Committed to liquid cooling
 

I have decided, after years of pondering, to commit to finding a liquid cooling solution to my RC vehicles.

I looked seriously into doing this years ago when we used to mod the original Mamba Max's for 6S use.
Back then there were no readily available micro sized fluid pumps that would fit the bill.
It ended up going no where, mostly because of the sheer cost of the endeavor.

Today is a different story.

My financial situation has improved, the pumps seem available, and there is more reason now than ever to complete a functioning system.
My primary goals are a solution that would allow me to put an ESC in an enclosed waterproof container and use it in extremely harsh environments without damage or heat issues.
I want water cooling for the ESC, motor, servo's and be able to use the same system from 1/10 scale to 1/5 scale.
I envision this system being used on SC trucks that are ROAR limited to small motors that could benefit from high levels of cooling.

I have a few unique ideas that I have not seen done yet that I think could make this a very clean and functional proposition.
First would be using the aluminum chassis as a radiator, I mean it is a large hunk of aluminum with air flow across it when the RC is in motion.
For higher capacity cooling needs I could use a traditional fan and radiator setup.
Second would be using purpose built "negatives" of the ESC's heat sink to conduct the temps into the cooling system.
For example removing the fan assembly from a MMM and inserting a cooling sink (chunk of aluminum with liquid cooling passages machined in and sealed for coolant flow) that fits onto the MMM's sink with a fine layer of non conductive thermal paste.
I don't want to negatively affect the warranty, and have a system that is easy to remove for swapping the system to a new vehicle or ESC.

The chassis could be turned into a radiator by attaching a liquid sink that is basically a thin aluminum block with cooling channels machined into it and sealed for coolant flow.
It could be attached to the chassis with either thin thermal tape, thermal adhesive, or bolted down with a thin layer of thermal grease.
For a smooth motor a standard boat cooling coil could be implemented. And for finned motors I could develop a special coil that matches the fins better.
I'm also curios if it's possible to create sealed end bells for brushless motors that could pass the (non conductive) coolant directly through the motor to cool the inside vs the outside.
Servo cooling could be a scaled down version of the chassis sink.



I expect the project to go slowly and take the better part of 3 months to complete.
I have Autocad 2011 software (inventor, mechanical, showcase, vault, ect).
After a month or so I should have around $400 to start piecing it together... though the budget is certainly not limited to $400.
When it's all said and done I would like to be able to put together the entire system for under $200

I really want to pull this off successfully and I'm going to make it happen.
Any comments, ideas, criticisms, or wisdom welcome :yes:

I was looking into this before. I know it's not necessary or even very practical, but I just think it's a neat idea and would be a cool project.

Check out these pumps: http://www.micropumps.co.uk/

oops

If you have coolant flowing internally through the motor your windage losses (or whatever you call them when its not wind turbulence creating drag but liquid) would be huge and the IO would rise astronmically water for example is around 840 times thicker than air, and oil the traditional electrical liquid coolant is even thicker and more draggy.

Most folks stopped using cooling coils in boats when we escaped from brushed motors that had holes in the cans, coils can work if wound with the right pressure to flatten the bottoms nicely and the gaps are filled with copious amounts of heat sink grease but a typical shop bought coil fitted dry is of very little use compared to a water jacket that has water in direct contact with the can. If the fins go across the motor like castles making a water jacket is a sinch slide an "O"ring between the first and second fins and another between the penultimate and last fins then slide a tube from the plumbing store over the "O"rings and there you have a jacket, glue in a couple of bits of tube for inlet and outlet and your done.
or you can get them ready made for Nue, Castle, Leopard and Feiago from offshore electrics.

Instead of dealing with the losses from another pump drawing power, not to mention adding another point of failure, I would use heatpipes. Then again, a more efficient motor (even a 1% or two) can make a huge difference.

fans would be a much simpler way to cool things, and much more effective.

Basically if it's running too hot and needs cooling you are pushing it too far.

People really get hung up on internal combustion technology, i.e. multiple speed trannys and liquid cooling. Electric motors are apples to the internal combustion's orange.

I don't want to burst your bubble, but liquid cooling for electric land based r/c is just very impractical.

Now, it would be very very cool to see. In the end it would be nothing more than a novelty. I for one would like to see you accomplish your goal.

Boats have a constant supply of cool water.....

The chassis idea is interesting. I'm imagining some sort of sandwich chassis with channels for coolant to flow.

If you want to do it just to do it, then by all means git 'r dun. There is no practical use for such a system however.

Nativepaul, thanks for the info I wasn't thinking about just sealing the motor in a sleeve like that.

As for running fluid inside the motor, I think I'll pick up a cheapie throw away and do some testing since I have data logging capability now :-)

BrianG: I thought about heat pipes, but I think a forced flow could have a higher capability.

