Make a watercooled high kw motor and I'll buy it !

Get all your technical information about electric bikes here.
Ohbse   10 kW

10 kW
Posts: 815
Joined: Dec 26 2013 5:15pm
Location: Auckland, New Zealand

Re: Make a watercooled high kw motor and I'll buy it !

Post by Ohbse » Jan 22 2017 3:17am

Totally agree with John. If you're on the verge of constantly smoking a MXUS, even with the very considerable advantage of FF and Hubsinks, then you're using the wrong motor for your application.

You're saying you'd pay double for a motor with water cooling, why not pay only 20-30% more and get a substantially more efficient motor like a 50mm QS v3? Utilise your existing hubsinks, add some more FF and you'll have increased thermal mass, increased surface area to dissipate the heat, but critically you'll be generating substantially less heat in the first place. You'll go faster or use less energy for the same activity, all without the added complications of water, hoses, radiators, pumps and control headaches.

The answer is never to add inactive mass to a system to try to overcome some shortcoming. Taking that same mass/complexity and adding copper/magnet area will always deliver a superior solution. More active mass under less stress produces less heat as well as being able to dissipate it quicker.

Jestronix   10 kW

10 kW
Posts: 508
Joined: Jul 18 2015 4:59am

Re: Make a watercooled high kw motor and I'll buy it !

Post by Jestronix » Jan 22 2017 6:03am

I would have thought a bigger motor would just delay the inevitable, it will overheat a few mins later than the mxus. Loading up unsprung weight to the level where I would never overheat would require an one hell of a giant motor.

Water cooling pump , radiator and hose weighs nothing in comparison to Xtra KGs needed. But your right adding Complexity has it s down sides.

Really a mid drive is the ansewer, DD hubs are just a poor design as far as heat goes. water cooling can mitigate that and keep unsprung weight low.

Maybe I'm silly but I like the idea of a watercooled DD hub :)

User avatar
Offroader   10 MW

10 MW
Posts: 2466
Joined: Sep 08 2013 9:03pm
Location: USA

Re: Make a watercooled high kw motor and I'll buy it !

Post by Offroader » Jan 22 2017 6:13am

Jestronix: You are using an oil cooled motor with those attached heat sinks, and heat is still an issue for you?

It seems most people who have an FF + hubsinked motor seem to claim they have no overheating issues anymore.

Jestronix   10 kW

10 kW
Posts: 508
Joined: Jul 18 2015 4:59am

Re: Make a watercooled high kw motor and I'll buy it !

Post by Jestronix » Jan 22 2017 6:50am

Offroader wrote:Jestronix: You are using an oil cooled motor with those attached heat sinks, and heat is still an issue for you?

It seems most people who have an FF + hubsinked motor seem to claim they have no overheating issues anymore.
Yeh I have a few 10min steep climbes that put my motor in the 100c range , I have to take a break and let it cool off, oil made a big difference, the hub sinks added a bit more. But I still have to pull over and let it cool. Pushing a constant 3kw average on the climb soon pushes the 100c mark and quicker on those 35c days :)

FF I thought was not quiet as good as trans oil ? Far more contact area with oil

John in CR   100 GW

100 GW
Posts: 13823
Joined: May 20 2008 12:58am
Location: Paradise

Re: Make a watercooled high kw motor and I'll buy it !

Post by John in CR » Jan 22 2017 7:38am

Jestronix wrote:I would have thought a bigger motor would just delay the inevitable, it will overheat a few mins later than the mxus. Loading up unsprung weight to the level where I would never overheat would require an one hell of a giant motor.

Water cooling pump , radiator and hose weighs nothing in comparison to Xtra KGs needed. But your right adding Complexity has it s down sides.

Really a mid drive is the ansewer, DD hubs are just a poor design as far as heat goes. water cooling can mitigate that and keep unsprung weight low.

Maybe I'm silly but I like the idea of a watercooled DD hub :)
Not bigger, just better as in more efficient, so you make less heat. Size comes into play based on how much power you want.

Heat in our hubbies comes from 2 main sources, iron losses and copper losses. Iron losses result from the alternating magnetic field, so they're rpm based, and better thinner stator lams combined with a lower slot/pole counts can make them low enough to be of little concern except maybe for high speed cruising. This makes copper losses the main source of heat, and is current squared times resistance. The route to greater efficiency (ie less heat) is simple... use more copper. Unfortunately copper is expensive and hubmotors already do fine in their designed use on low power ebikes, so advances in hubmotors has been agonizingly slow.

