Idea for folks blowing RC ESCs.

Joined
Oct 28, 2008
Messages
14,539
Location
Manhattan Beach, CA, USA
Just a random idea. Cut the shrink wrap wrapper off the ESC. Find or make a little aluminum box the ESC fits into. Make an end plate on the box that has the needed wires pass through it, seal it up with silicone or bulkhead seals (if you're fancy). Throw a 3/8" nipple on each side of the box. Mount a little PC liquid cooling pump plumbed to the nipples so it has a high fluid flow rate through the ESC. Put the end on the box, sealed up with some silicone, fill the system with mobil1 oil, synthetic ATF, or if you wana get super fancy, fluorinert (a specially designed electronics cooling fluid).

The high flow rate of fluid should do a pretty good job of distributing the inherent hot spots in the controller into the whole system.

It might be enough to give the RC controllers decent reliability, maybe double the continuous safe current handling, and the total parts cost to give it a try is ~$50.
 
Well, I got a $10 CPU heatsink with fan from Fry's and attaching it to the ESC has given me damn good reliability. Powering the CPU fan is necessary if I want to keep the ESC relatively cool all the time, which can be done with a DC-DC converter. However, I've found that I don't need to power it since I've changed to a higher gearing ratio which has substantially brought down ESC temps (I still have the same top speed and far better hill climbing ability, so go figure).

My Castle Creations Phoenix-HV ESC has lasted over 1.5 years now and is still ticking.

Adding a heatsink, in my opinion, is necessary for a lot of popular configurations because rc airplane ESCs were not designed to sustain high bicycle loads for prolonged periods of time, and the ESC's heatsink is small because there's a huge demand for the lightest and smallest package possible in the airplane world, which goes against the needs of bikers. Ebikers need a hefty heatsink to help absorb prolonged high loads, which is precisely what a lot of "normal ebike controllers" have.
 
Luke, I like that idea
More b'cause it would be cool to watch every one scratching thier heads wondering if they really could use something like that. :lol:
funny enough I have a motor all drawn up that would also be suitable to be run fully imersed in a cooling bath. Based on Colasus 7kw stators (more than 1 hehe) It will require the magnetts to all be filled flush to eliminate hydrolic drag but with a proper sized shaft & a few orafices we can get some flow through the stators & run wild on the input....any thoughts?

1st I need a controller that will over heat the motor. (my Ice 160 never logs more than 10deg above ambient, But it is never run at partial throttle. I see more heat in the units when I let friends take a spin , but really never HOT. My Ice 100 has gotton pretty warm to the touch (nearing the end of of a long run with batteries getting lower than they should be)
 
Fully submercible pcs have been done since early 2000 you can also use mineral oil (what the pc guys use)

I am still baffled at the need for cooling, i must be an exception to the rule as i have never got my HV160 esc hotter than ~110F according to the data logging. Is this too hot Luke? I would be more than happy to build a plexi box and pop the rc ESC in fill it with non-conductive fluid and pump it through a pc heat exchanger using a Laing pump...everyone of my pcs since 2001 has been watrcooled i aint new to it :)

Your cost estimate needs to be quadrupled to buddy a decent pc pump alone is over the 100 dollar mark nowadays.

KiM
 
I think Luke's idea is not as much the cooling of the controller as a whole as to help distribute the heat evenly throughout it, removing it's hotspots (even though this will warm up other parts of it). It'll result in cooling the whole thing, too, but primarily it'll help remove the heat from the FETs in places on them that don't contact the heatsinks, and help cool the PCB itself, too, for instance, as well as capacitors and the like.


Also, I first read the thread title as "Idea for blowing folks RC ESCs" and thought "but we don't need more ideas for that, it happens often enough as it is!", but had to look just to see what wild thing was within. :lol:
 
Exactly Amberwolf. The goal is uniform temp, something critical to the survival of something that has a component thermal run-a-way mode. Controller could read cool as a cucumber its whole life, and it only takes 1 component to enter thermal run-a-way, and poof! In a split second its toast.
 
you mean something like this? http://www.tcscooling.com/

that is a system for an rc car, but it would be much more useful on a bike as it would be a substantial weight on a car, but not noticeable on your bike. i was just intending on sinking my controller to the aluminium frame its on and adding a fan, then see how she goes from there.
 
