Re-wind of a turnigy 80/100 (Now-tutorial w/Video)

[Jeremy Harris]

If silicone will do the job then that has to be the best option, as at least you could still get the motor apart again if you had to. It's probably well worth doing so research to see how effective it might be.

I'm also intrigued by the 6 tooth wind, as it offers the possibility of lowering the total winding resistance. I guess the limiting factor has to be core saturation - just how many amp turns will a stator tooth take before saturating.

Jeremy

[Miles]

It really is a ballancing act of copper & magnets & ellectricity after all is said & done.

I guess the balancing starts with the ratio of Copper to Iron when designing the motor?

Great tutorial! Thanks, Todd.

[Miles]

If silicone will do the job then that has to be the best option, as at least you could still get the motor apart again if you had to. It's probably well worth doing so research to see how effective it might be.

I found this:
http://www.sinus-electronic.de/datenbla ... 31K_EX.pdf
http://www.silicone-solutions.co.uk/The ... h%20V3.pdf

[texaspyro]

I wonder if you can get something that's highly heat conductive, yet still an insulator that could be mixed with a high temperature resin and infused around the windings?

Most of the thermal epoxies for CPUs, etc use aluminum oxide as the filler. The CPU people want like dollars per gram for the stuff. I went to the local ceramics supply store and bought a pound of aluminum oxide for less than $5 and make my own thermal epoxy. You can also get diamond powder (natural or man made) for almost reasonable amounts...

[bobc]

Gents,
I can't help feeling that the best way of conducting the heat out of the stator slots is along the copper windings themselves. Whatever you put into an epoxy matrix will be hobbled by the epoxy content. Al oxide is 50W/mK heat conductivity compared to ~370-400 for copper. Al nitride is a LOT better at 150W/mK. (as suggested diamond is great at 2000W/mK!). the most disappointing stuff is heatsink compound at less than 1W/mK - but this is considerably better than air (which it replaces when used).
I've heard of thermal conductor filled plastics going up to 20W/mK, but these materials' mechanical properties are poor enough to make them practically unusable - otherwise they would have made interesting combined housing/heatsink for some of our procucts at work.
Oh yeah - in the olden days some high power components used to be filled with beryllium oxide - but I have no idea what its thermal properties are - check google - ooh 330W/mK good stuff - hideously poisonous mind you

[Biff]

Interesting idea of making your own thermal epoxy. I wonder if the aluminum oxide or diamond would risk shorting coils.

The standard thing to use is something like this

http://www.epoxies.com/therm.htm

Product 50-3150FR is about as thick as chocolate syrup or ketchup (according to Wikipeida http://en.wikipedia.org/wiki/Viscosity )
Product 50-3100 is about as thick as peanut butter

I think those are the best two products from that supplier, there must be more suppliers of products like that, those kind of epoxies are very common in the motor building business. It is important to get vacuum on the parts when you apply the epoxy, that way all the air is removed to ensure that there is good conduction within the coil itself. If there are air pockets in there you will have local hot spots that can be significantly hotter than the rest of the coil which could cause problems if you are pushing the limits of the motor.

to bobc .. the idea of using the conductor itself to remove the heat is attractive, but the length of the copper and small cross section of the leads makes it infeasible. There are large motors which actually have hollow copper wire, and they pump cooling liquid right through the center. I haven't found any hollow conductors that would be suitable for small applications.


-ryan

[texaspyro]

Aluminum oxide and diamond are two excellent insulators. Unless your diamonds are boron doped blue diamonds... they conduct and are used as pacemaker electrodes.

Another good possibility is silicon carbide... The best possible conductor is isotopically pure C6 diamond... twice the conductivity of natural diamond.

[Biff]

Aluminum oxide and diamond are two excellent insulators. Unless your diamonds are boron doped blue diamonds... they conduct and are used as pacemaker electrodes.

Another good possibility is silicon carbide... The best possible conductor is isotopically pure C6 diamond... twice the conductivity of natural diamond.

Thats neat. Did you ever do any tests to measure the thermal conductivity of the epoxies you mixed?

[bobc]

Thought about this a bit more: How about putting a single copper tube in amongst the windings, zig zag it all the way round the motor & bring it out with the power wires. Pump water through this. Now you will get effective cooling of the windings.
PS the highest thermal conductivity I've seen for a thermally loaded paste has been in single figures (W/mK) (I think that had a load of silver in it) (why is it always expensive stuff....??)

[olaf-lampe]

Thought about this a bit more: How about putting a single copper tube in amongst the windings, zig zag it all the way round the motor & bring it out with the power wires. Pump water through this. Now you will get effective cooling of the windings.
PS the highest thermal conductivity I've seen for a thermally loaded paste has been in single figures (W/mK) (I think that had a load of silver in it) (why is it always expensive stuff....??)

I had a similar idea: put watercooled aluminium plates on the heads of the stator. ( same shape as the lamination ) before you wind it.
Maybe a combination of both is the best way... zigzagging pipes is not as easy as it sounds...
-Olaf

[Jeremy Harris]

... zigzagging pipes is not as easy as it sounds...

Plus it'd probably cause increased eddy current loss and take away winding space that could be used for more copper to lower the winding resistance that would reduce losses.

Jeremy

[Thud]

Active cooling:
Given the design or the motors, I would think it would be easyer to encase the entire unit in a liquid tight enclosure.
Just some attention to details to remove all the "pumping" surfaces & sharp edges that would create drag & a shaft seal on the output. An coolant input through the stator & a few proper designed orafaces for flow...

Again, that would be for a much larger motor than my 7kw rebuilt turnigys

[Jeremy Harris]

Active cooling:
Given the design or the motors, I would think it would be easyer to encase the entire unit in a liquid tight enclosure.
Just some attention to details to remove all the "pumping" surfaces & sharp edges that would create drag & a shaft seal on the output. An coolant input through the stator & a few proper designed orafaces for flow...

