Unique 200A ESC cooling system

[questtek]

Two of the new threads in this forum have prompted me to provide a few details on a technology I have been working on for the past several weeks. The threads involve:
1. The Inexpensive Mystery 200 Amp ESC
2. Cooling a motor via water injection

As an ocean engineer and chief scientist for the EXOMOS submersible factory in Dubai, I can tell you first hand that running both PM and brushless motors in a non-conductive oil is standard practice. All deep water submersibles use brushless motors in a non-conductive oil bath, either mineral oil or better, biodegradeable oil obtained from certain flowers.

I run inrunner and outrunner motors in non-conductive oil and even water all the time.

This thread will hopefully get some of the great minds on this forum thinking regarding my approach to making inexpensive speed controllers work: like the sub-$50 Mystery 200 Amp ESC.

I purchased 8 of these units for testing. I have 7 still fully operational and I believe the method I use to cool them has a lot to do with keeping them alive and kicking.

I have included pictures to show how this is done but quite simply here are the following steps:

1. Remove the shrink wrap from the Mystery 200 A ESC and soak in Corrosion X for several hours. Let it drip dry.
2. Build a non-conductive bottom plate and use small but long screws to bolt the esc to a fan sandwiched between the stock mystery heat sink and the lid from a small tupperware container as shown. This permits the heat sink to be firmly mated to the top of the FETS and pulls the fan directly onto the heat sink for optimum cooling
3. Punch, do not drill, holes in the tubberware container top for the motor, battery, control, and thermistor wiring as shown. (The units I use are 2 for $1 at the local 99 cent store)
4. Fill the bottom mating part of the small tupperware container with common mineral oil. It should be at the level of the bottom of the heat sink
5. Attach the top and mount into your FE. Note that I keep the gas engine mounting surfaces in the Mean Machine and design an adjustable motor mount around it. Some simple dremel tool machining provides a perfect fit for my tupperwate, oil-baithed Mystery el Cheapo.

One of the set ups I have been using is a 580 in a Mean Machine. I have made lots of runs on this thing pulling 170 plus amps with no problem.

Concept behind this Technique:
The non-conductive oil provides 100% cover for heat transfer.
The movement of the boat circulates this fluid and provides good heat transfer to the stock heat sink.
The fan pulls the heat out of the heat sink.
The fan is directly connected to the power input terminals of the ESC . ( A normal 12 vdc computer fan will work just fine on a 4-cell LiPo set up.) It continues to run and cool until the battery is disconnected.

I used this ESC cooling set up in the first test of the new Insane Hydro with Mark. Worked great, and the Eagle Tree showed cool temps. Downside was that this new boat cornered so well in the turns that centripital force caused some of the mineral oil to push out the punched holes in the lid for the wires so we had a tiny bit of mineral oil on the bottom inside of the hull. Good because this will prevent the fiberglass from rusting.

Hope you fine this useful and it stirs the imagination with guys like Jeff W. who had been playing with ESC cooling for a bit.

[Boatman]

This is just going to be hard for everyone to grasp. You say you run your motors also submerged in water? That just goes against everything we have been taught and told as to date. Brilliant if it works but electronics and water just seem like a big NO.

[Efis]

What's the temp of the esc with regular water cooling vs. the mineral oil/fan configuration?

Blake

[questtek]

What's the temp of the esc with regular water cooling vs. the mineral oil/fan configuration?

Blake

Good question and I have not done a direct comparison but that would indeed be interesting. I have a few of these water cooled from some previous experiments and postings in this forum I have done. If I have time I will compare them running the same set up in my Mean Machine with the 580. It pulls lots of amps so it would be a good test.

If you read through the forums I feel sorry for the guys who have ESC's that burn up. Major damage and repairs may be avoided with the ESC in a non-conductive mineral oil. Just a thought.

[questtek]

This is just going to be hard for everyone to grasp. You say you run your motors also submerged in water? That just goes against everything we have been taught and told as to date. Brilliant if it works but electronics and water just seem like a big NO.

You can run a motor in water ONLY IF YOU MADE ADEQUATE PREPs TO THE MOTOR BEFORE YOU ATTEMPT TO DO THIS. I would NEVER take a stock motor out of the box and run it in water.

My approach is to soak it in corrosion X. Fully sumberged for hours. I then run it in Corrosion X. Since it is pretty expensive, you can save this and then bath the motor in mineral oil.