Why not run water cooled Inrunners with lowKV and diameters like the Astro - then a simple capilary pumping action could be used to circulate the water through a tiny transmission cooler (like for tiny import cars in city traffic)? Could even plum your own fancy rad attached to the front stem, good windflow.

-Mike
PS: I'm only half kidding!
 
Thats awesome. I'm gonna use the Perfluorohexane formulation (boiling at 56c) for a two phase cooling system on my x5. just inject a tiny stream into it and let it boil off! Side benefit? Perfluorohexane is a solvent, for pearly clean stators!
 
NorCalTuna said:
Thats awesome. I'm gonna use the Perfluorohexane formulation (boiling at 56c) for a two phase cooling system on my x5. just inject a tiny stream into it and let it boil off! Side benefit? Perfluorohexane is a solvent, for pearly clean stators!


If you re-wound your x5 to use 200C rated windings, you could not only decrease the heating amount by ~50%, but you could also use plain water for evaporative cooling. :) Much higher latent heat of vaporization with water would let you run a much smaller tank, with refills available everywhere. :)

But you're certianly right about perfluorohexane being able to keep the motor at a nice cool temp through pretty much any amount of heating if you could just keep dribbling it onto the stator. Also safe to breathe (as I'm sure you knew).
 
The most insane part is this is real:

http://www.youtube.com/watch?v=2NmU7VKd3VA

Perfluorohexane is medically distributed under many names and even sold in cosmetics- a 58 buck per ounce "aging serum"

which brings us to the real problem.... the ridiculous, eye popping cost of the stuff. 250gm (about 83cc?) for 255

A liter of fluorinert is much better, 385, but still too expensive to boil off!
http://www.parallax-tech.com/fluorine.htm

so... where can I get my stator rewound then?
 
If this stuff is so expensive, it would be useful to build a heatpipe-like coolingsystem with it.
Catch up the vapor, cool it down (outside the hub) and reuse it again.

BTW: hobbyking introduced a new 240A watercooled ESC. I hope, they learned from the monster2000 desaster.
http://www.hobbyking.com/hobbyking/store/uh_viewItem.asp?idProduct=12442

-Olaf
 
liveforphysics said:
Exactly Amberwolf. The goal is uniform temp, something critical to the survival of something that has a component thermal run-a-way mode. Controller could read cool as a cucumber its whole life, and it only takes 1 component to enter thermal run-a-way, and poof! In a split second its toast.

This is good for those who have this risk.

But, assuming you have phase currents under control through sufficient limiting and/or sufficient gearing and/or sufficient winding resistance and/or sufficiently low operating voltage, then "thermal runaway" shouldn't be an issue as it hasn't been for me and $10 costs a whole lot less than $100+. (Plus without the bulkiness of a pump and what-not)
 
The thermal problems are probably going to occur (if they do) under conditions where the drive is operating in it's non-ideal range, and so not able to keep cool via the planned parameter controls. If you're running at partial throttle a lot (especially in the heat), either because traffic is too slow and/or stop-n-go, or because the battery is low and you dont' want to draw too much current from it and cause it to sag below LVC, or some other similar condition, then the controller would heat up more than under it's designed-to-be-normal conditions.

Additionally, when current-limiting kicks in to protect the motor or battery during hills or start-from-stop conditions, it causes similar partial-throttle conditions that can quickly heat the controller's weak spots (FETs, primarily, but also capacitors).

At that point, having something that can at least spread the heat around to remove hotspots would be really nice, and could save a controller that would have blown otherwise, at least for short bursts of heat (which seems to be what kills them, from my interpretations of various threads describing the problems I've read).
 
Back
Top