Definitive Tests on the Heating and Cooling of Hub Motors

[John in CR]

ok, John, I fully understand your exhaust side with the blades creating depressure, I just wanted to see your intake approach/positionning holes near the stator axel or near windings, I don't think I saw it on your pics...

This motor is smallish diameter, so I put them everywhere. Zappy has had really good success swiss cheezing most of the intake side. One thing I would do differently if I was venting a common hubbie is that I would cut 6 narrow slots angled slightly rearward when viewed at the top of the rotation and tap 6 bolt holes next to them. The idea is add blades from the outside on the intake side too. Those blades would only protrude on the inside and I would bend them rearward. They will accentuate the spinning flow inside the motor on that side and deflect it toward the stator for a higher coefficient of convective heat transfer. I used dual sided interior blades on all my previous ventilated motors on both sides. Increasing turbulence and velocity right at the stator, especially the end windings makes success simple. Plus it may even create a low pressure region at any intake holes near the blades further aiding fresh air flow.

FWI, I don't normally even turn my temp sensor on anymore, but I just did an errand run 8 miles round trip with only a 5 minute stop. I did lay on the throttle more than usual, mostly teaching moto's that no, they can't pass the fat gringo on the silent little bike looking thing. No hills but it is a pretty consistent downhill grade of about 2% on the way and uphill on the return. I was used an average of 61wh/km (98wh/mi), so an abundance of fun riding with my high efficiency rig, since long-term average is 35-40wh/km. After the uphill ride home I turned on the cooking thermometer, which I believe reads a few degrees high after I butchered the sensor tip installing it, the motor temp was 73°C.

Regarding intake holes/slots, the only thing I would shy away from is putting them from the windings to the perimeter, because I don't want any exhaust venting on the intake side, because I believe that will greatly diminish the flow through the magnetic gap. Our magnets are the most temperature sensitive material in our motors, so since I've gone to one sided intake and other side exhaust, I will never go back. I wouldn't even care if someone had temp sensors on the stator saying it was slightly cooler with 2 sided exhaust, I want flow thru the gap if I can create it.

 

[migueralliart]

I just wanna say the effects of just going with holes in the covers are not negligible.

There is a benefit just not as extreme as other approaches.

For people not running constantly at those power levels drilling holes might not be a bad idea.

 

[John in CR]

I just wanna say the effects of just going with holes in the covers are not negligible.

There is a benefit just not as extreme as other approaches.

For people not running constantly at those power levels drilling holes might not be a bad idea.

Look earlier in the thread. Justin ran quite low power levels and overkill with the holes didn't net much cooling. His thermal limiter didn't stop kicking in until he added interior vanes/blades. Of course any ventilation will help, but a moderate hill on a hot day at moderate power is likely to kill any hubbie with only holes if the hill is long enough. How many fans have you ever seen that are just round holes in a plate?...None, because it won't move much air.

If you're going to open your motor to the elements, for a small amount of additional effort you can turn your motor into an effective centrifugal fan. Note in my first post on the topic, the slotted cover isn't the way to go. I believe just 6 exhaust holes or larger slots plus blades will be more effective.

 

[migueralliart]

You can see that the addition of the milled perimeter slots with no "intake" holes had a measurable effect on how long it took the motor to hit thermal rollback, increasing from 19 minutes to 26 minutes. Opening up the intake holes on the same side as the machined slots had only a very marginal improvement over no intake holes. However, opening up the holes on the opposite side cover instead had a huge effect. You can see that with the opposite side cover holes open, the motor never got hot enough to hit thermal rollback. This is a similar observation to the first set of tests with simple drilled holes in each side plate, where there effect of holes on both side covers was much more than twice the improvement you get from having them one just side plate.

The same setup using the 30 smaller holes rather than the 10 elongated slots performed good but not quite as well, still hitting the thermal rollback at the 40 minute mark.

Here is some of the early results posted by Justin.

