Definitive Tests on the Heating and Cooling of Hub Motors

Get all your technical information about electric bikes here.
User avatar
MrDude_1
100 kW
100 kW
Posts: 1007
Joined: Aug 17, 2012 3:07 pm
Location: Charleston, SC

Re: Definitive Tests on the Heating and Cooling of Hub Motor

Post by MrDude_1 » Oct 30, 2017 8:09 am

zzoing wrote:If you take aluminium mesh with .7mm thick strands and you roll it into a kind of bandana arrangement, you can glue it with arctic adhesive around the outside of the motor and increase the heat dissipiation by 2x, it has the advantage that when you throw water on it will act as a sponge too. If you stick a roll of it to the hub at 45 degrees, all the strands at the base of the roll will be directly conducting heat away from the hub. it can multiply the hub surface area by 5 times for 20 dollars including 2x arctic thermal adhesive. it's made from aluminum oxide and boron nitride and rubbery glue.

I sometimes find myself wetting plants and moss in the forest at the top of a mountain to cool the motor down, aluminium mesh would be cooler.
I need a picture of what you did to visualize this... sounds interesting.
Warning:This post is being read via light, a substance known to the state of California to cause cancer.

User avatar
madin88
10 MW
10 MW
Posts: 2328
Joined: May 27, 2013 2:02 am
Location: Austria

Re: Definitive Tests on the Heating and Cooling of Hub Motor

Post by madin88 » Oct 30, 2017 12:10 pm

zzoing wrote:If you take aluminium mesh with .7mm thick strands and you roll it into a kind of bandana arrangement, you can glue it with arctic adhesive around the outside of the motor and increase the heat dissipiation by 2x
brilliant idea^^

User avatar
Allex
100 MW
100 MW
Posts: 2898
Joined: Dec 05, 2011 8:46 am
Location: Stockholm, Sweden

Re: Definitive Tests on the Heating and Cooling of Hub Motor

Post by Allex » Oct 30, 2017 3:18 pm

Meanwhile in Russia, they take another way to cool down the motors.
It seems that ferrofluid is not enough for them so they actually use Active cooling and it is far more complex one.

The original creator "Chupa"
http://electrotransport.ru/ussr/index.p ... le;u=20897
Came up with the idea to use double-fans for servers, not any fans but a very specific type that has a very high pressure and sealed bearings for long life. At the air intake they install Airfilter.
Image

Image

This creation is for a MAC 9T motor. He runs a MAC setup and pumping it with over 6kW(8kW peak) with a kelly 72121 The motor never reach above 75C at 75km/h And I've seen about 80km/h in the below video, thats insane on a Mac(it does of course have modified gears)

Image

Image

Image



Image

Image

They use this with MXUS 3000 motors as well, and at 15kW peaks it never overheats.


And here is a run on with the mac ,you can see the speed on the right side of handlebar, temp is on the stem, showing a maximum of 70C at about 75km/h jump to 28:50
Last edited by Allex on Nov 07, 2017 3:36 am, edited 1 time in total.

User avatar
Allex
100 MW
100 MW
Posts: 2898
Joined: Dec 05, 2011 8:46 am
Location: Stockholm, Sweden

Re: Definitive Tests on the Heating and Cooling of Hub Motor

Post by Allex » Oct 30, 2017 3:33 pm


User avatar
Cowardlyduck
10 MW
10 MW
Posts: 2290
Joined: Jun 26, 2011 8:41 pm
Location: Canberra, ACT, Australia

Re: Definitive Tests on the Heating and Cooling of Hub Motor

Post by Cowardlyduck » Oct 30, 2017 4:45 pm

DAM! Thanks for posting that Allex.
That makes my fan cooling setup look like a gentle breeze!