Ohbse   10 kW

10 kW
Posts: 815
Joined: Dec 26 2013 5:15pm
Location: Auckland, New Zealand

Re: Make a watercooled high kw motor and I'll buy it !

Post by Ohbse » Jan 22 2017 4:21pm

Jestronix wrote:I would have thought a bigger motor would just delay the inevitable, it will overheat a few mins later than the mxus. Loading up unsprung weight to the level where I would never overheat would require an one hell of a giant motor.

Water cooling pump , radiator and hose weighs nothing in comparison to Xtra KGs needed. But your right adding Complexity has it s down sides.
Bigger/better motor will convert more of your input energy to torque, thus creating less heat in the first place. Because it's physically larger it has both more thermal mass to absorb that heat and more surface area to remove it.

Picture a bucket (Hub motor) dangling from a hook. A hose fills the bucket with water (Heat) at a constant rate. In the bottom of the bucket is a hole. If the flow in is larger than the flow out of the hole, the bucket will overflow (overheat). You're having to wait while your bucket empties. There's a few ways to solve this problem, you can either buy a bigger bucket (more mass to absorb heat, takes longer to overheat), you can reduce the flow of water into the bucket (turn down the power OR improve efficiency) or you can make the hole in the bottom larger (improving heat dissipation with FF/Hubsinks for instance). Improving all three of these factors by even an incremental amount can potentially completely eliminate your overheating. That might only require a couple of KG, which will be directly comparable to the total weight of all the water cooling bits and pieces plus coolant.

Buy a QS 205 50 v3, add hubsinks and FF and I would be surprised if you have to stop any more. These things are significantly better than the Cromotor V2 which in turn are better than the MXUS. I own all of the above, I can't overheat the QS at all, even riding at 50wh/km into huge head winds at 35% over the speed limit. I would have smoked the MXUS in the same circumstances before I got half way home and the Cromotor would have been 100c+. QS was ~40c.

If you're concerned over the increase in weight, take the new hub to a machinist and mill away some material. Remove the inactive portions of the spoke flanges, thin the steel by 20%. Pocket out (but not through hole) the side covers in a spoke shape. You can probably get damn close to the weight of a MXUS, only a much higher percentage of that mass is actually contributing towards being a motor (e.g Copper)

Jestronix   10 kW

10 kW
Posts: 508
Joined: Jul 18 2015 4:59am

Re: Make a watercooled high kw motor and I'll buy it !

Post by Jestronix » Jan 22 2017 4:58pm

It does make sense , bit from a 40 to 50 wide DD making my over heat problems go away ? I think I may get 30% further up my hill ? Efficiency increase based on more copper means only 25% less heat gen ? And a longer over heat process.

It makes total sense why in ur application it made all the difference. But for me ? Up hill constant load 1000m climb

Ohbse   10 kW

10 kW
Posts: 815
Joined: Dec 26 2013 5:15pm
Location: Auckland, New Zealand

Re: Make a watercooled high kw motor and I'll buy it !

Post by Ohbse » Jan 22 2017 6:21pm

Jestronix wrote:It does make sense , bit from a 40 to 50 wide DD making my over heat problems go away ? I think I may get 30% further up my hill ? Efficiency increase based on more copper means only 25% less heat gen ? And a longer over heat process.

It makes total sense why in ur application it made all the difference. But for me ? Up hill constant load 1000m climb
I was pushing a constant 3-4kw for probably 20 minutes of riding at high speed, probably at higher speeds than your hill climb, but still...

It's not just the 25% increase in stator width (though that alone may be enough to let you reach stable temperatures), the QS v3 has more copper per mm of width. The difference in active mass is probably more like 35-40%. If you consider that it can then remove that much lower amount of heat more quickly you can see how it adds up to a big difference. Note I'm not running FF or Hubsinks on this motor either! I simply haven't been able to get it hot enough to justify adding them.

sketchism   100 W

100 W
Posts: 168
Joined: Feb 17 2016 8:35am
Location: GC, Australia

Re: Make a watercooled high kw motor and I'll buy it !

