For who try/have tried direct drive with outrunner


#202

i think i’ll keep pushing ahead a bit… i love the simplicity of that solution (my fist try was overly complex and ended up under powered, so now i’m going for simple and overpowered :slight_smile: ),
next step is to fill the gaps between the magnets, so the rotor give up less energy to the water, and try to kind of channel the water through: small circular gap at the front only, and channels at the back using the centrifugal force to “pull” the water through the stator .


#203

That is a good plan to make a smooth inner surface in the rotor and only some holes to let the water out. Also the spokes should be covered somehow. You should not cover the stator with too much material, just a thin coating, otherwise it will prevent the heat transfer to the water. Do not fill the space between the slots in the stator, its the best cooling surface you have and it leaves sone space for the water to move through.
I measured the no load current under water of my very first motor with a cap which prevented any pumping and had an additonal power loss of just 90W. My actual motors have the spokes covered from the outside, leaving holes between the spokes at the circumference. Another good idea is to use ceramic bearings and to cover the single bearing at the motor mount with a cup and an o-ring. Maybe its also possible to make a simple sealing at the other end, i use 2RS bearings which provide some sealing. You can keep the spokes of the motor mount open to let fresh water in, it will help to cool the leads and winding head.
I just saw you left a message in Direct drive outrunner with direct water cooling


#204

I have lost track of who has successfully built what. Who has successfully build a E-foil using a out-runner that does not use direct water cooling?


#205

Maybe you? Tell us more.


#206

No, I wish. Its been about 8 months of learning, reading, waiting for parts, and occasionally building. I think I am on the right path, but I was hoping to hear more dry(possibly oil filled) out-runner success story’s. But I guess it would be even better to be the first success story.


#207

Currently I am procrastinating drilling some holes in my new mast to run the wires.


#208

Haha just do it. I hacked mine up yesterday with some ugly holes. Need to tidy up with the dremel.


#209

Anyone tried anything like this? With a flex drill bit that would sit in the board and connect to the propeller. No need to have electronics in the water then.

image


#211

So I started to test my Alien powerglider outrunner motor inside a soon to be oil flooded aluminum housing. I discovered something interesting and really annoying. The spinning magnets used in the outrunner design are causing induction heating of the aluminum housing. With no oil inside the housing and only the housing (aluminum tube) over the motor, the outside will quickly get too hot to touch. I have verified that the outer bell of the motor is not rubbing on the inside of the housing tube so it must be induction heating. The aluminum housing can be super hot and then if you slide it off quickly and feel the motor housing it is cool to the touch. At wide open throttle and no housing over the motor, the motor pulls about 70W no load. Doing this same test again but with only sliding the housing over the motor it draws about 180W no load. So I have a 110W heater!

Has anyone else run into this issue before? Any ideas how to fix it? I borrowed an induction heating book from another engineer at work and am trying to figure it out. So far all I have figured out is that It appears to be due to “joule heating” where the susceptor does not need to be ferrous.


#212

You could use a plastic tube for the motor section, something with a bit of temperature stability. I use my outrunner in a PLA shell, but it‘s only a protective shell as mine is watercooled. If the steel part around your magnets (the rotor bell) would be thicker, there would be less field escaping. But I can not estimate that right now. Would be hard to thicken that part, too. I mean I can only imagine turning a steel tube and pressing it on the rotor.


#213

I have a 80 mm Alien out runner in a 3.5" OD aluminum tube. Since I put it in the tube I have not run it much. You got me wondering if the same thing will happen to me. How long do you run it. Under load, or just free wheel? RPM’s? Voltage? I will try to duplicate.


#214

I just tried to duplicate your problem but I ran into a snag. I had a wire burn up. I don’t have time to fix it now. In a week I will repair it. I am interested in what you find out is the problem.


#215

The field of the rotor magnets reaches the stillstanding aluminium pipe, as they are turning, the change of field in an conductor creates a current, leading to a field which pushes against the root cause field, braking the rotors rotation. You can build nice eddy current brakes like this.
There is not a lot you can do, here are some options:
Larger diameter will weaken the brake force but leads to higher drag.
Interrupt the current in the aluminium pipe by making thin slots in the pipe and filling them up with epoxy.

On the other hand: 180W at full speed is not a nightmare. The power is generated inside the aluminium, not inside the motor. How much additional powerloss do you expect after filling the housing with oil? How thin is that oil, how small viscosity can you get? Maybe destilled water could be an option, or a cooling lubricant?


#216

Thanks for the reply. I have thought about using a piece of clear polycarbonate or acrylic fir the time being but worried that it might not be a good enough thermal conductor to conduct heat to the water when the motor is running at full load


#217

You know what you are talking about @PowerGlider! I have run into the same solution ideas reading through the book I mentioned on induction heating. I agree that the power loss is not a total killer but far from ideal. I am somewhat concerned with how hot that housing might actually get even in the water. If the temperature differential between the inside of the aluminum tube and the hot oil is not big enough it might not do a very good job of keeping things cool. I will let you know what the losses due to the viscosity of the oil are. Its fairly thin at room temp and I imagine will get much thinner at higher temps.


#218

Anyone tried one of these to batter cooling issues?


http://www.weiku.com/products/17156974/BL3650_2300KV_Brushless_Inrunner_Motor_for_RC_Boat_WaterCooled_.html
https://pt.aliexpress.com/item/BL3656-1800KV-RC-Brushless-Inrunner-Motor-Water-Cooled-1100059/490768706.html


#220

An outrunner in oil at 7Nm 4000RPM will produce a lot of heat, 10-20% of the input power. The addition of another cylindric wall around the turning bell is somehow overkill, it does not reduce the hydro friction in the complete system by any means. On the other hand it separates the inside from outside with some benefits regarding the microbes…
I underestimated this eddy current, former motors i designed and use have thicker flux ring. So the stray field is rather wide with these light flux rings. 10-20% of these 180W losses happen in the motor windings. In Germany you can by iron wire in different qualities and makings. You could wind one layer around the motor bell with some epoxy e.g. and e.g. kevlar strings. Such a wound wire layer has good effects on the stray field. Use 0.4mm wire to get 1mm more diameter.


#221

It sounds like PowerGlider and you understand what is occurring. I would like to duplicate your test and see how my set up reacts.
When I am running at full 6S, no load, I don’t see a obvious heating issue. What is your full throttle voltage? How long do you have to run it before it is hot to the touch?


#222

We need 12s for the 100Kv motor to get enough RPM for any prop to drive. With 6s you will not get enough power and the prop would have to have 10-20" pitch which is unrealistic high for the diameters we want to use. The powerloss is dependent on the air gap and velocity and the thickness of the aluminium pipe.


#223

Working on mounting the motor to the mast now. Then I will try at 12S.