Home
JAQForum Ver 20.06
Log In or Join  
Active Topics
Local Time 20:18 28 Mar 2024 Privacy Policy
Jump to

Notice. New forum software under development. It's going to miss a few functions and look a bit ugly for a while, but I'm working on it full time now as the old forum was too unstable. Couple days, all good. If you notice any issues, please contact me.

Forum Index : Windmills : Staggered stator connections

     Page 1 of 3    
Author Message
Gizmo

Admin Group

Joined: 05/06/2004
Location: Australia
Posts: 5004
Posted: 11:23am 03 Oct 2007
Copy link to clipboard 
Print this post

I'm planning to put up a dual stator windmill this weekend, and was contemplating how to configure the stators.
I was going to connect both stators the same way, both as 7 poles in series, 3 phase, 80 series wire. This would give a low cut in speed. Then I though it would be better to configure the windings into different coil counts, ie one series of 6 poles, another of 5 poles, etc.
After looking at a few options, I've come up with the following. I have two stators, each with 14 poles per phase. As the windmill speeds up I want the strings of coils to cut in at different speeds, remembering at the lowest speed there is less power so I want to draw less power from the mill, and at higher speed there is more power so I can increase the load.
The lowest speed cut in will be a string of 6 poles, this should supply a couple of amps under 100 RPM. Then as the windmill speeds up, if the turbine is making the power to overcome the 6 string load, a string of 5 poles will start to make power as it reaches battery voltage. After that I have two strings of 4 poles in parallel, starting to make some good watts now and the wind has picked up. Lastly there are three strings of 3 poles, for max power in very strong winds.
Each of the 4 different configurations will need its own 3 phase rectifier. So I end up with 4 power sources each reaching cut in speed at different RPM's.
This is a non linear load, as the wind increases the load increases exponentially, sort of thing.

What do you think? I know the Air-X windmills use two windings of different coil counts, one for low speed and one for high speed, so I think it should work.

Glenn


The best time to plant a tree was twenty years ago, the second best time is right now.
JAQ
 
vasi

Guru

Joined: 23/03/2007
Location: Romania
Posts: 1697
Posted: 11:33pm 03 Oct 2007
Copy link to clipboard 
Print this post

Hi Glenn,
If you want, make a sketch. Something similar with Star/Delta switch based on rpm? With a microcontroller and relays? It neeeds to be placed on the mast... is a big project. But is nice, I just realised, we don't have yet a TheBackShed star/delta switch based on picaxe!


Off topic!
Glenn,
your forum can have something like rss feed, or atom feed? I would like to put recent titles (posts) on my site to guide romanian people on your site ....Edited by vasi 2007-10-05
Hobbit name: Togo Toadfoot of Frogmorton
Elvish name: Mablung Miriel
Beyound Arduino Lang
 
Gizmo

Admin Group

Joined: 05/06/2004
Location: Australia
Posts: 5004
Posted: 07:40am 04 Oct 2007
Copy link to clipboard 
Print this post

Hi Vasi.

No switching, just a simple case of different cut in speeds.



The outputs of the four brigde rectifiers are all connected in parallel. In light winds only the 6 and possibly the 5 pole groups are generating enough voltage to provide a charging current, as the wind picks up the other groups come online as their voltages rise.

What it means is that instead of a single RPM = Charging Voltage, I have 4 different RPM's to reach charging voltage.

Glenn
The best time to plant a tree was twenty years ago, the second best time is right now.
JAQ
 
vasi

Guru

Joined: 23/03/2007
Location: Romania
Posts: 1697
Posted: 07:53am 04 Oct 2007
Copy link to clipboard 
Print this post

But the 6 poles in series continues to generate and can be to much voltage? If is not too big, the battery is not in danger, but then, 3x3poles poles don't generate enough volts. I think ...Edited by vasi 2007-10-05
Hobbit name: Togo Toadfoot of Frogmorton
Elvish name: Mablung Miriel
Beyound Arduino Lang
 
Gizmo

Admin Group

Joined: 05/06/2004
Location: Australia
Posts: 5004
Posted: 08:50am 04 Oct 2007
Copy link to clipboard 
Print this post

Yes thats right, but the 6 poles can only make 1/3 the amps of the 3 X 3 poles. Each pole of the 6 pole will be working harder as the windmill speeds up, but its own internal resistance wont let it exceed the wire capability, ie, it wont burn out. A single coil in a 80 series is only good for a few amps, its when we parallel them up we get the high currents.