I'll post up what I've found for pumps, very small, low wattage, and plenty of flow...

-P- I'm not arguing what fans can do :-) but they don't work very well packed full of mud LOL

The idea is to take a non waterproof ESC like the MMM, put it in a sealed foam lined box and not have heat issues.
EDIT: without voiding the warranty.

I know there was a thread somewhere around here where a guy put an MMM in a silicone oil bath in a sealed box in his Summit. Might want to look around for that, I'm pretty sure it worked out and the oil dissipated the heat (but did void warranty). Probably would be a mess if you ever had to take it out of there, but just brainstorming here. Then you could focus the water cooling on the motor.

How do those pumps I linked compare to the others you've found so far?

Either way if your going to dissappate heat you need airflow over a surface, all that energy just doesnt vanish when you have liquid cooling.

Your going to need something to move air over your radiator.. so weather a fan is directly cooling your motor, or cooling the radiator- its still a fan. and still would get covered in mud.

:lol: good job SC, those are the same pumps I was looking at :intello:

They were actually around the last time I looked, but the company was fairly new at that point but the cost of them was INSANE!
They've come down a bunch in price since then.

B16, that was the idea of using the chassis for cooling. No need for a fan in that situation.

What model pump were you thinking? M200P2 or M200P4 look good.

Whoops, my apologies- would help if i read through the whole post:neutral:

http://www.micropumps.co.uk/TCSimage...rge%20Gold.jpg

15x16x22mm plus the fittings, 6v at 1.6w (0.27a) so it could be ran easily off the BEC. Would flow 200ml/min at 2.8psi. ~$65 usd plus shipping.

I'll try to get a few for testing purposes.

Edit: it's the M200P4, but I'll probably get one of the larger series to test as well.

well, git 'r dun:yes:

nitro temp gauges would be a relatively easy way to keep an eye on what you are actually accomplishing. just wrap the loop around the heatsink or the can or servo or whatever.

B16, my fault for making the post SOOO long :tongue:

I have a real nice small temp probe that I use with my fluke meter, I should be able to sneak it into an ESC case or inside a motor that I'll sacrifice for testing the liquid inside theory.

Here's something to look at;

http://www.tcscooling.com/Resources/item2p.gif

http://www.tcscooling.com/index.html

Thanks PBO, I had e-mailed this company back in the day about their products (specifically the pump) and never get a reply from them.
I guess they were changing ownership or something.

I'll shoot them another E-mail as it seems their site has been updated and maybe someone is running the company again.

Sent TCS an e-mail, we'll see if they respond this time.

P.s. I think they are using the same pump....

Just something else to keep in mind if this does work out and you want to improve the system.... a pump capable of moving more liquid volume isn't always better, just look through some high performance engine building forums or 1:1 car forums and you will learn that this is 100% true. If you push the coolant through the motor/engine too quickly, it can't absorb the heat, it just flows through and does very little to actually cool the motor/engine. And considering this is going to be such a tiny system, if you don't get the results you hoped for at first, try to slow the pump down some and see if it does any good. The system boats use is just a simple force fed system, and is rather tunable with water inlet/intake sizes, and adjusting the pump speed essentially does the same thing here, or run some kind of restrictor system.

Sounds like a fun project, would be cool if you managed to pack this into a small and effective enough system to be considered practical.

Yah Krawlin, I actually put a system in my chevelle with a moroso pump that only cycled when the coolant hit 185'f then off again at 160'f. Cured the heating problems I was getting just running a restrictor plate for a t-stat and the standard water pump. It would usually be on for about 15 seconds then off for ~2 minutes at idle. I put the temp switch in a PVC pipe thread adapter from 3/4 to 3/8" in the cylinder head to keep it reading the coolant vs the temp of the head itself. Never got hot running from LA to Vegas in 2-1/2 hours...

Good call on the info though :-)

I like your "reverse heatsink" shaped adapter idea for the MMM. I would like to see that item mounted to the al chassis, then the MMM flipped over and stuck to that. That way you get the thermal mass of the al chassis to draw the heat off the esc. Same could be done with the motor, just make a half round motor cradle that contacts the can plenty and attaches to the chassis. Not sure a servo really needs to be cooled? Maybe just a mount setup that clamps to the al center section that most good servos have?

While water collling seems cool I see it as being too complex and finnicky. I do like BrianG's heatpipe idea, and they sound pretty easy to make. And no moving parts either.

Yah, I guess I could do the reverse heat sink machined into the bottom of a sealed aluminum box and put the box on the chassis.
I'd have to come up with some adapters to make the wires come out the right way...

It's an interesting idea Linc :yes:

Seems like al transfers heat pretty well, and the chassis has a large mass and surface area. Air flow under the chassis must help cool it of? Maybe one of our more nitro experienced users could chime in? Since the nitro mill is attached to the chassis via a metal mount (on 1/8 buggies and truggies) heat must transfer? Must help cooling, more metal for the heat to disapate thru?

