Go to Otherpower.com Home Page Go to Forcefield Shopping Cart Go to Wondermagnet.com Home Page
Front Page - [Homebrewed Electricity-- (wind) (solar) (hydro) (steam) (controls) (storage) (mechanical)] - Classifieds - Site News
Everything - Newbies - [Remote Living-- (housing) (heat) (light) (water)] - Reviews - Diaries - Our Products
Stator heat dissipation discussion


By sdscott, Section Mechanical
Posted on Thu Aug 30, 2007 at 06:13:46 PM MST
for Air-Core alternators.

I am starting a new thread as the resent "Stator overheat problem" Post got a bit long,  and it is interesting to have peoples views and ideas about stator construction for `Air-Core' alternators and the inherent heat problem associated with typical resin-based potting.
Wouldn't it be great to have a readily available compound for potting that would set rigid, with no metallic qualities, and dissipate heat faster than the copper can produce it! Well such a compound exists, but I have no idea on supply and cost.
I do a far bit of auto repair, and if you have ever seen a 'blower-motor resister pack' on any late model car, you will agree, the ceramic wafer that the trace-resistors are etched on is the perfect material for casting our air-core stators.
The ceramic is very strong (I cannot brake a 6inch long 3/32nd wafer of this with my hands... barely bends), it has no noticeable expansion when heated with a torch, and I have a hard time repairing a bad solder joint on these as the temperature of my 1200 watt soldering gun cannot keep-up with the heat conduction and dissipation of the ceramic board.   Thoughts?
Cheers!

Keven L wrote;
Thermal conductivity values of fiberglass and epoxy are quite low, viewed here at (http://www.engineeringtoolbox.com/thermal-conductivity-d_429.html )0.04 and 0.35 respectively.
The fiberglass used is this rotor was acting as an insulator and holding all the heat in the coils causing them to cook.  I believe a mortar mix utilizing a very very fine sand would provide the workability and thermal characteristics needed for the coils.  It would not be subject to eddy currents and is somewhat workable after curing.  It also has a thermal conductivity of 1.73 - that is 5 times greater then epoxy and 43 times better then fiberglass.  This alone should help eliminate overheating.  Mortar has an added benefit of being so cheap and easily obtained.  
Racing brake rotors use vein pumps to force air over the surface of the rotor and could be implemented to apply limited cooling as well.  Let me know your thoughts.

Stator heat dissipation discussion | 23 comments (23 topical)

Re: Stator heat dissipation discussion (3.00 / 0) (#1)
by sdscott on Thu Aug 30, 2007 at 12:14:19 PM MST

Kevin, I have been using triple-ought silica-sand in my molds, and it does seem to work well.  A potting mixture of 2.5~3 parts silica, 2 parts talc, and 1 part resin has noticeably better heat transfer (using primitive testing methods of a hot-plate under a cured blank and your hand on top, and measure time-to-temp). It also mixes easier than talc alone because of the aggregate, and hardens with a rock-like solidity.  I do not advise mixing in chopped fiberglass as it will not gain you much more in strength and tends to make the mixture more viscous.  I do use the fiber-glass mat on either side of the stator.  I also always paint the finished stator with a couple of coats of epoxy-paint... good practice?

The draw back is; it is hard on your drill bits!  So I've been putting pugs in the mold where I need the holes to be (wooden-dowels of the same diameter as desired hole), after the mold cures I drill the dowels out. (Check my posts for more information.




Re: Stator heat dissipation discussion (3.00 / 0) (#2)
by bob g on Thu Aug 30, 2007 at 12:27:44 PM MST

seems like this topic was covered in some depth over on the awea forum a few years ago, and they were mixing carborundum with the resin

from what i remember the carborundum has a very high heat transfer rate and is readily available.

and yes i am sure it would be very hard on drill bits as well :)

bob g



Re: Stator heat dissipation discussion (3.00 / 0) (#3)
by sdscott on Thu Aug 30, 2007 at 12:49:12 PM MST

What site is awea?

Sodium may be a better powder additive than talc, as it has a heat-conductivity equal to Iron.  silica/sodium/resin ...?

[ Parent ]



Re: Stator heat dissipation discussion (3.00 / 0) (#4)
by Flux on Thu Aug 30, 2007 at 01:19:07 PM MST

I think you may have more than a little problems if you really mean Sodium.