I work in I.T., and have worked with (and right next to :roll: ) a whole bunch of servers over the years. While their fans can and do push a lot of air, not only are they extremely LOUD, but they also have a really annoying pitch/whine to them.
I think those kind of fan cooling setups would only work well for riding in a noisy city with traffic or at high speeds on the highway where other noises drown out the fan noise. They would completely ruin the ambience of the outdoors if you used them in the forest or bush.
No doubt the cold Russian air really helps these setups work well also!

Seeing these setups really validates my suspicions about fan cooling having the highest potential cooling capability. Don't get me wrong, I still love FF and hubsinks...I have a set + 50ml of FF ready to go for my next build...but the FF + hunbsink combo has a thermal dissipation limit that is not too hard to reach if you really start pushing things. This is less the case with fan cooling...if things start getting too hot, just push more air through. :)

I've tried it with limited success, but I still think the absolute best cooling is a combination of all 3. FF + Hubsinks + fan cooling. If the windings could be better protected without reducing the thermal conduction to the FF and air, this combination would be the best by far. I'm actually still running my Leaf motor like this currently, but I fully expect the FF to be contaminated and the windings to be very dirty when I finally get around to opening it up.

Cheers
High Power LiPo wiring harnesses - 4P - XT90, HXT4mm, 5.5mm. 200Amp+ capable. Global shipping.

Modified Stealth Fighter - Force air cooled Leaf motor @ 5KW, Heat-sinked Adaptto Mini-E. Battery = 20AH, 72V LiPo.
Cowardlyducks - Stealth Fighter Videos

BikeE recumbent commuter - Golden motor, 6Fet Infineon, 20AH 12S LiPo, with on-board solar charging.

BLAKE'S BUILDS - The stuff I make and modify.

User avatar
amberwolf
100 GW
100 GW
Posts: 23936
Joined: Aug 17, 2009 6:43 am
Location: Phoenix, AZ, USA, Earth, Sol, Local Bubble, Orion Arm, Milky Way, Local Group
Contact:

Re: Definitive Tests on the Heating and Cooling of Hub Motor

Post by amberwolf » Oct 30, 2017 9:19 pm

Cowardlyduck wrote: I work in I.T., and have worked with (and right next to :roll: ) a whole bunch of servers over the years. While their fans can and do push a lot of air, not only are they extremely LOUD, but they also have a really annoying pitch/whine to them.
Regarding computer fans in general, but especially high-volume fans:

FWIW, many of them are louder and whiny because of the closely-spaced grills front (and back), or close placement to components that cause enough blockage of airflow that the fan motor is loaded more, or blades are deformed in airflow, or a partial vacuum occurs behind the fan, etc. (like putting your hand over the hose of a vacuum cleaner).

Sometimes it's caused by insufficient intake openings (when the fans are exhaust) or vice-versa.


Removing those grills (which are only needed to protect the careless from themselves, including those that don't tie wires down inside the case away from fans) decreases noise and improves airflow. Increasing the size of the intake holes (or exhaust) at the other side of the case from the fans does the same thing.

Changing from screens/metal with lots of little punched holes to larger directed vent holes helps too.


Back-to-back fans are often problematic by nature; they generally cause the vacuum (or overpressure area) that makes for the bad whine. Turbulent air between the two sets of blades does it too. :/

User avatar
Rix
100 GW
100 GW
Posts: 5951
Joined: Mar 29, 2012 11:26 am
Location: Fallon NV USA

Re: Definitive Tests on the Heating and Cooling of Hub Motor

Post by Rix » Oct 30, 2017 11:03 pm

Allex wrote:Meanwhile in Russia, they take another way to cool down the motors.
It seems that ferrofluid is not enough for them so they actually use Active cooling and it is far more complex one.