Post by sketchism » Jan 22 2017 7:30pm

Offroader wrote: I would love to try FF + hubsinks but I worry I won't be happy with it and then I have a motor full of FF fluid and now I can't convert over to air. I am not sure if you can get all of that out of the motor and it will make dirt stick inside.


i have to be very careful throwing guarantees around but i guarantee there is no problem removing FF later on if you decide not to use it, you can just wipe it off with a rag and mild solvent

and although this isnt a 'guarantee' if anyone bought a HubSink, installed it correctly with thermal paste and FF, and it still didn't do what it was meant to do (which hasn't happened yet but who knows) i'd be happy to refund them.


the 2 vids here are essentially non stop wheelies and hill climbs at 18kw https://www.facebook.com/groups/endless ... 704185062/
Last edited by sketchism on Jan 24 2017 5:34am, edited 1 time in total.
Don't overheat your ebike hub motor - www.hubsink.com (beta testing still open)
https://hubsink.com HubSink - Hub Motor Heatsinks

https://australiandroneco.com Australian Drone Co - Aerial Videography

Nothing is fool proof to a sufficiently talented fool, trust me we'll find a way :idea:

Jestronix   10 kW

10 kW
Posts: 508
Joined: Jul 18 2015 4:59am

Re: Make a watercooled high kw motor and I'll buy it !

Post by Jestronix » Jan 23 2017 4:29am

Makes sense can't wait for ebikes.ca to test.


Mmmm wheelies and speeding down public roads :( thinks long term dis will bring crack down on ebikes prefer to see off-road actions .

I'll prob go QS for next build. Qbix or futr build.

User avatar
ferret   1 kW

1 kW
Posts: 430
Joined: Aug 04 2013 1:58pm
Location: Israel

Re: Make a watercooled high kw motor and I'll buy it !

Post by ferret » Jan 23 2017 6:24am

A bike with a smaller wheel will require less torque from the motor to achieve the same speed on the same grade, the motor will draw less amps and heat less.
It will reduce your top speed, so you will need a higher voltage or a lower motor turn count to compensate.

Avner.
Don't believe everything you think.

Cove GNG-Spot build thread

Astral Rune build thread

Jestronix   10 kW

10 kW
Posts: 508
Joined: Jul 18 2015 4:59am

Re: Make a watercooled high kw motor and I'll buy it !

Post by Jestronix » Jan 23 2017 7:26am

ferret wrote:A bike with a smaller wheel will require less torque from the motor to achieve the same speed on the same grade, the motor will draw less amps and heat less.
It will reduce your top speed, so you will need a higher voltage or a lower motor turn count to compensate.

Avner.
Yep probably the best option, this also = out though? A motor has its perfect sweet spot where it's efficiency is perfect and thus heat generation is less. So say a 2T in a small wheel will give u what a larger wheel with a 6T will give u yeh ? So really a higher voltage setup with a med to hight turn motor in a small wheel is the best option yeh ?

A 20" wheel with a 4T and maybe 130% overspeed should give a billygoat and a cheetah yeh ? I run a 5T in a 26" with a 2.8 tire, so it's not the best climber heat wise.

I'm edging on ah 16" wheel and a moto rim :) doubt I could fit spokes in ?

User avatar
Offroader   10 MW

10 MW
Posts: 2466
Joined: Sep 08 2013 9:03pm
Location: USA

Re: Make a watercooled high kw motor and I'll buy it !

Post by Offroader » Jan 23 2017 4:13pm

Jestronix: Can you do me a favor. I'm trying to determine how well oil cooled motors transfer the heat to the exterior of the motor.

What I have found when using internal fans to circulate the air inside my MXUS is that once the hub gets hot when stopped, if I drive for just 30 seconds the entire hub feels lukewarm when I touch it. It makes no difference how fast my fans are.

This made me believe that heat sinks don't have that much of an impact when you are riding. The hub easily cools down on the exterior.

Could you get your motor burning hot to the touch, maybe stop for a few minutes after the motor is fully heat soaked. Then ride for 15-60 seconds, and then quickly stop the bike and feel the hub motor and see if it feels burning hot or just lukewarm, meaning it cooled down really fast. I believe you will notice that the exterior of your hub will cool down very fast when you are moving.