Carborundum may be ok, alumina and marble flower should also work. Any filler will be better than resin alone. All you can hope to do is increase the effective surface area.

I can't see any of this stuff making enough difference to let you run at the crazy 40% efficiency many of you are aiming for. Even motor conversions will not stand running at this efficiency on a continuous rating. Much of this all comes down to the effective duty cycle compared with the intended rating. If you call it a 1kW machine and it does that for short periods under very variable conditions it may be fine. Put it on another site with consistent wind without turbulence and keep the same winding at 1kW for 10 mins and it will fry. Also the effect during furling is critical. All my machines reduce power as they go hard into furl, I seem to suspect that some here are holding up the power level or even continuing to rise with increasing wind speed. That causes real problems.

Flux

[ Parent ]



Re: Stator heat dissipation discussion (3.00 / 0) (#5)
by sdscott on Thu Aug 30, 2007 at 05:31:02 PM MST

Flux, please expand your thoughts on Sodium... I seem to remember high-end gas engines using Calcium/Sodium filled valves. Just a thought, and I'm interested in your knowledge, so please elaborate.



Re: Stator heat dissipation discussion (3.00 / 0) (#6)
by Warrior on Thu Aug 30, 2007 at 09:32:03 PM MST

I think the heat issue will never be solved, because IMHO it has nothing to do with the potting material, whether it be resin, sodium, or all the other things that people have mentioned.

The stators are burt because they are overdriven. It's as simple as that. There's no mysterious or magic material that will save it from this.

All that needs to be done is take great care and be sure the furling system is working/designed properly.

Yes, the machine will have less output when furled, but if you already are having high enough winds to make it furl your battery bank is probably full by now and you're just dumping the excess power, so why have it make more power if it's going to dump?

Also, I've seen "professional" made turbines like Proven for example burn out, and these have laminated stators which can take serious abuse.

As I said in the begining, it's more a matter of protecting the machine in high winds rather than finding some magical material that will remove heat from the stator fast enough to keep it from toasting .

Warrior
Warrior__ "Why can't Murphy's Law be used to my advantage??"
[ Parent ]



Re: Stator heat dissipation discussion (3.00 / 0) (#8)
by Flux on Fri Aug 31, 2007 at 12:57:16 AM MST

You are absolutely right. At best all they can do with this dream potting stuff is gain
a few % increase in output for a given size of machine.

Low speed machines are always very large compared with the normal industrial speed ones and in many cases the sizes here are a fraction of a continuously industrially rated machine. To squeeze such a lot out of a small size, it means working at class H or above and inevitably even then the things are very intermittently rated and if it was not for the nature of wind power they would all be fried in hours.

There is obviously a significant cooling from the wind but I suspect that there is nowhere near as much wind around these alternators as many imagine. At the low rotational speeds conventional fans within the alternator are far from effective.

Perhaps someone might try moving the alternator back a few feet behind the centre of the blades where there may be a lot more wind, especially during furling. This would need a bearing at the back of the prop and an extension shaft but would not be difficult.

Simpler cheaper and easier be a bit more generous in the design and make it furl properly. I have adopted the electronic converter approach that lets me run at sensible efficiency so I really can't give that much advice from practical experience about squeezing the maximum from these poorly matched schemes.

For battery charging it is perfectly true that in times of high wind things are nearly always dumping and if it furls early it doesn't matter. If you want power for other than battery charging you can match heaters far more effectively without needing to work at constant volts with all the associated stator loss problems.

You can't get a quart out of a pint pot.
Flux

[ Parent ]



Re: Stator heat dissipation discussion (3.00 / 0) (#20)
by Kevin L on Fri Aug 31, 2007 at 02:59:18 PM MST

Flux,

There is plently of air to carry away the small amount of heat generated in the coils.  The problem is the medium between the coils and the air is a poor conductor of heat.  Fiberglass only conducts slightly more then 1/10 of what epoxy will.  If the air is to carry away the maximum amount of heat the stator must approch a temperature near what the coils reach and distribute this heat over a larger area for maximum heat disapation.  I dont think the idea is to vastly increase power production, but simply stator preservation from stormy nights.  You can furl and disconnect the load, but why not design better properties in from the start.  Mortar is strong when cured properly and would provide this advantage.  Read post below. (Post #19)

Kevin L

Value and judge each moment for its merits as you shall not pass this way again.