Again, that would be for a much larger motor than my 7kw rebuilt turnigys

Good idea, this is exactly what the Prius does. Toyota immerse both motors in the transmission oil and fit a cooling system. It allows them to push pretty high current through windings that don't look like they would normally take it.

It would be easy enough to see the effect of oil pumping/circulation losses, just by running a motor suspended in a tank of light oil. I bet they wouldn't be too high, even leaving the motor standard. If the motor was hooked up to a chain reduction drive and the whole thing immersed in light oil I bet it make for a very reliable drive system. It'd probably be fairly quiet, too, as I bet the oil would damp out a lot of noise.

Jeremy

[bobc]

... zigzagging pipes is not as easy as it sounds...

Plus it'd probably cause increased eddy current loss and take away winding space that could be used for more copper to lower the winding resistance that would reduce losses.

Jeremy

I was thinking pretty small pipes guys.... 1.5mm bore say - in copper that would bend easy enough & shouldn't take too much room.
Thinking about it, it would be better to take 2 pipes round in opposite directions to cancel temperature gradients.
yes it would have to be varnished & terminated with plastic to prevent shorted turn(s)
You don't need a lot of water to cart a LOT of heat away....

PS read this - it's all true
http://www.dansdata.com/goop.htm

[texaspyro]

Did you ever do any tests to measure the thermal conductivity of the epoxies you mixed?

Nope, I used the epoxy to mount temperature sensors (LM35's) to high power (50-100W) LEDs (see viewtopic.php?f=2&t=23033&p=351762&hilit=dalewheat#p354447). I have used similar material to pot up some fairly high power electronics modules.

The thermal input to the sensors is through its leads, not the TO-92 plastic package. I left the leads full length and covered them and the sensor with my thermal epoxy. Seems to work very well.

The sensor readings matched a thermocouple spot welded directly to the surface, so the thermal conductivity is pretty reasonable.

[texaspyro]

On the subject of winding motors and detecting shorted wnidings, etc... I have more than a few Tektronix 1502 and 1503 time domain reflectometers (TDRs). See http://www.prc68.com/I/Tek1502.shtml for some background.

These send a very fast rise time pulse down a wire and wherever there is a short/open/impedance anomaly in the wire, some of that pulse gets reflected back to the TDR where it is detected and displayed on a CRT or LCD screen. They also contain a chart recorder or plotter output module. The resolution of the 1502 on short runs is around 0.6" (the pulse rise time is measured in picoseconds).

I have not tried using one on a motor, but it would probably work rather well. You could certainly compare the various windings for matches and differences.

[phyllis]

I know thas some steam boilers (SAAB, in the 70's I think) have been made with lots of parallell 1mm inner diameter pipe.
But what about galvanic corrosion if it is used for watercooled windings? Anyone know?
Mcmaster has some high conductive copper pipe but even the smallest has 6 mm2 copper cross section (OD 3.18 mm ID 1.55 mm)

Heat pipes are very conductive and you wouldn't need a pump.

Since there is so much open space in the stator (even with a good wind) it should be possible to encapsulate only the stator, in fiberglass or something, and run liquid coolant through it in direct contact with the windings. Then there wouldn't be any friction loss due to liquid between the stator and rotor.

EDIT: anyone know how to calculate roughly the drag between rotor and stator with various viscosities fluid in between?

[Ludo91]

My 80- 100 HTX 180 Kv died toady... loose stock windings didnt mange to take the heat...

So its time to rewind!

I was thinking to to rewind it with a copper tube, something with 1.8mm ext diameter and .5 external diameter so i could pump coolant in it.... Seems some of u had the same Idea, I`ll let u know if I find something intresting my friends

[phyllis]

Condolances! Let me know if you find some suitable tube, and either way how it goes!

[Ludo91]

For sure! anyway it was a pretty expected death since i was brutally abusing of it

edit: I am sending inquries to many producers over the net, how many meters of wire do i need to rewind a htx 80 100 motor assuming 8 turns of single wire? tnks

[Miles]

I was thinking to to rewind it with a copper tube, something with 1.8mm ext diameter and .5 external diameter so i could pump coolant in it.... Seems some of u had the same Idea, I`ll let u know if I find something intresting my friends

http://cgi.ebay.co.uk/COPPER-CAPILLARY- ... d2ffceb7ee Something to experiment with....

[Jeremy Harris]

I was thinking to to rewind it with a copper tube, something with 1.8mm ext diameter and .5 external diameter so i could pump coolant in it.... Seems some of u had the same Idea, I`ll let u know if I find something intresting my friends

http://cgi.ebay.co.uk/COPPER-CAPILLARY- ... d2ffceb7ee Something to experiment with....

Have you looked at the Reynolds number for that tube bore at any realistic flow rate? My guess is that solid copper wire would be more effective at shifting heat, as I doubt that you'd get enough fluid flow through that small a bore to make any significant difference to to the stator temperature.

Jeremy

[Ludo91]

Jeremy I share your concerns about the difficulti in pumpoing fluid in the tubes.
Depending on it I may choose a different winding configuration (i`ll rewind it in its actual config, delta, but I may split each phase in half as regards the fluid circolaton).

Right now I have college tests coming up but as soon as they are over I`ll but the tubes, pperform the tests and post results on here

Miles thnks 4 the link

[texaspyro]

Not to mention the near impossibility of winding a motor without kinking it and blocking the bore...

[Miles]

Not to mention the near impossibility of winding a motor without kinking it and blocking the bore...

Maybe you could fill it with MCP70 alloy before you wind it?