Then maybe see if you can ever get the hub motor to feel burning hot to the touch while keeping speeds above 15MPH.

If you can then see how fast it takes the motor case to get burning hot again after your bike is stopped, assuming it is not burning hot when you stop and check it.

What I am trying to find out here is if there is a bottleneck on the heat transfer out of the motor. Meaning, the external motor shell will quickly reduce temps far faster than it can get heated internally, even with OIL. Because if the motor feels just lukewarm after you touch it immediately after stopping your bike, you are not heating your hub motor case fast enough.

This is what I have found, but I did not use oil cooling.

This would mean that the bottleneck in your setup is not external convection (The transfer of heat to the air outside the motor), but the convection or conduction inside the motor.

This would mean it would benefit you to somehow add heatsinks inside the motor.

I hope you see what I am getting at here and can check this for me, Thanks.

sketchism   100 W

100 W
Posts: 168
Joined: Feb 17 2016 8:35am
Location: GC, Australia

Re: Make a watercooled high kw motor and I'll buy it !

Post by sketchism » Jan 23 2017 10:50pm

Jestronix: Can you do me a favor. I'm trying to determine how well oil cooled motors transfer the heat to the exterior of the motor.

What I have found when using internal fans to circulate the air inside my MXUS is that once the hub gets hot when stopped, if I drive for just 30 seconds the entire hub feels lukewarm when I touch it. It makes no difference how fast my fans are.

This made me believe that heat sinks don't have that much of an impact when you are riding. The hub easily cools down on the exterior.

i know you wanted to check a real world example and if you can use a flir cam or a cheap laser temp guage thats a good idea, just for discussion's sake here is the answer to that question directly as shown during thermodynamic modeling , 98C vs ~165c in the same operating conditions (i.e oil filled hub producing X amount of excess heat)

The hub itself is intrinsically bad at shedding the heat by itself as shown in this test with the same amount of heat generation with and without the sinks, making some even from LED heatsinks on ebay is vastly more effective than hoping the shell will shed that heat effectively.

Image
Don't overheat your ebike hub motor - www.hubsink.com (beta testing still open)
https://hubsink.com HubSink - Hub Motor Heatsinks

https://australiandroneco.com Australian Drone Co - Aerial Videography

Nothing is fool proof to a sufficiently talented fool, trust me we'll find a way :idea:

User avatar
Offroader   10 MW

10 MW
Posts: 2466
Joined: Sep 08 2013 9:03pm
Location: USA

Re: Make a watercooled high kw motor and I'll buy it !

Post by Offroader » Jan 24 2017 3:18am

skethism, I agree with that if the motor is not spinning and the bike is stopped the hub sinks will work very well. The motor will also get very hot like that when not riding.


What I have found is that if I am riding at say 20-25MPH, the hub shell without any heatsinks will be lukewarm / cool to the touch. A spinning hub motor will shed the heat so fast while riding the bike will cool down super fast even without hubsinks.

Even if I let my bike sit for a few minutes so that the exterior gets burning hot to the touch, it will cool the exterior in 10-15 seconds of riding.

I want to use another high speed fan approach on my MXUS, which worked very well in my Cromotor.

I have to decide for my new motor if I want to give up this setup for FF + hubsinks.

Here is a basic diagram of my cooling setup, it blows the hot air out continuously at high enough temps which will start to burn your hand after a few seconds. This is some serious heat shedding and I can't believe that FF + Hubsinks will work as well. But it might just work as well.

What I need to know is how hot the exterior of the motor is getting at cruising or even slow speed riding. If the motor shell is still burning hot when riding then the hubsinks and FF are working. If the motor feels cold to the touch on the outside then the hubsinks won't help unless you are basically stopped.

Image

User avatar
Samd   1.21 GW

1.21 GW
Posts: 3667
Joined: Jun 28 2011 11:08pm
Location: Ballarat, AU
Contact:

Make a watercooled high kw motor and I'll buy it !

Post by Samd » Jan 24 2017 5:55am

Offroader you're not factoring the lack of heat path into your assessment of hubs of speed. Before ferrofluid heat had to travel via the bearings.

Especially relevant to your last sentence.

Using conduction to the case then maximising convection at the periphery is preferable to trying to perform convection inside.