They use double-fans for servers, not any fans but a very specific type that has a very high pressure and sealed bearings for long life. At the air intake they install Airfilter.
[i
This creation is for a MAC 9T motor. He runs a MAC setup and pumping it with over 6kW(8kW peak) with a kelly 72121 The motor never reach above 75C at 75km/h And I've seen about 80km/h in the below video, thats insane on a Mac(it does of course have modified gears)

here is a run on with the mac ,you can see the speed on the right side of handlebar, temp is on the stem, showing a maximum of 70C at about 75km/h jump to 28:50
Sheesh, and here I thought I was on the edge with the idea of running 3kw through mac 10 motor with a metal gear and oil bath cooling.

User avatar
rberger
100 mW
100 mW
Posts: 35
Joined: Mar 09, 2015 2:10 am
Location: Saratoga, CA

Re: Definitive Tests on the Heating and Cooling of Hub Motor

Post by rberger » Oct 31, 2017 3:16 am

If one were to have Statorade and hubsinks, would having just a good fan blowing over the hubsinks be a significant help vs. just the Statorade and Heatsinks?
My use case isn't high peak power but long steep hill climbs on a trike at ~15mph with a Crystalyte H3540.
2011 ICE Adventure Trike Converted to 1500W DD Hub Pedalec in 2017

User avatar
macribs
1 GW
1 GW
Posts: 3132
Joined: Jul 22, 2014 5:59 pm

Re: Definitive Tests on the Heating and Cooling of Hub Motor

Post by macribs » Oct 31, 2017 11:12 am

For those of you riding hard, climbing steep and often get into situations where temperature rises so much you need to stop I can think of one solution that will make the cooling even more effective then FF+heatsink yet still silent.

When the dd hub gets warm, the FF will transfer the heat to the heat sink. Then the liquid inside the pipes will transfer into vapor. And the gas get pushed to the cooler part of the pipe, along the out perimeter of the heat sink. As the vapor/gas rapidly cools
down in contact with the large surface area of the fins from wind/air the gas turns into liquid again, and in essence you got a closed loop liquid cooling ontop of the FF+Heatsink combination.

Heat pipes are very well suited to remove vast amounts of heat.
WikiP wrote:Due to the very high heat transfer coefficients for boiling and condensation, heat pipes are highly effective thermal conductors. The effective thermal conductivity varies with heat pipe length, and can approach 100 kW/(m⋅K) for long heat pipes, in comparison with approximately 0.4 kW/(m⋅K) for copper.


Heat pipes are silent, maintenance free, highly efficient heat conductors. Very well suited to add another level of heat removal from our DD hubs.

In fact, if there was enough demand sketch once said he would do a "macribs edition" of the heatsink, with heat pipes.

User avatar
bigbore
1 kW
1 kW
Posts: 389
Joined: Sep 25, 2009 7:43 am
Location: Italy

Re: Definitive Tests on the Heating and Cooling of Hub Motor

Post by bigbore » Oct 31, 2017 1:03 pm

Rix wrote:
Allex wrote:Meanwhile in Russia, they take another way to cool down the motors.
It seems that ferrofluid is not enough for them so they actually use Active cooling and it is far more complex one.

They use double-fans for servers, not any fans but a very specific type that has a very high pressure and sealed bearings for long life. At the air intake they install Airfilter.
[i
This creation is for a MAC 9T motor. He runs a MAC setup and pumping it with over 6kW(8kW peak) with a kelly 72121 The motor never reach above 75C at 75km/h And I've seen about 80km/h in the below video, thats insane on a Mac(it does of course have modified gears)

here is a run on with the mac ,you can see the speed on the right side of handlebar, temp is on the stem, showing a maximum of 70C at about 75km/h jump to 28:50
Sheesh, and here I thought I was on the edge with the idea of running 3kw through mac 10 motor with a metal gear and oil bath cooling.
Is my opinion only that they ride in a dangerous way trough the traffic? I don't ride as a grandma my Honda VTR 1000 F but it is a real motorbike at least.