The voltage wont go much above battery voltage because the relatively high resistance of the 6 poles in series cant compete agains the low internal resistance of a battery.

The F&P's are air cooled and I've never heard of one suffering damage from overload, tough little buggers. Good thing about a windmill is the faster the wind the harder the coils works, but because the wind is faster its getting better cooling around the coils.

All fun stuff.

The site doesnt have a RSS feed or such. If I want to link to a story I just copy the address from the address box in the browser.

Glenn
The best time to plant a tree was twenty years ago, the second best time is right now.
JAQ
 
Gill

Senior Member

Joined: 11/11/2006
Location: Australia
Posts: 669
Posted: 12:27pm 04 Oct 2007
Copy link to clipboard 
Print this post

Seems like a good idea Glenn.
It steps the loading which is like MPPT.

I have gone down a different track. Vasi's suggestion is what I have been working on for the last 6 months. I have gone a little bit more in that I am using dual x 7 phase F&P's. This then gives me gens in series and parallel as well as Star and Delta making a total of 5 configurations. This will provide from HiVolts for low cut in to All In Parallel for HiAmps. I have been toying with using one gen only to reduce load at startup as your line of thinking has suggested but I'll wait to see performance figures of test run first. (only PICAXE programming to step into that extra setting.)

I don't think you'll be utilizing the full potential of all coils all the time. Isn't that against the RULES???
[That's a joke fellas]

The one thing needed for both our attempts to suck the marrow out of winds of all strengths is the elusive variable pitch prop. I have a sketch of one that is controlled by the masthead picaxe. This gets away from the centrifugal force control that most go for. Also gets away from furling as picaxe feathers or reverse thrusts prop at excess rpm. Unfortunately I have no machining resources.

Good idea Glenn give it a whirl. At the speed you work you'll have yours flying in no time.
was working fine... til the smoke got out.
Cheers Gill _Cairns, FNQ
 
Gill

Senior Member

Joined: 11/11/2006
Location: Australia
Posts: 669
Posted: 10:32am 06 Oct 2007
Copy link to clipboard 
Print this post

Glenn,
How are you going to space the arrangements on both stators? I can't see any balanced arrangement there, but maybe you have a trick up your sleeve?

I am also doing the exercise bike thingie at the moment so am keen to try 2 of your four. Thinking your 6 and 4 x 2 as the alternative to 7 x 2 (cutout switch on one bank of 7).
Please keep us posted on progress. Ta.
was working fine... til the smoke got out.
Cheers Gill _Cairns, FNQ
 
herbnz

Senior Member

Joined: 18/02/2007
Location: New Zealand
Posts: 258
Posted: 07:34pm 06 Oct 2007
Copy link to clipboard 
Print this post

Hi Glen
I feel you are on to a very practical arrangement. The properties of the Fp is that the 6 poles will generate current to a value that is about 200 ampturns then allow more speed to bring in the 5 poles. The six poles will still produce their current but will no be added to by the 5poles.Wereas if you only had 6 poles the mill would start its run away as more input power is applied.
Note it is the ampturns in stator that oppose the flux in rotor that limits current in fp machines when these are equal we have no more flux.
Herb
 
KiwiJohn
Guru

Joined: 01/12/2005
Location: New Zealand
Posts: 691
Posted: 07:49pm 06 Oct 2007
Copy link to clipboard 
Print this post

  Gill said  
I am also doing the exercise bike thingie at the moment so am keen to try 2 of your four. Thinking your 6 and 4 x 2 as the alternative to 7 x 2 (cutout switch on one bank of 7).


Gill, I made an exercycle with an F&P and early on realised the F&P was capable of much greater output than I could input so I cut half the stator away. This instantly halved the cogging drag and made subsequent experimentation quite easy.
 
Gill

Senior Member

Joined: 11/11/2006
Location: Australia
Posts: 669
Posted: 04:44am 07 Oct 2007
Copy link to clipboard 
Print this post

John,
Cutting away the unused half and getting less cogging is a plus. On the other hand, an exercycle needs to be able to have the load variable for a given rpm.
It was an earlier description of your cycle that gave me the idea of switching in or out a heavier load.
With Glenn's I could run 6 in series then switch in parallel an extra 4 in series, then another 4 for a short burst.