[ Parent ]



Re: Stator heat dissipation discussion (3.00 / 0) (#9)
by Flux on Fri Aug 31, 2007 at 01:04:52 AM MST

I am not that familiar with sodium but I seem to remember from my school days that it is pretty unstable, oxidises very rapidly in air and does spectacular things in the presence of moisture.

If you can tame its physical properties you still have the almost impossible problem that it is a metal and a good electrical conductor. You will build yourself a nice eddy current brake and possibly an effective liquid metal magnetically operated pump. I can't see how it could be any better than any other metal, all of them would need to be finely laminated and sodium would be more of a problem in this respect than any of the common metals.

Use in engines is a totally different issue.

Flux

[ Parent ]



Re: Stator heat dissipation discussion (3.00 / 0) (#11)
by sdscott on Fri Aug 31, 2007 at 05:40:03 AM MST

Yes, Sodium is an alkali metal, but it is not so scary an element as you suggest.  Most people recognize sodium in its combined salts form, and it can be very active in some of these forms.  However, it is very abundant in nature, hence cheep and relatively stable even in the presence of moisture (although I believe we have existing problems with moisture in our current potted stators that need attention), and its resistivity is twice that of Aluminum (which is already a poor conductor and used successfully in stator potting solutions for cooling).  Use in engines is exactly for pulling temperature away from the valve faces, quickly, to the stems for dissipation.

I am not suggesting that this collaboration of thoughts be used for squeezing maximum power from our mills.  I am interested in the continued and ongoing efforts to refine our processes to better the overall design qualities of the home-brew RE generators.  We are not all trying to run our machines full-out. Case in point; DanB's latest burnout... everyone can agree that he is defining most 'best practices', and his stator may have survived with a few mods to the design.

It is an ever-evolving system, and I do not believe we have done all we can do, yet :)   I do appreciate your vast knowledge and understanding, Flux.

Other changes may need to be discovered...

[ Parent ]



Re: Stator heat dissipation discussion (3.00 / 0) (#15)
by Tritium on Fri Aug 31, 2007 at 08:51:18 AM MST

Sodium compounds abound in nature NOT pure sodium metal which is extremely reactive. Sodium metal if dropped into H20 liberates large amounts of hydrogen and heat. Read that BOOM! Been there done that in science class. Teacher almost had a heart attack when a student dropped a walnut size chunk in a large glass graduated cylinder full of water. Thank goodness no one was hurt by the explosion which took out the celing. The teachers paddle did hurt a bit though I suspect.

Thurmond

[ Parent ]



Re: Stator heat dissipation discussion (3.00 / 0) (#17)
by sdscott on Fri Aug 31, 2007 at 10:56:09 AM MST

Okay, Sodium metal will not be good.  What else might be?

[ Parent ]


POOF! (3.00 / 0) (#10)
by wdyasq on Fri Aug 31, 2007 at 04:54:58 AM MST

Sodium filled valves have a hollow stem 'half filled' with sodium which melts and carries some of the vale heat to the valve guides and others areas.  It was a common on exhaust valves of some engines.

I never worked in a shop where one of the valves was cut open. I understand it was a bit more excitement to cut one open than one wanted. Some claimed it was more excitement than was survivable.

Ron
Adventure is just bad planning." -- Roald Amundsen
[ Parent ]



Re: Stator heat dissipation discussion (3.00 / 0) (#7)
by MaryAlana on Thu Aug 30, 2007 at 10:17:53 PM MST

Readily available and used for potting electronics and transformers http://www.mgchemicals.com/products/832ht.html 375ML is under $30 so not overly expensive if you do some things to cut down on the amount of resin used like keep the centers of the coils open. I have been doing some destructive testing on a coil I wound with it. No numbers on the failure point yet because my power supply trips the crowbar circuit and I haven't been brave enough to connect it to a battery to see how hot it gets before it melts.



Re: Stator heat dissipation discussion (3.00 / 0) (#13)
by snowcrow on Fri Aug 31, 2007 at 07:18:09 AM MST

Great find MaryAlana!!!