My findings, as I mentioned to Mark F. are the following:
1. At low speed the motor appears to run more efficient due to better lubrication. This is based on Eagle Tree RPM vs amp curves.
2. At higher RPM's the motor is LESS efficient, probably due to the viscous drag. The decrease is a few percent and depends on the motor, RPM, etc.

I hope to post some pics of the motors I am using underwater. I have a dual set up on an Aeromarine Advenger Cat.

Please keep in mind many of my applications are scientific and just to make a FE run at optimum speed. I post this information primarily as "Food for Thought" and very much welcome comments and suggestions like yours. Thanks!

[Brushless55]

one reason for the high amp draw is your COG looks way forward
move your batteries back and your motor and esc will run cooler without this stuff

[befu]

One of the things that we tried in RC warship combat was filling the servo's with mineral oil. Keeps the water out and they work, but the rotated really slow. So instead, we used a 3m conformal coating on the circuit board. I forget which one, but it is a sticky brown goo. Coat the board well and it would run underwater just like a dry one. Motor and pot were the only thing left unprotected.

Nice idea on the larger esc. For me, I am just adding a cooling plate with thermal paste and then potting in epoxy. The largest one I have done is my $25 90/100 amp turnigy with my added BEC. has worked fine in my mean machine, but have only pulled up to 90 amps so far. Runs very cool.

I am into the smaller boats so all of my ESC's are single layer fets, the 90/100 might be two, do not remember.

Keep up the ideas, we like them! I remember people saying outrunners were only for planes and spinning larger diameter air props. Hah!

Brian

[questtek]

one reason for the high amp draw is your COG looks way forward
move your batteries back and your motor and esc will run cooler without this stuff

Thanks for the tip. If you notice the two red tape marks on the hull it is the balance point I use. Does this look correct to you? I just set the batteries in for the pic to show the lay out. Appreciate the helpeful suggestion.

[questtek]

One of the things that we tried in RC warship combat was filling the servo's with mineral oil. Keeps the water out and they work, but the rotated really slow. So instead, we used a 3m conformal coating on the circuit board. I forget which one, but it is a sticky brown goo. Coat the board well and it would run underwater just like a dry one. Motor and pot were the only thing left unprotected.

Nice idea on the larger esc. For me, I am just adding a cooling plate with thermal paste and then potting in epoxy. The largest one I have done is my $25 90/100 amp turnigy with my added BEC. has worked fine in my mean machine, but have only pulled up to 90 amps so far. Runs very cool.

I am into the smaller boats so all of my ESC's are single layer fets, the 90/100 might be two, do not remember.

Keep up the ideas, we like them! I remember people saying outrunners were only for planes and spinning larger diameter air props. Hah!

Brian

Great. I have tried the conformal coating Silicon spray and it works great. I have also filled the servos with Corrosion X. It has a pretty low viscosity so I really did not notice any speed or torque reduction....but I am using cheap HiTec servos.

Mark F loaned my his water cooled Little Screamer Outrunner to do some tests with. However, to get the RPM's....and thus reasonable speed.... you need lots of volts. Mark recommended 10,000 mAhr at 10 cells. That is four 5-cell, 5000 mAhr pacs at a 30 C or better rating! I am having to think twice about this type of battery investment just for some testing. Seems to turn out that the battery investment is 3 to 4 times that of the motor.

[m4a1usr]

Some great stuff going on here. I advised some of the top racers back in 2005 to build a closed/ containerized system using FC-770 3M fluorinert and they could get some serious performance increase. It fell on deaf ears.

I still have a gallon laying around and played with doing just what you are now. It could be a real winner with a small circulation system. But what I have learned over time is stick with KISS. Size my components so they run at their best eff. Then I dont need to add unnecessary items or rely on complexity to allow them to operate beyond their limits. But it sure is fun playing around with this kind of stuff no matter what!

John

[daveives]

Take a look at:

https://forums.offshoreelectrics.com/...ad.php?t=14754

which shares some ideas with this thread.

[Brushless55]

Thanks for the tip. If you notice the two red tape marks on the hull it is the balance point I use. Does this look correct to you? I just set the batteries in for the pic to show the lay out. Appreciate the helpeful suggestion.

I have a 9XL power MM on 4s and the packs are all the way back.. for me it's perfect!

