise how hard the gen would be to turn when wired in delta.  Maybe I was doing something wrong, but it certainly cleared up which way I was going to wire the final version.  I live in a low wind area, and what I do get tend to come a bit gusty.  

I feel it would be better to have the turbine turning easily most of the time and then accelerating to cut in rather than missing most of the gust waiting for it to start turning.  So I'm going for star.

I've come quite a bit further since I took this picture, I'm almost ready to cast, so more to follow.

[dinges]

Strange.

My generators, both axial fluxes and motor conversions, don't becomes easier or harder to turn in either star or delta. I suspect that you have circulating currents, because either 1) not all coils have an equal amount of turns or 2) you made a wiring error.

You could also settle for 'open delta' : that way, you rectify each phase separately with a bridge rectifier and then parallel the DC outputs.

If I were you I'd build it in such a way that all 6 outputs of the phases (3 phases with each 2 connections) are available at the outside. That way, you could later decide for star, delta or open-delta. Could even add an automatic switch in the future for changeover, if you'd like. Builders of most 3 phase induction motors do the same: in the wiring box one can wire the motor for 220V delta or 380V star (I live in Europe); see the drawing in the bottom-left:

http://www.anotherpower.com/gallery/dinges/scan_angus_3phase_connection_resized

It's what I have done with my small axial fluxes:

http://www.anotherpower.com/gallery/album46/stator_mount_back_resized

The 6 bolts are the outputs of the stator. In another picture, you can see the wires to connect the stator for either star or delta:

http://www.anotherpower.com/gallery/album46/finished_gen_backside_resized

Doing it this way gives you the option later, when the mill is flying, to play with parameters to optimize it for your particular situation/setup.

[wdyasq]

A 78rpm record is usually a bit more stiff. I don't know you can stay with Andy Williams.

Ron

[Capt Slog]

Obviously I couldn't see an error, but I'm prepared to accept that there was one there!

I connected the 'end' of one phase to the 'start' of the next and so on.  And put the rectifiers on the corners.  However, I'm confused every time I see the delta layout, because to me it always looks like the whole thing is just one big short circuit.  From this perspective, I wasn't exactly surprised to find that it didn't turn as well as in star.

As I thought I'd read this somewhere, I did a search just now, this is from the FAQ at Scoraig wind.

Q. I have made an alternator and wired it delta  as described in your brake drum book (yes its a bit old but I like the drum type) a-f  b-d  c-e and I notice that it becomes slightly harder to turn when turning it by hand on the work bench when no load applied to it. when I connect it star it shows no increased resistance to turning . if I break a connection say b-d  and connect a volt meter between them it shows a small voltage, is this normal or are the windings not electrically balanced?

A. Yes this is the main disadvantage of using delta connections.  With a hand-built alternator it is very unlikely that the phases will be perfectly balanced in voltage and phase angle (timing), so there are some parasitic currents around the delta itself.  This can make it hard to start up.  Once the rectifier starts to work there will be third harmonic currents as well.  So delta is less efficient than star.

As I've already joined the three tails of the phases for star, I'm not sure if I can go back a stage.  I don't think I'd seen 'open delta' before, but now I've had a look it could be the way to go.

[Capt Slog]

Yeh wdyasq, but some of those 78s are actually worth something!

[dinges]

The FAQ is right, and it has been mentioned on this board before too: try to make the coils as identical as possible, i.e. the same number of turns.

That way, in delta, you won't have troubles with circulating currents.

Perhaps the fact that, in my axial fluxes, I notice no difference in star and delta (when unloaded) as I turn the rotor, is proof that I did a good job winding the coils...  Fact is I paid great care when winding to get the same number of turns for each coil.

I remember now that in my latest motor conversion (the 130W) it -DID- become hard to turn in delta;  this may have been due to either my magnetic poles being too wide (and thus introducing 3rd harmonics in the wave shape) or that the (original) coils in the motor are not all identical. So, as a solution, I rectified each phase separately and combine the DC outputs in parallel.

I still stand by my point that it would be wise to have both ends of each phase available to the external world; that way one has a few options to choose from when finally deciding on stator configuration (star; delta; open delta; rectifying & paralleling individual phases). I understand it's no longer an option for you though, as the design is 'committed' already

I realize now I made a mistake in my previous response; rectifying individual phases is probably NOT what I called 'open delta'; see here for a brief overview:

http://www.elec-toolbox.com/usefulinfo/xfmr-3ph.htm

[Flux]

I don't like this use of the term open delta for the connection using a rectifier for each phase.

Open delta has a specific meaning and is best left to describe what it has meant for many years. There seems no generally accepted name for the other option and it seems reasonable to describe it a Jerry connection here where most people understand what is meant.

I hope this makes things clear. Star,delta and open delta would use a 3 phase rectifier.

Jerry connection has each phase ac connections to a single phase bridge rectifier and the dc connections are paralleled.

If the waveform is completely sinusoidal and the voltages and phase angles are equal then on no load there will be no circulating current with delta. Invariably there is some lack of balance and almost invariably there is some harmonics present on the waveform so you will have a circulating current that will cause drag below cut in.

Beyond cut in the waveform distortion of the rectifier will result in harmonics that can't be avoided and a 3 rd harmonic at least will circulate within the delta even if it was ok on no load.

Open delta is useful for instrumentation and a few light duties but it has no application for a power producing alternator.

Jerry connection doesn't suffer from circulating currents below cut in, its operation beyond cut in is a bit complex.

Flux

[willib]

I agree with Flux and Peter , there should be no difference in drag on the alt ,whether its in star or delta, before cutin.

the fact that in delta it looks shorted is just that ' it looks shorted '

but electromagnetically , if it is balanced , you will feel no difference in either configuration

 sure your coils are wired right?

[Flux]

I didn't intend to give the impression that there will be no drag in delta below cut in.

Very few alternators will be completely balanced and generally there will be harmonics in the waveform. In most cases there will be noticeable drag but it will be nothing like the drag from a short circuit, or a load such as a lamp.

Flux

[willib]

I did

the minigen 1 had no noticable change between open circuit star , and delta .

[Flux]

Yes it can happen but not everyone will be that lucky especially with coils stuck on with tape. Still maybe he has made a connection mistake, depends on how bad the drag is that he is complaining of.

The effect is likely to be greater with large machines and low resistance.

Normally you can even feel drag when paralleling coils on one phase on machines that are reasonably well built. Without toothed cores to distribute the flux it is difficult to get perfect balance.

Flux

[Capt Slog]

I saw a diagram somewhere else of the so-called Jerry connection, and merely because of it's shape I assumed that this was what was meant by 'open delta'.

I had a look at my final coils last night when I got home (yes I do most of this typing at work!) and even though I had made the common conection for a star set-up, I still had sufficient wire to be able to cut this out and then bring out six ends.  So this is the way I'll go and then be able to make the final choice when its put together.