Sent from my iPhone using Tapatalk
http://ballaratebikes.com/

ImageImageImage
"Knowledge is knowing that a tomato is a fruit. Wisdom is knowing that a 165mm rear shock can't deliver 220mm of rear swingarm travel." - Oprah Winfrey.

sketchism   100 W

100 W
Posts: 168
Joined: Feb 17 2016 8:35am
Location: GC, Australia

Re: Make a watercooled high kw motor and I'll buy it !

Post by sketchism » Jan 24 2017 6:01am

Offroader wrote:skethism, I agree with that if the motor is not spinning and the bike is stopped the hub sinks will work very well. The motor will also get very hot like that when not riding.


What I have found is that if I am riding at say 20-25MPH, the hub shell without any heatsinks will be lukewarm / cool to the touch. A spinning hub motor will shed the heat so fast while riding the bike will cool down super fast even without hubsinks.

The defining factor of the difference though is the FF here creating a thermal path from the stator to the magnet ring/sink and out to the turbulent air

EDIT: sam beat me
Don't overheat your ebike hub motor - www.hubsink.com (beta testing still open)
https://hubsink.com HubSink - Hub Motor Heatsinks

https://australiandroneco.com Australian Drone Co - Aerial Videography

Nothing is fool proof to a sufficiently talented fool, trust me we'll find a way :idea:

User avatar
Offroader   10 MW

10 MW
Posts: 2466
Joined: Sep 08 2013 9:03pm
Location: USA

Re: Make a watercooled high kw motor and I'll buy it !

Post by Offroader » Jan 24 2017 1:12pm

I'm just saying I would like a test of the exterior motor case temperature when riding at 15 MPH+ speeds with FF or oil.

I agree when stopped the exterior of the motor gets hot.

I have found if you ride at 15-20MPH+ speeds the exterior of the motor cools down very quickly without hubsinks, so quickly I question the usefulness of hubsinks on a bike at speed.

From my own tests using forced air convection inside the motor, I could not get the exterior of the motor hot when the bike is in motion. I even tried to keep the winding temp at 115C and had my fan speed on maximum, still could not get the exterior hot.

If I stopped the bike, then yes the exterior motor temp got very hot, but still took some time to do so. When I used NO forced air convection on my MXUS, I could not get the exterior of the motor hot even when left sitting for a long time.

There is a bottleneck in the FF or magnet gap.

John in CR   100 GW

100 GW
Posts: 13823
Joined: May 20 2008 12:58am
Location: Paradise

Re: Make a watercooled high kw motor and I'll buy it !

Post by John in CR » Jan 24 2017 9:17pm

Jestronix wrote:So say a 2T in a small wheel will give u what a larger wheel with a 6T will give u yeh ?
Nope. They're exactly the same motor and just require a different voltage and motor current for identical results (power, rpm, efficiency, torque, etc) at the same gearing (ie wheel size). As a practical matter though we bump into the available controller voltage limit, so slow wind motors can never keep up with fast winds. If you can live with the smaller wheel in terms of looks and ride, a smaller wheel will win every time with greater acceleration and efficiency for a given power level. That's why escooters are made by the millions, but no emotos with hubbies are manufactured, since users will burn them up from being too steeply geared.

That's not to say we can't get away with it on our much lighter ebikes than scoots and motor, especially lightweight riders, but look at the heat problems performance chasers deal with.

User avatar
Samd   1.21 GW

1.21 GW
Posts: 3667
Joined: Jun 28 2011 11:08pm
Location: Ballarat, AU
Contact:

Re: Make a watercooled high kw motor and I'll buy it !

Post by Samd » Jan 24 2017 11:42pm

John is fighting the good fight against the fallacy that different windings of the same motor produce different torque. I am thankful for it.

All the mxus winds for example produce almost the same torque - the 4t is slightly higher than the others due to the slightly higher copper fill.

Its just that a motor should be selected to be in its most efficient speed band. I still suggest say a 6t mxus to people who want to cruise slowly on soft sandbeaches for example.

With respect to heatpaths abovethe bearings only provide surfaces of several square centimetres.

Ferrofluid provides many orders of magnitude higher then the fins take it away from the precise point where it is at risk of degaussing the magnets.

So we are in a new age. No more chasing heavy v2 or v3 aluminium stators.