I also found a heat conductive potting there as well:

 http://www.mgchemicals.com/products/832tc.html

Maybe a combination of the two epoxies would work well. If the coil wires were coated with the high epoxy during the winding process and then cased in the heat conductive epoxy, or the other way around, it could be a wonderful thing! Maybe even use carbon fiber instead of fiberglass? It would be worth doing a few test coils to found out!

Blessings, Snow Crow

[ Parent ]



Re: Stator heat dissipation discussion (3.00 / 0) (#14)
by sdscott on Fri Aug 31, 2007 at 08:26:46 AM MST

Looks good... however 2liters is $110can.  Still doable thou...

[ Parent ]


Re: Stator heat dissipation discussion (3.00 / 0) (#16)
by MaryAlana on Fri Aug 31, 2007 at 10:12:03 AM MST

I have some of the TC here, working temp isn't very high compared to the HT

[ Parent ]


Re: Stator heat dissipation discussion (3.00 / 0) (#18)
by Sundog on Fri Aug 31, 2007 at 11:28:48 AM MST

VW's used sodium filled exhaust valves for awhile.  They didn't really work any better than "normal" valves, so they quit using them.  Also, it really made a mess of things when the head of the valve would break off and rattle around in the engine.  

I'd hate to think that a reaction of the sodium with moisture while inside that magnesium crankcase would do....  

[ Parent ]



Re: Stator heat dissipation discussion (3.00 / 0) (#12)
by sdscott on Fri Aug 31, 2007 at 06:12:07 AM MST

I have also focused on getting more air-flow around the stator and magnets to aid in cooling.  A couple of things I've done that I believe can assist is; Drill a 2-3inch hole in the center of the blade-hub to allow the oncoming wind to be 'sucked' into the alternator core, and expelled by the fan action of the magnet-rotors.  Also, don't pot the magnet rotors. Instead, use industrial adhesives to weld the magnets to the rotors.  This will provide more surface area for better cooling of the magnets and better fan-action for moving the injected air around the stator.










Re: Stator heat dissipation discussion (3.00 / 0) (#19)
by Kevin L on Fri Aug 31, 2007 at 02:45:43 PM MST

OK just found the new post.
  1. ) Sodium reacts with water as stated and is not good.  (Kinda fun though)
  2. ) Salts dissolve in Water and would need a protective coating.  If it is epoxy or resin it would reduce the desired heat dissipation.  Probably not feasible.
I don't think we need to vastly increase the heat dissipation, but I think all can agree that the thermal insulation properties of fiberglass are exactly the opposite of what a coil needs, that is liberate the heat from the coils.

Epoxy isn't horrible, as some heat transfer does take place, but a greatly increased heat transfer to the atmosphere will reduce the internal temperature of the coils and prevent the wire from failing on those occasions when the stator is overdriven.  Mortar mixed with fine sand would do exactly this.  Metal sleeves could be inserted during molding for the mounting holes, with hard rubber bushings applied to reduce vibrations.  The stator could transfer 5 times more heat through a given thickness, and would not fall apart if overheating did occur like the fiberglass stator that we viewed.  It is interesting that the fiberglass stator photos did show that most of the heat buildup was directly around the coils.  The black charred fiberglass was directly above the coils thus most heat transfer is around this general region.  If mortar were used weight saving could be had by reducing the size of the stator to the area needed for mechanical strength of the unit, as weight may be the primary disadvantage of using mortar.  (Hole in center and mortar or epoxy tabs for mounting.)  

I'm driving to Florida from Ohio tonight for vacation and wont return for two weeks. Taking the kids to Disney World. (Do they pay for promotions on Otherpower?) Wish I could contribute more to the tread, as I value the opinions many of you have on this subject.

Kevin L

Value and judge each moment for its merits as you shall not pass this way again.


[ Parent ]



Re: Stator heat dissipation discussion (3.00 / 0) (#21)
by scoraigwind on Sun Sep 02, 2007 at 01:27:33 AM MST

As usual I come a bit late to the debate here, but I find that Flux has been talking a lot of sense again.

As far as heat conductivity goes I do think it is important to use plenty of talcum powder in the resin mix.  This helps to prevent the casting from heating as it sets and and crackingas it cools.  But more importantly it helps it stay cool in operation.