I do see why you want water cooling, putting the esc in a closed box will require something as no air will be able to circulate around it.

If you decide to use a motor sleeve, offshore electrics has one for just about any motor out there. Also, Grossi makes a radiator for their liquid cooled engines, but sit down before you check the price! $150 for just the damn radiator!

http://www.grossishop.com/usa/index....&product_id=97

-JB

The only issue I can see with using the chassis as a cooling heatsink, is that the chassis flexes quite a bit during normal running on an offroad vehicle, and this would more than likely lead to problems.

BP-Revo, I figured I could use a thermal tape for connecting to the chassis as it's would give some flex :-)

I got a call this morning from TCS!!! :surprised: Talked to him a bit about the scenario and he's going to send me some info via e-mail today :party::rules::party:

This may be a very good thing :intello:

Do you want to be able to submerge your vehicles in water? What is the ultimate goal?

DrKnow will have to answer that himself, but from PMing with him a bit I think a lot of it is just about doing it because it's a cool concept. Not extremely practical or necessary, but cool nonetheless. I love stuff like this. Also, trying to completely get rid of fans.

I was looking at some of the little radiators TCS has, they are pretty neat. I could mount one of these right on the front bumper of my 1/8 GT, put a cut out in the body with a little mesh for debris protection, and it would look just like an intercooler on a real 1:1 car, and have plenty of airflow with no fan needed.

http://www.micropumps.co.uk/TCSAccessories.htm

-P- the main goal is to be able to seal off a MMP, MMM, or XL in a waterfroof case without heat issues. The purpose of the waterproof container would be so I could run my RC's in any environment without worry of water damage, which of course voids the warranty.

As an example, say I went to the track and ran just after they watered or was bashing in the mud. Then I take the car home and just take a garden hose and some compressed air to it.

Well, that would definitely be cool.

I fear Linc's suggestion of inverting it may end badly. It was designed knowing heat rises and the components are stacked in a way that the heatsink is on top to pull heat up as well as the fan pulling air up and out. You'd have to keep it awfully cool to compensate for that.

Additionally, motors and servos are pretty water proof the way they are. A little silicone around the case seam and some heavy grease around the output will waterproof a servo. A little tape and silicone will waterproof a motor.

I think in order to make this a reality you need to focus all of the cooling ability on the esc. Your cooling system will only remove so much heat, and I fear adding the servo and motor into the loop will over tax the system.

Focus on the esc, and if you end up with ice cold temps, then perhaps you can add the motor or servo.

Also since motors can run a little hotter than esc's trying to keep the motor cool enough that the esc won't overheat could really tax the system.

I'd like to see you accomplish this. Good luck.

I don't think this is much of an issue. Hot air rises (because it is less dense). That doesn't mean that heat rises in general. A heat sink physically connected to heat-generating electrical parts will absorb heat no matter which side it's connected to.

Another consideration - with DrKnow's overpowered set-ups, his trucks may spend as much time on their lids as upright - negating the effect of how the esc is oriented.

The only downside to mounting the heatsink to the chassis I can think of is stress on the components. Con't forget the heatsink is thermal-epoxied to the FET casings - and that's it - so all the stress from jumping, rough terrain, and bad landings will be transferred directly to the FETs, which are relatively fragile. Unless you plan to do something different, the FETs will also be supporting the weight of the entire ESC and all the G forces will be acting on the FETs.

This is true. The esc really needs to be mounted in some manner, then the heatsink can just be used to remove heat.

I do worry about the shock, not sure how that will affect stuff.

I do know that if you do not mount an esc sollidly you will have issues with it. That is whay I use double sided tape and zipties whenever possible.

Yeah, it would be different if the heatsink was actually mounted to the PCB (and/or case) and the sink was also touching the FETs.

I think the first thing I'll try is a liquid heat sink that attaches (in mirror) to the MMM's heat sink. Then put that whole assembly in some foam tucked into a lightweight watertight case (without the MMM's plastic case) that with any luck shouldn't be too much larger than the MMM with it's fan...

The motor cooling was more about helping HOT motors without using a fan. Like these goofy SC trucks trying to put out 1500 watts on 2S.
I think if you had the ESC as the first component in the loop it would work, assuming the chassis has enough thermal mass to dissipate the combined heat, and like Harold said the motor can be a little hotter.

I have thought about trying something like this:

http://i52.photobucket.com/albums/g4...kid/MMMWCJ.jpg

Build a water jacket around the HS and have the in/out on opposite corners on top. It doesn't require modding the MMM, except for cutting off the top of the case that holds the fan. I think the material would have to be ~1mm thick to fit between the HS and the capacitors on one side.

Nice setup jeff ! Love that little pump omg loll sooo small !!