Check out justins new thru-axle front motor. Its basically a copy of the 6.4kg mxus motor with 29mm stator. But it weighs almost half.

This innovation could see a return to lighter steel stators that don't require expensive casting tooling that aluminium stators do.




Sent from my iPhone using Tapatalk
http://ballaratebikes.com/

ImageImageImage
"Knowledge is knowing that a tomato is a fruit. Wisdom is knowing that a 165mm rear shock can't deliver 220mm of rear swingarm travel." - Oprah Winfrey.

Jestronix   10 kW

10 kW
Posts: 508
Joined: Jul 18 2015 4:59am

Re: Make a watercooled high kw motor and I'll buy it !

Post by Jestronix » Jan 26 2017 7:15pm

John in CR wrote:Jestronix, you're asking for the wrong thing. The best way to deal with heat is to not make it to begin with. That means a more efficient motor. People jump thru ridiculous hoops to make crap motors disperse heat better, when for the same price they could have had hubmotors to run at up to 10-15kw in stock sealed form without unnecessary and barely effective liquid cooling of an outrunner. Sure they took some work to fit, but that pales in comparison to the effort needed to make low efficiency motors dump their heat. Ask for better hubmotors, not motors with liquid cooling.
Weight is an issue for me in the rear hub, those 10kw things are massive! i want less unsprung weight. without the complications of mid drive.

User avatar
macribs   1 GW

1 GW
Posts: 3488
Joined: Jul 22 2014 5:59pm

Re: Make a watercooled high kw motor and I'll buy it !

Post by macribs » Jan 26 2017 9:42pm

Jestronix get in touch with the dude behind this motor, according to some of the comments it seems he does custom work too.
Iirc his 45kw motor with water cooling is 17 kilos, 3500 RPM. But if he builds custom motors I am sure you get to decide what factors to prioritize so that you get a kv that suits your need. This is not dd hub but would work for mid drive. </obsolete text>

[youtube]2LNfDI3QUpQ[youtube]



That video has since been made private, so I guess custom work is out of the question. Might wanna look into dlc 28 motor or the dlc 28 clone, that samd mentions further down on this page. Btw samd even got a kick ass controller (mobipus 72200/72600) to match. Sine wave Field weakening with real dashboard ie on a smartphone or tablet





..
Last edited by macribs on Feb 07 2017 4:32pm, edited 9 times in total.

Jestronix   10 kW

10 kW
Posts: 508
Joined: Jul 18 2015 4:59am

Re: Make a watercooled high kw motor and I'll buy it !

Post by Jestronix » Jan 27 2017 4:48pm

Imagine the cost on this ! Custom custom, insane motor though.

User avatar
Samd   1.21 GW

1.21 GW
Posts: 3667
Joined: Jun 28 2011 11:08pm
Location: Ballarat, AU
Contact:

Re: Make a watercooled high kw motor and I'll buy it !

Post by Samd » Jan 28 2017 3:36pm

Jestronix, Sketch isnt far from you and has a dlc28 clone and mobipus 72600 controller for building up his dirtbike. It'll do 42kw no prob. 22kg.


Sent from my iPhone using Tapatalk
http://ballaratebikes.com/

ImageImageImage
"Knowledge is knowing that a tomato is a fruit. Wisdom is knowing that a 165mm rear shock can't deliver 220mm of rear swingarm travel." - Oprah Winfrey.

User avatar
macribs   1 GW

1 GW
Posts: 3488
Joined: Jul 22 2014 5:59pm

Re: Make a watercooled high kw motor and I'll buy it !

Post by macribs » Jan 28 2017 8:05pm

Samd wrote:Jestronix, Sketch isnt far from you and has a dlc28 clone and mobipus 72600 controller for building up his dirtbike. It'll do 42kw no prob. 22kg.

That is some sweet controllers, and serious cooling on both. Great vid btw. Look forward to see you power up that 72600 :twisted:
That 72600 seems by you video to be a lot smaller then the Apt 96600, do you remember the size? I really liked the dashboard, so freakin cool. That tablet will be awesome dashboard. Keep posting vid's sam, look forward to see more.

Just for fun, do you know the price for the dlc28 clone? 22 kilo seems a bit high. Iirc the dlc28 was 15 kilos? But hey, its a mid drive so really no reason to worry about some extra kilos if the price is right.

Post Reply