Here's how I see Glenn's assorted coil arrangement's performance:



I see it's key features being that it's output power is a closer match to the winds power (velocity^3) than a single linear generator. Also that it needs no external control.

Unfortunately this once again brings me back to the point of prop selection being a compromise.
was working fine... til the smoke got out.
Cheers Gill _Cairns, FNQ
 
KiwiJohn
Guru

Joined: 01/12/2005
Location: New Zealand
Posts: 691
Posted: 06:23am 07 Oct 2007
Copy link to clipboard 
Print this post

Dont worry Gill half a F&P is plenty to load me to exhaustion!

In my case I was not so much interested in exercise but rather in producing a practical 'pedal generator' to operate a HF radio rig.
 
Trev

Guru

Joined: 15/07/2006
Location: Australia
Posts: 639
Posted: 11:05am 07 Oct 2007
Copy link to clipboard 
Print this post

Hi Glenn,
That is very smart thinking, outside the square as some say. Would love to see how it goes.

Currently I have my double wired in 2 pole series, each motor rectified seperately and then at the bottom of the tower can connect in series (low wind - & is efectivly 4 poles) or parallel (high wind).

With the common low wind here I think I was better of with the 2 motors in 7 pole series, though a little dissapointing in the higher wind that happens twice a year.

Trev @ drivebynature.com
 
Gizmo

Admin Group

Joined: 05/06/2004
Location: Australia
Posts: 5004
Posted: 11:19am 07 Oct 2007
Copy link to clipboard 
Print this post

Thanks for the feedback guys.

I've put the mill up with the staggered dual stator, added an entry to my diary here http://www.thebackshed.com/windmill/forum1/forum_posts.asp?T ID=751&PN=1.

Its working!

Glenn
The best time to plant a tree was twenty years ago, the second best time is right now.
JAQ
 
Warpspeed
Guru

Joined: 09/08/2007
Location: Australia
Posts: 4406
Posted: 09:13pm 07 Oct 2007
Copy link to clipboard 
Print this post

Technically, the best solution would be a continuously variable switching power supply to convert the rectified dc wind machine voltage to the required system voltage.

The simplest design would probably be a buck converter, so the rectified wind machine voltage should be arranged to be as high as possible. (within reason!)

The duty cycle of the switching supply would then reduce this higher dc voltage to a lower voltage, while at the same time increasing the output current.

Some simple software could then monitor the output current from the switching regulator, and constantly adjust the PWM duty cycle to keep the whole thing operating at the maximum possible efficiency over a very wide wind speed range.

The way to control the duty cycle would be for the software to measure and record the instantaneous output current, then make a very small change to the PWM duty cycle in one direction.

If the output current has increased, keep that change. If the output current has decreased, make the next change to the PWM duty cycle in the opposite direction. So the software slowly ramps the PWM duty cycle up and down by a very small amount, seeking to hold the output current right at the efficiency peak.

This all assumes there is a battery to more or less keep the system output voltage fairly constant. It is just a case of monitoring the battery charging current and keeping that at the highest possible value for any given wind speed.

It should make an interesting project for someone. I am a retired power electronics design engineer, but am too lazy to do all this myself. Living in the suburbs, I do not have a wind machine to test it with, or try out ideas on, so it would be difficult here to fully develop.

But if anyone wants to have a go at this, I will be happy to assist in the design. It should not be too difficult, I have built similar things before.Edited by Warpspeed 2007-10-09
Cheers,  Tony.
 
Gizmo

Admin Group

Joined: 05/06/2004
Location: Australia
Posts: 5004
Posted: 09:40pm 07 Oct 2007
Copy link to clipboard 
Print this post

Sounds like an interresting project Warpspeed, I might look into that. Its something a PicAxe could handle, you can set its PWM output and it continues to run in the background while the processor performs other instructions. Will give it a go. Could use some help in the Mosfet and inductor selection/design. While its easy to get a Mosfet switch an inductor, its a lot harder to make it last under extremes of load and voltage without going BANG!

Glenn
The best time to plant a tree was twenty years ago, the second best time is right now.
JAQ
 
Warpspeed
Guru

Joined: 09/08/2007
Location: Australia
Posts: 4406
Posted: 09:58pm 07 Oct 2007
Copy link to clipboard 
Print this post

No problem with the MOSFET power switching part, or the inductor design, that is right up my alley.