Not to start an argument but not everyone here is looking to push their motors into thermal rollback past the 40 minute mark. Plus you will need a really BIG battery to do so.

Like I said there is clearly a benefit. Just not good for all builds/people.

 

[John in CR]

I stand corrected. I missed that Justin finally opened that motor up enough to avoid thermal rollback without blades, but I do understand those running modest power don't need the same flow my motors need. There are lots of ways to skin a cat, and putting giant holes in a hubmotor would be my last choice to pick up a few hundred watts additional cooling. I'd rather keep the motor sealed an double the cooling by doubling the surface area of both covers both inside and out, something no one has even tried yet.

John

 

[pendragon8000]

I'd rather keep the motor sealed an double the cooling by doubling the surface area of both covers both inside and out, something no one has even tried yet.

John

Are you talking about heat sink? With vanes?

Btw
I moded my hub as previously mentioned with large 16 an 19mm holes near the disc mount and 12mm holes around the perimiter of the gear side as was shown to be quite effictive in the definitive results , however after doing jumps in the park then a quick 5 min trip on a path I found myself in a fun run with 100s pedestrians my speed dropped and the motors thermostat clicked at 110°C . Unexpected... I guess it was due to low speed high power repeated acceleration over a few 100 meters after a moderate flogging.
(Bloody fun run!)

 

[Merlin]

...just my 2 cents,
but AIR is the bad'est way for a heat getaway...
So a closed hub with "Fins" to the Rim maybe cool down the Case...but as we allready know a 60°c Hub Case has on Cruising a near double Temp inside.
Take that uphill and you have tripple Temps inside.

Heat"Pipes" work only in direct contact with the element to cool.

On Justins tests i see that 1" Holes are alot of difference. (Both sides)

my own experience on a 3540/4080 is that i can cruise (a really long time) with just 1KW .... with more the temp is rising.(over 120°c)
My Tempprobe is connected at the end of Phase wire -> starting Windings. So i think it is the most Hot Point in my hub.
(theres no Problem to drive with peaks 6-7KW...but cruising 10-12km (1KW) will heat up the Motor over 120°c)

 

[itchynackers]

I can't wait until Justin gets back to some "definitive tests" here. Tired of logging in to read frivolous junk. Where are you Justin?

 

[Merlin]

This "frivolous junk" is SOMETIMES the difference Between Theory and Reallity.........

 

[pendragon8000]

I can't wait until Justin gets back to some "definitive tests" here. Tired of logging in to read frivolous junk. Where are you Justin?

Lol. Yeah I apologize to anyone that feels the same as I've put up a lot of comments along the way, some of my own reports and some questions about results.
As Merlin says, I hope I'm contributing to this thread by giving real life experience in regard to different methods that have been tested in the lab. So in future ppl may benefit from reading nd decide what's best for them.

However itchynackers I to am very keenly awaiting more "definitive tests".

 

[Chalz]

I have been lightly following this thread, so forgive me if I missed some of it. Has anyone tried to cool the motor like this build?

http://endless-sphere.com/forums/viewtopic.php?f=6&t=49545&p=780575#p780575

It's not the best, but it's simple, cheap and fairly effective. It would be nice to see some test data on something like that.

 

[amberwolf]

To find out if that method has been tried in this thread, you'll have to look thru this thread's posts.

I know it has been discussed, if not in this thread then in others, but I don't know if any methodical / comparative testing was done on it.

But it will still be good to have your test data on your version of it, once that is available, posted here in this thread.

 

[Obiwan007]

I like the idea of the evap cooling method since it can easily be combined with virtually any of the other methods to increase the overall cooling effect. I think it may be able to be refined. For instance if the strap holding the felt were perforated, maybe a lightweight metal mesh. Also the conductivity of the saturation material might be an area to experiment with. I have recently finished venting my side covers on the "Electric Jeepster".