[exotica]

...another great idea on aiding the heat issue with these cheaper ESC. Im going to have a play around with this, still utilizing a bigger heatsink as ive done in my recent thread.

[questtek]

...another great idea on aiding the heat issue with these cheaper ESC. Im going to have a play around with this, still utilizing a bigger heatsink as ive done in my recent thread.

While heat sinks and water cooling are all great, many of us have still had ESC's that have caught fire resulting in boat damage. I believe there is much less chance of this happening when the ESC is operating in a non-electrically conductive but good thermally conductive oil bath. Consider also that the heat sink just removes heat from the top surface of the top layer of FETS on multi-layer ESC's. The inside FETS do not have direct contact with the heat sink but when in an oil bath there is a path for thermal conduction and thus better internal cooling. Just some thoughts.

[BakedMopar]

I like this idea. It's nice that someone is thinking outside the box. Some people just don't like change!

[Jeepers]

I have a 9XL power MM on 4s and the packs are all the way back.. for me it's perfect!

Brushless,Questtek is making more power than you are, I will bet he has to run his cg that far forward just to keep it on the water. He is running 4s2p and pulling 170amps. Things change when you start putting more power in the water.

[tanasit]

Thanks for sharing. After I burnt my Etti 150A esc months ago and still have the smell left today, everytime I open the hatch I just can't stop thinking of the better way for cooling. I also have one of the cheapo 200A esc, which I added 2 water colloing plates as well as a fan with custom shroud and I even added the air scoop up front and the vent out back using the right angle 1/2" pvc plumber pipe. In winter I can plug them with the cap. So far no heat problem and the white pipe blended in for the not so ugly look.
At one point, I almost cut a square hole on the hull bottom and block it off with aluminum sheet that will attach to the cooling plate or directly to the FET board. If I don't care much about the look, I may have install the esc on the outside altogether.

[Jeff Wohlt]

Cool stuff! I am not sure when an ESC blows that it will not blow the top right off, since we know it can almost blow hatches off, that you will not end up with a big mess. Maybe no fire but you will have fluid everywhere...which is not bad since it can't really hurt anything.

Might be some neat ideas for boxes or bags for this system. Lipo bag filled and sealed. I would think some small boxes would work well and just seal them up good first with good RTV sealer then drill a small hole to fill it and then cover that hole.

Keep it going! Fun stuff. I knew questtek would hop on this challenge.

[Jeff Wohlt]

I can see a small oil cooler radiator mounted on top the hull to get a water spray and cool the oil. Questtek....did you seal around the fins on your shot up there? I guess you could and make it sealed. Nice job!

[JMSCARD]

Brushless,Questtek is making more power than you are, I will bet he has to run his cg that far forward just to keep it on the water. He is running 4s2p and pulling 170amps. Things change when you start putting more power in the water.

Yeah I have a NEU 1515 1.5y I run on 6s2p in my mean machine... I have to put the COG a little more forward to keep my boat from becoming a plane:)... NEU Torque causes serious front end lightness:)

[befu]

Some great stuff going on here. I advised some of the top racers back in 2005 to build a closed/ containerized system using FC-770 3M fluorinert and they could get some serious performance increase. It fell on deaf ears.

I still have a gallon laying around and played with doing just what you are now. It could be a real winner with a small circulation system. But what I have learned over time is stick with KISS. Size my components so they run at their best eff. Then I dont need to add unnecessary items or rely on complexity to allow them to operate beyond their limits. But it sure is fun playing around with this kind of stuff no matter what!

John

Back in the mid 90's I remember helping machine parts for a data acquisition for NASA on thier high altitude scram jet, a system that used a flurornert fluid for cooling the electronics. that was some "cool" stuff. That stuff could leak out of ANY small opening and boil off in the atmosphere. I am sure they have better stuff now, but the oil is definately easier to get and probably easier for the hobbyist.

Brian

[Steven Vaccaro]

For you electronics wizards, what causes a esc fire. I've seen full infernos and just some smoke. Is it just a chain reaction of one fet failure?

[befu]

Not sure why to even use an aluminum heat sink. Just remove the wrapping and heat sink. Seal and submerge in the oil. If the oil gets too hot, add the cooling to that. make the case aluminum and as mentioned, mount it in the bottom of the hull or sponson so it stays cool, cools the oil. Or run cooling lines through the case.