The measurement of the dc battery charging current in the output does not need to be extremely accurate, the software is only hunting for a maximum. Offset errors, or thermal drift are not going to cause a problem, so something fairly simple should be quite o/k.

How about a linear hall effect device with a five volt output driving the a/d on the PIC? That would provide complete dc isolation for the current sensing, is simple, and should do the job.

The MOSFET gate can be driven through a gate driver chip, straight from the PIC, and does not need to be isolated. The whole thing could be fairly simple.

Cheers,  Tony.
 
Gill

Senior Member

Joined: 11/11/2006
Location: Australia
Posts: 669
Posted: 10:54am 08 Oct 2007
Copy link to clipboard 
Print this post

It would be nice to develop a MPPT Controller('Maximum Power Point Tracking' for those not up with the jargon). I have searched other Forum to find much discussion by armchair experts but never a schematic let alone a DIY Project. Commercial wind units are available(USA) for a price but their use does not appear to be as wide spread as the solar units.

I'm not familiar with buck converters or gate driver's so can't help there but I can cheer from the sidelines
Edited by Gill 2007-10-09
was working fine... til the smoke got out.
Cheers Gill _Cairns, FNQ
 
herbnz

Senior Member

Joined: 18/02/2007
Location: New Zealand
Posts: 258
Posted: 06:35pm 08 Oct 2007
Copy link to clipboard 
Print this post

2MPPT controller would definatly be a plus. It would be best to aim for the highest possible input voltage and step it down. and the use of DC is necessary because of the high frequency AC. The main limitation of FP is the definate current limit when the stator Ampturns approach equal to the Rotor Ampturns. Usuing a high voltage DC would allow less current / coil and more power. also the windmill can be sited further from the batteries.
By the way Gill your graphs would show a flaterning off as this point is approached 200 ampturns is my experiance ie if 80 series has 60 turns per coil max current any series connectin 200/60 = 3.3amps in glens case The 6 branch would stop 3.3 amps then 5 branch would start adding its 3.3 amps as mill speeds up then thnext branch 2// would add 6.6 then 3// would 9.9 amps atotal possible output 19.8 amps.

Herb
 
Warpspeed
Guru

Joined: 09/08/2007
Location: Australia
Posts: 4406
Posted: 09:47pm 08 Oct 2007
Copy link to clipboard 
Print this post

Highest possible input voltage will always be achieved with the wind machine completely unloaded, which is not really what is desired.

What is required is maximum possible output power. If the output voltage has a battery to stabilize the output voltage, maximum power will always coincide with maximum battery charging current. Current is always a much easier parameter to measure than power.

I just cannot start another full time development project right now. I am developing completely from scratch, a 15 Kw RF induction heating furnace for smelting metals for home metal casting.

There is a huge amount of interest in this induction heating project and I have only just started. That thread on another Forum has now had 19,000 hits, and I am falling behind on work with it as it is.
http://www.cnczone.com/forums/showthread.php?t=13545

But....

I have the MPPT hardware pretty well figured out in my head for a practical controller. The software algorithm for it is a different problem altogether, that needs a lot more thought.

If there are enough interested people we could start a fresh thread, and kick around a few ideas. Any takers ?

It would be easy to post a proposed schematic, describe how parts of it should work, and give a few design equations so that others could adapt it to their own particular needs.
Cheers,  Tony.
 
herbnz

Senior Member

Joined: 18/02/2007
Location: New Zealand
Posts: 258
Posted: 04:45am 09 Oct 2007
Copy link to clipboard 
Print this post

Hi Tony This needs a new topic I for one want to be involved
I was not meaning to control on voltage we would go for max battery current. What I was saying was the FP is very prone to current limit if we are stepping down a higher voltage to battery volts the current in the stator would be less therefore more power before current limit it will also be necessary to use capacitors to reduce current peaks as FP can not deliver.
Getting higher voltage from FP however is really governed by speed as increasing turns only brings in current limlting sooner. 200 ampereturns.
however speed increase increases frequency and iron losses so around we go hence reason i am looking at iron powder.
I would start new topic but subject "max power controller" Or max power point tracker ??
Herb
Edited by herbnz 2007-10-10
 
     Page 1 of 3    
Print this page
© JAQ Software 2024