I know that this turbine design has a reputation for overdoing it, and as the designer I must take some responsibility, and make my apologies.  The solution is to reduce the weight of the tail and also to ensure that the tower is truly vertical.  Even a very small tilt in the tower can have a big effect on the operation of this design becasue of its very small angle on the tail hinge.  I am now moving to using a 20 degree angle on the tail and keeping the tail lighter.

Furling behaviour does depend on blade speed, and it will furl better when stalling (running slowly) whereas once it begines to run fast it will tend to seek the wind and run faster, creating a vicious circle.  I have never seen this get out of control with such a large offset (8 inches) on a 12 footer but I have seen it with a 6 inch offset.

I am inclined to agree that 1000 watts output continuously is a bit ambitious although 1000 watt surges would not be a problem. To get higher power out of these machines it would be very helpful to change the alternator voltage for different windspeeds.  Flux does this with his electronics.  The machine we built in France last month did it using a grid tie inverter (windy boy).  You can also do star/delta switching but it's not very satisfactory because the change is too large - a smoother change is what you really need.

If what you are after is a reliable supply of battery charge then my suggestion is to make sure you furl early and keep the power down in stronger winds.  There is little to be gained from high power is strong winds when the batteries are already full, but plenty of things that can go wrong with a hot stator, fast blades and everything working toward the limit.

If you are after high power in those high winds for whatever reason, then it makes sense as Dan says to use bigger magnets and make a more efficient alternator.  Of course this will stall pretty heavily as the wind increases, so the blades will not be working at their best.  Or to talk with Flux about using a booster circuit between the turbine and the battery in low winds, so that it is more efficient all round.  But introducing this sort of electronics has its own cost - especially in terms of simplicity and reliability.

It depends really on whether you want high windspeed operation to be high power or just better survival.  I think that's the first decision to make.

Hugh Piggott http://www.scoraigwind.co.uk



Re: Stator heat dissipation discussion (3.00 / 0) (#22)
by blueEnergy on Wed Sep 05, 2007 at 05:57:56 PM MST

Thanks to Flux, Hugh and all the contributors on this subject.  We're talking all your comments into consideration.

My gut tells me that we need to just focus on achieving earlier, more reliable furling to reduce power output in high winds.  Our systems are for battery charging so the extra energy that comes in storms is of little value to us.  Keeping our systems low-maintenance and operational in remote locations is of value to us.

Hugh, can you elaborate on your new steeper angle for the tail vane, or point me to somewhere where you've discussed this.

Best,
Mathias Craig
Executive Director
--
blueEnergy
www.blueenergygroup.org
[ Parent ]



Re: Stator heat dissipation discussion (3.00 / 0) (#23)
by Kevin L on Sat Sep 15, 2007 at 07:33:01 AM MST

scoraigwind,

I disagree, while talc powder does have good thermal conductivity, when mixed with epoxy the increase in overall thermal conductivity will be minimal.  The best method to reduce overheating of the stator is to find a better heat conductor to tranfer the heat to the air.  Of course furling would provide some protection, but as we have seen many times on this board, overheating of the coils is still occuring during storm conditions during full furl.  Using a meduim with a low thermal transfer rate such as Epoxy (0.35 W/m K) or Fiberglass Board (0.04 W/m K) is inviting the prospect of overheated coils.  Motar cement(1.73 W/m K) is a significantly better conductor of heat, and since heat is the primary cause of coil failure, it does makes sense to look at the possibility to use mortar as an epoxy replacement.

Kevin L

Value and judge each moment for its merits as you shall not pass this way again.

[ Parent ]



Stator heat dissipation discussion | 23 comments (23 topical)
Display: Sort:
Menu
· create account
· How to use the board
· FAQs
· search the board
· Google search the board

Login
Make a new account
Username:
Password:

Total Views
  141 Scoop users have viewed this posting.

Related Links
· http://www .engineeringtoolbox.com/thermal-conductivity-d_429.html
· Also by sdscott

Powered by Scoop
You must be a registered user to post here. It's easy and free, and the link is on the upper right side of your page.
All trademarks and copyrights on this page are owned by their respective companies. Postings are owned by the poster, but may be deleted or moved at the ADMIN's sole discretion. The Rest © 2009 Forcefield.
You can Email the board ADMIN here. PLEASE include the username you signed up with!