It turned into a marathon project because I first pinched and grounded my hall wires when reassembling the motor and then got the hall wires mixed up after that repair. All is good now however. It had been my intention to do some testing before the upgrade so i could compare the result afterward. but i found a short window in my schedule before i could do some testing and decided to just "get er done". I have only had the the opportunity to take it on one ride since the upgrade but the cooling is significantly better. If I had to guess, (I know how you all hat that) I would say 20-30% better.

 

[John in CR]

Better hope that felt never runs dry, because then you've got a nice thick insulator helping roast your magnets.

 

[cal3thousand]

Better hope that felt never runs dry, because then you've got a nice thick insulator helping roast your magnets.

Is there another material that could deliver the water and still allow some heat transfer if dry?

 

[dan974]

looks like a scooter cooling solution

or side plate as a blower...

 

[speedmd]

Is there another material that could deliver the water and still allow some heat transfer if dry?

Porous metal possibly. http://www.eetimes.com/document.asp?doc_id=1280108

 

[Chalz]

Yes- There are lots of other materials to try. I am using a polymer crystal found in my babies diaper. Pics on my build page.

http://endless-sphere.com/forums/viewtopic.php?f=6&t=49545&p=780575#p780575

This diaper material will not dry out over an entire day. Not as if the felt ever did. These are pretty low risk experiments- nothing to be scared of. If you survived making a wiring harness without shorting a battery, the you should be able to handle monitoring the moisture of felt and temperature of your motor with your finger. I'm looking forward future experiments from someone with better test equipment on this thread. Please let me know if you dare to try....

 

[nechaus]

Just got onto this thread, Wooo, great information, i am learning alot,
This data will help me on my next build, fantastic that its being quantified


Anyone thinking of putting 3d printed ABS or PLA inside your hub motor, I would not risk the plastic melting over the windings and magnets causing a certain mess inside, unless you are sure you will not reach hi temps, Nylon or Polycarbonate might do it tho...
id prefer aluminium as it will absorb heat better


RE. Drilling the hub covers with the case on is not a good idea hahah, its not that you might hit some internal wiring, but first thing i think of is scrap metal going inside the hub motor...


I cant wait to see the results of liquid cooling.
I have a very small radiator in my shed which i have been keep in effort to try and liquid cool one of my magic pies...

 

[Cowardlyduck]

Thought I would just through it out there, as I'm unlikely to try it myself.
I had the idea to put EDF type inn-runner fans into the ~25mm holes of the stator in my HS4065. It would also work for any HS35xx motor, or any other motor with 25mm holes through the stator support.

I managed to find these fans if anyone wants to give it a try.
http://australia.rs-online.com/web/p/axial-fans/3811392
http://www.newark.com/micronel/d241l-012gk-1/fan-24mm-12vdc/dp/95K6059

I would try it myself, were they not so expensive just for one fan!

Cheers

 

[itchynackers]

Looks like lots of folks are chomping at the bit for the next posts on the testing to see which ways may be best to explore further.
Open/sealed, wet/dry Me too.

Hey guys, unfortunately I had to put a pause on this testing process just as it was getting interesting due to some prior project commitments with more imminent deadlines. There is a small chance we'll be able to run one of the oil cooled tests this weekend, but otherwise we'll need to put things on hold for a good 5-6 weeks before things can resume.

-Justin

It's been like 8 months since your last post Justin. Will there be any other test data released (any liquid results)? Can you give us a hint?

 

[parabellum]

Desperately waiting for oil cooling test results.

 

[Arlo1]

Desperately waiting for oil cooling test results.

Just added ATF to a hub motor about 1/4 full no load amps were 6.5 before and 8.3 after. Its high but a cheep motor and cheep controller for a drift trike. I will try another one with dyno tests in the future and we will put a temp sensor in the windings as well.

 

[h0tr0d]

Parabellum, is this good enough?

 

[parabellum]

Parabellum, is this good enough?

Thanks for the graph. I have 2 oil cooled hubs and know there is huge improvement in cooling the stator, I just like those w/degree numbers so much.