User avatar
madin88
10 MW
10 MW
Posts: 2328
Joined: May 27, 2013 2:02 am
Location: Austria

Re: Definitive Tests on the Heating and Cooling of Hub Motor

Post by madin88 » Oct 31, 2017 1:29 pm

bigbore wrote: Is my opinion only that they ride in a dangerous way trough the traffic? I don't ride as a grandma my Honda VTR 1000 F but it is a real motorbike at least.
russia has differnt laws.. i believe it is not illegal at all how they ride :lol:

User avatar
John Bozi
1 MW
1 MW
Posts: 1903
Joined: Apr 09, 2013 1:46 am
Location: Brisbane, Australia
Contact:

Re: Definitive Tests on the Heating and Cooling of Hub Motors

Post by John Bozi » Nov 18, 2017 2:59 pm

That's a nice video of mate-ship riding bikes enjoyed it.

All the complexity of that set up on a Mac however boggles my mind though. For the type of riding in the vid a bigger hub and the usual sinks and ff would keep just as cool if not cooler. Open road riding is the right gearing so heat should be a non issue even without any mods on a large hub.

Great share though, the ingenuity of the human mind to work with what we got is very interesting.

Thanks Alex for the share.

User avatar
madin88
10 MW
10 MW
Posts: 2328
Joined: May 27, 2013 2:02 am
Location: Austria

Re: Definitive Tests on the Heating and Cooling of Hub Motors

Post by madin88 » Nov 19, 2017 7:05 am

John Bozi wrote:
Nov 18, 2017 2:59 pm
All the complexity of that set up on a Mac however boggles my mind though. For the type of riding in the vid a bigger hub and the usual sinks and ff would keep just as cool if not cooler. Open road riding is the right gearing so heat should be a non issue even without any mods on a large hub.
Maybe the benefits of less (unsprung)weight had been the reason for the modifications.
in terms of torque the MAC can deliver it would at least need a direct drive hub with 7-8kg
the MAC can be laced into bicycle rims, whereas for bigger hubs it needs moped rims for a durable wheel.

User avatar
John Bozi
1 MW
1 MW
Posts: 1903
Joined: Apr 09, 2013 1:46 am
Location: Brisbane, Australia
Contact:

Re: Definitive Tests on the Heating and Cooling of Hub Motors

Post by John Bozi » Nov 19, 2017 1:37 pm

Nothing wrong with lighter obviously, but it is road riding.

Bullfrog
1 mW
1 mW
Posts: 15
Joined: Apr 08, 2017 10:37 am

Re: Definitive Tests on the Heating and Cooling of Hub Motors

Post by Bullfrog » Nov 30, 2017 11:29 am

Below are the details and results for my experiment using Low Viscosity ATF to cool a MAC IGH motor. The small diameter rear wheel was used because with an IGH motor, changing the diameter of the rear wheel is the only way to change the gearing between the electric motor and the ground. A word of caution...it only takes a tiny amount of ATF on your brake rotor to cause a serious loss of braking capability.

Bike - Copy.jpg
Bike - Copy.jpg (206.09 KiB) Viewed 256 times

MAC ATF Cooling Experiment

Objective:
The objective of these experiments was to determine if adding Automatic Transmission Fluid to a MAC Internally Geared Hub (IGH) motor can sufficiently reduce the temperature of the motor to prevent potential damage when riding off road.
Equipment used and starting conditions:
A. The motor used was a 12T MAC Internally Geared Hub motor installed in a 20” rim. The wheel assembly was mounted in a Mongoose Terrex bicycle. The rim is 57 mm wide (internal) and the tire is 4" wide.
B. A Cycle Analyst v3 was used to display the temperature and the wattage.
C. The battery used was a nominal 52v 14s6p configuration.
D. ATF being used was “low viscosity” and meets the Dexron VI and Mercron LV spec.
E. The ambient temperature for the entire experiment was 20 degrees C (67 F).
F. The starting motor temperature was 70 degrees C every time the experiment was conducted. This was accomplished by riding to the starting point and if the motor was too hot then it was allowed to cool until it reached 70 degrees C and if it was too cool then the brakes were applied along with the throttle and the bike was allowed to roll slowly until the motor temperature was 70 degrees C.

