GringoInChina wrote:One of the things you may want to put in your water cooling circuit (you prolly already thought about it anyway) is a little 'surge tank', mounted up high, above the motor height and above the pump. It prolly should be vented to atmospheric (and you could use that to fill the system too). With a little air space in it to help purge the air. Maybe have the radiator dump into that and then use another tube coming from the bottom to feed the pump (which you're probably planning on placing low somewhere).
Are your windings as dark as they look in the pictures (very warm motor)?
Yeah, it'll probably just be a section of tube on a T-fitting in-line with the loop in a handy place where I can fill it when the need arises.
Dunno. It's secondhand. But I hear some of the C'lyte's windings are just dark. These seem to be sort of bonded with what's probably a thermal epoxy, so I wouldn't be surprised if that's why they look sort of black-ish.
liveforphysics wrote:Link- A properly designed pump for a closed loop cooling system would have no ability to pump against a head. In otherwords, in a closed loop (zero gravity head), that thermaltake pump could have entirely different performance.
True! Which is why I didn't eBay the Thermaltake on the spot. However, there IS the resistance to flow introduced by the tubing itself, which is determined by the inside diameter, length, and how rough it is inside. I tried getting as close to that as I could with my setup via having the water enter and exit at the same gravitational potential, but, thinking back, the longer section of tube really should have been on the exit, not the entrance. Probably test it again like that, just to see how much of an effect that had...
liveforphysics wrote:This is tough for me to tell you, because I see a lot of work in your project, but you should know before spending more effort and resources on something. I work as an engineer to design cooling systems (and power generation) for datacenter devices. I specialize in the physics of thermodynamics and fluid dynamics. I can tell you this setup is going to be of no benifit to cooling the motor. Energy can only flow when it has a temp difference. The temp difference needed to be transfering any signifigant energy through that amount of insulation from the windings to the water would be so great that the magents would be de-gaussed, and the windings toasted before the cooling would begin to help.
Lul, too late. It was effectively done when the silicone dried.
I'm not expecting it to go through the vinyl tubes. I'm expecting it to go through the metal stator and for the water to take care of it from there. Maybe I should have posted a pic of how it would look between the spokes of the stator...
liveforphysics wrote:Get some 1/8" copper ref tubing. Don't try to do criss-cross nonsense. Don't try to do a bunch of tight bends either. Roll the copper around the outside edges. Just 2 passes on each side will do. Then grind your slot big enough to let that tubeing can pass through the axel slot. Grind a slot of each side if you must. Epoxy the loops to the inside of the stator with thermal epoxy. I always use artic silver for my projects, but it's kinda pricey for large jobs, so cheaper stuff would be better suit this project. You can also make your own. Mix aluminum and/or copper dust into long cure-time epoxy until you can't physically get anymore into it, and it turns into play-dough. Pack it in tightly around the tubes to get as good of contact as possible with the inside of the stator.
I'm still considering doing that, actually. The middle bit of the stator should be fine, since it's contacting the water directly, but heat from the outer edges of the stator (where the windings loop into the next slot; near the side covers) will have to go through about double the iron to be removed from the motor.
It wouldn't be too hard at all, since I'd just leave the middle assembly in place and splice in the copper loops, but I'm curious to see just how much a difference it could have. When I test it, I'll probably have six or more different thermistors on it to map out how much temperature difference there is between the side the water enters on, the side it exits on, the windings for either side, the median point between them, and the axle. Plus I'll probably monitor the temperature of the water going into the hub, coming out of the hub, going into the radiators, and going out of the radiators.
If I figure I'd benefit enough from adding a few passes of copper (which probably won't take much to convince me of), I'll do it. If there's not really very much temperature difference between the middle of the stator and the left and right halves, then...uh...I'll still probably do it anyway, just so the copper tubing I have left won't feel like a total waste.
liveforphysics wrote:A low flowrate setup like that will need to operate off a high temp difference to be transfering a useful amount of energy. That means no vinal.
*shrugs*
I figure if these motors can live totally enclosed with how much power SOME people here are putting through them
, mine should be okay with even crappy water cooling.
Also: More stupid ideas are popping into my head. How much would it cost to get a die that would stamp out stator laminations for a motor, ya figure?
LOL @ 55.2 dBA. Isn't that almost normal talking volume?
Sorry but I couldn't resist!
) and using this:
Word to the wise: Do NOT order that size of Zippy from HobbyKing if they're not in stock and you want them within a reasonable timeframe.
I'll measure it again when I actually get the cooling loop set up, but I'm not sure the long intake on the Thermaltake was having THAT much effect, since the water level in the bowl I was using to feed it was a few inches above the pump. Plus, while the Shur-Flo's was 3/8" ID while the Thermaltake's was 1/4" ID, the Shur-Flo's was about double the length.