Quick, someone with a CNC mill make a case. Finned on the inside to absorb heat from oil. Bolt on cover with O-ring seal.... hmmm, wish I had gotten that cnc mill at work....

Lots of potential here

Brian

[questtek]

One of the things that we tried in RC warship combat was filling the servo's with mineral oil. Keeps the water out and they work, but the rotated really slow. So instead, we used a 3m conformal coating on the circuit board. I forget which one, but it is a sticky brown goo. Coat the board well and it would run underwater just like a dry one. Motor and pot were the only thing left unprotected.

Keep up the ideas, we like them! I remember people saying outrunners were only for planes and spinning larger diameter air props. Hah!

Brian

I thought I would do a bit of submerged component testing. The set up is shown in the attached picture. I have a small aquarium filled with tap water plus an older Futaba servo, Mystery 200A ESC and brushless motor that is ready for in water testing. All these components have been soaked in corrosion X.

The video can be seen at:

http://www.youtube.com/watch?v=Ndibf...=youtube_gdata

They say you learn from stupid mistakes and bad assumptions, thus making me the smartest dumb guy in the forum! I failed to take into account that my 40 year old home in SoCa. was built with aluminum wiring and iron pipe plumbing!

Enjoy the video as you shake your head and roll your eyes!

[questtek]

Cool stuff! I am not sure when an ESC blows that it will not blow the top right off, since we know it can almost blow hatches off, that you will not end up with a big mess. Maybe no fire but you will have fluid everywhere...which is not bad since it can't really hurt anything.

Might be some neat ideas for boxes or bags for this system. Lipo bag filled and sealed. I would think some small boxes would work well and just seal them up good first with good RTV sealer then drill a small hole to fill it and then cover that hole.

Keep it going! Fun stuff. I knew questtek would hop on this challenge.

Jeff, I know both you and I have put a lot of thought into this. Maybe we just have the record for the most burnt ESC's!

In my testing I first put the 200A Mystery into a tupperware container with heatsink and non-conductive oil. I Eagle Tree'd the running of the system and found that the oil became too hot. Even with the boat movement to circulate the oil the temp just keept going higher. This is when I realized that using a fan would be a great help and it turned out to be true.

You can try just putting the ESC in a bag, etc however I am concerned of the continued oil temp build up since there is no way to remote the heat it continues to conduct from the FET's and ESC.

You and I both have a good system using a small pump to circulate cool water over the ESC but I just wanted something a bit simplier. This is a "works in progress" and I certainly welcome your input and suggestions.

[questtek]

Not sure why to even use an aluminum heat sink. Just remove the wrapping and heat sink. Seal and submerge in the oil. If the oil gets too hot, add the cooling to that. make the case aluminum and as mentioned, mount it in the bottom of the hull or sponson so it stays cool, cools the oil. Or run cooling lines through the case.

Quick, someone with a CNC mill make a case. Finned on the inside to absorb heat from oil. Bolt on cover with O-ring seal.... hmmm, wish I had gotten that cnc mill at work....

Lots of potential here

Brian

Please see my comments to Jeff. Your idea is good with the plate on the bottom of the hull however on a cat is the flat inner bottom of the hull in contact with the water? Go get that cnc mill!

[daveives]

There is a difference between old servos and the newer servos. The feedback pot in the old servos had a much lower resistance than in the newer servos. Thus what works for an older servo will usually not work for a modern servo. The newer servos (using a lower pot current) can fail when wet by running to one end and burning out, while the older servos work fine. Also, salt water on a newer servo plug will stop operation after a minute or so as rapid battery-driven corrosion happens (much faster than normal water driven corrosion.) Of course, one logically uses an older (expendable) servo to test with. I am up to about a dozen burnt ESC's, so I like what you are trying to do.
Also, does the fan air get into the Tupperware (ie: could water get in?) or is there an undiscussed seal between the inside and outside. Thanks

[ray schrauwen]

Hammond makes some nice Aluminum cases to make a sealed unit.

Nice thread Dave!

[tanasit]

RE: Enjoy the video as you shake your head and roll your eyes!

Well, I didn't do either of those but smiled and admired your test and learn something new. Thanks...

[Doug Smock]

With all due respect to the OSE scientists, I think I'll continue to run my equipment at or just below their specifications.

Last week I had to replace a 208v 3 phase 1 hp evaporator fan motor in a -20'F warehouse freezer. The stator was cooked.
I wonder how that happened at that temperature??

Doug