Procedure
A. Record the full throttle no load wattage with no ATF added.
B. Add .5 ounces of ATF and record the full throttle no load wattage.
C. Lean the bike 30 degrees from vertical and visually determine if any ATF runs out of the fill port with the fill port in the 6 o’clock (lowest) position. The fill port being used was one of the brake rotor bolt holes.
D. Record the full throttle no load wattage.
E. Ride the test section and record the temperature.
F. Add .5 ounces of ATF and repeat steps C thru E.
If at any point ATF runs out of the fill port when the bike is leaned 30 degrees with the fill port in the six o’clock position, the experiment is complete. A 30 degree lean angle is beyond any expected riding condition but was selected to allow a sufficient safety margin to prevent leakage under normal operating conditions.

Test data/results
Based on past experience the temperature of the motor with a lithium based grease in it for lubrication resulted in motor temperatures around 110 C when the test section used for this experiment was ridden.

Run #1 - Internal components of the motor were clean and no ATF was added:
Full Throttle No Load Wattage: 90 w
Temperature at the end of the test section: Not ridden due to a lack of lubrication in the motor.

Run #2 - 0.5 ounces total ATF in the motor:
Full Throttle No Load Wattage: 90 w
Temperature at the end of the test section: Not ridden due to a lack of lubrication in the motor.

Run #3 - 1.0 ounces total ATF in the motor:
Full Throttle No Load Wattage: 100 w
Temperature at the end of the test section: 105 C

Run #4 - 1.5 ounces total ATF in the motor:
Full Throttle No Load Wattage: 90 w
Temperature at the end of the test section: 110 C

Run #5 - 2.0 ounces total ATF in the motor:
Full Throttle No Load Wattage (FTNLW): 80w
Temperature at the end of the test section: 110 C

Run #6 - 2.5 ounces total ATF in the motor:
Full Throttle No Load Wattage (FTNLW): 100 w
Temperature at the end of the test section: 96 C

Run #7 - 3.0 ounces total ATF in the motor:
Full Throttle No Load Wattage (FTNLW): 110 w
Temperature at the end of the test section: 86 C

Run #8 - 3.5 ounces total ATF in the motor:
Full Throttle No Load Wattage (FTNLW): 107 w
Temperature at the end of the test section: 83 C

After adding 3.5 ounces total ATF to the motor a very small amount of ATF was observed departing the motor when the bike was leaned 30 degrees therefore this ended the experiment by definition.

Conclusions
A. 2.5 to 3.5 ounces of low viscosity ATF can aid in cooling a MAC IGH motor.
B. Minimal increases in the viscous drag were observed as expected but considered acceptable based on the decrease in temperatures.
C. 2.0 ounces or less of ATF does not appear to add any cooling benefit.
D. More than 3.5 ounces of ATF could lead to leakage and the associated safety concerns regarding braking.

Recommendations
A. Conduct similar experiments in a laboratory setting where all of the variables could be controlled, monitored and recorded. This experiment was conducted as carefully as possible utilizing the available resources.
B. Utilize additional fluids to determine if low viscosity ATF provides the best benefit based on viscous drag and temperature reduction.

Supporting Data and information
Dimensions:
1. Motor Case ID ~ 5.85”
2. Outer Motor OD ~ 5.76”
3. Internal Case Width ~ 1.55”
4. Initial fill volume is .5 ounces…this should provide some lubrication for the planetary gears but not provide any additional cooling. The approximate volume between the Motor Case ID and the Outer Motor OD is 1.24 cubic inches which is equivalent to .7 fluid ounces.
5. ID of the motor copper windings ~ 3.68”
6. OD of the motor copper windings ~ 5.15”
7. Break Rotor bolt circle diameter ~ 1.68”
8. With the Mongoose 20x4” tire being used: 10 mph = 150 rpms of the rear wheel.
9. ATF being used is “low viscosity” and meets the Dexron VI and Mercron LV spec.
10. Permatex Ultra Gray was used to seal the interface between the hub and the motor cover.
11. Permatex Ultra Black was used to seal five of the six brake rotor bolts. The sixth bolt was not sealed and use for the fill port.
12. No leakage of ATF was observed other than that stated in iteration #8.

These experiments were conducted on 22 November 2017.

User avatar
wturber
1 kW
1 kW
Posts: 313
Joined: Aug 23, 2017 8:52 pm
Location: Fountain Hills,AZ
Contact:

Re: Definitive Tests on the Heating and Cooling of Hub Motors

Post by wturber » Nov 30, 2017 4:55 pm

Very nicely done. Not an area that I'm personally concerned with, but it is nice to see some well thought out testing like this. Testing like this can be pretty time consuming - which is probably partly why we don't see more of it.
"Commuter - DC Booster"
Iron Horse 3.0 hardtail - 48V / 1000W / 470rpm generic Chinese DD Hub motor (ebay)
7 x 36v 4.3ah 10s 2P battery packs - Wangdd22 1500W 30A DC Boost Converter delivers 54v and about 1175 watts peak
53T/42T Sakae Road cranks - 30mph+ on flats
https://endless-sphere.com/forums/viewt ... =2&t=90369

Bullfrog
1 mW
1 mW
Posts: 15
Joined: Apr 08, 2017 10:37 am

Re: Definitive Tests on the Heating and Cooling of Hub Motors

Post by Bullfrog » Nov 30, 2017 10:10 pm

Thanks...have to give some credit to the guys at Grin Tech: http://www.ebikes.ca/.

They have been very helpful and IMO are some of the sharpest around.

markz
10 GW
10 GW
Posts: 4737
Joined: Jan 09, 2014 11:38 pm
Location: Alberta Canada

Re: Definitive Tests on the Heating and Cooling of Hub Motors

Post by markz » Nov 30, 2017 11:10 pm

Either there was too much ATF fluid weight when wheel was spinning or the fluid somehow grew in mass then therefore leaked.

Different viscosity and weight of ATF. It seems the cheapest cooling solution if you got some laying around in your garage.

User avatar
zzoing
10 kW
10 kW
Posts: 659
Joined: Feb 01, 2008 12:26 am
Location: Fr SE

Re: Definitive Tests on the Heating and Cooling of Hub Motors

Post by zzoing » Dec 10, 2017 3:28 pm

Hi, Sorry I took a while to see your messages about an aluminium bandanna around the hub, I drew a pic on the previous page.

You know how you can buy 100g of nanotubes and graphene quite easily these days, depending on the grade, like 90% and so forth...

Perhaps it would be possible to anodize nanotubes straight into the hub, someone was talking about anodizing aluminium and I was looking at nanotube prices recently and it would be cool to anodize a hub with nanotubes :D the 1st guy to have a carbon nanotubes back wheel would be kindof crazy. and it probably doesnt transfer to air that fast, perhaps a 130% thermal advantage?

User avatar
amberwolf
100 GW
100 GW
Posts: 23936
Joined: Aug 17, 2009 6:43 am
Location: Phoenix, AZ, USA, Earth, Sol, Local Bubble, Orion Arm, Milky Way, Local Group
Contact:

Re: Definitive Tests on the Heating and Cooling of Hub Motors

Post by amberwolf » Dec 11, 2017 11:33 pm

markz wrote:
Nov 30, 2017 11:10 pm
or the fluid somehow grew in mass
Huh? I'm sure you meant something else, because mass can't increase on it's own--you'd either have to add more to do that, or convert energy into mass (not a likely thing to happen inside a hubmotor ;) ).

Post Reply