Cooling fans inside Hub motors

Great stuff offroader!

I also have new plans. After I recently accidentally pumped 75V through my 7 of my 7.4V rated rated fans (so 10.7V each) and fried them, I now plan to replace them with 12 fans in my Leaf Motor.
This way, when running 75V each fan only see's 6.25V, and if I jump back down to a 50V setup, they should still work at 4.16V each, but be much quieter.
With 12 fans, 6 on each side of the stator in a push/pull configuration, the air flow should be much improved, even with them running at slower speeds, meaning less noise + better cooling...win win :D
I'm not going to get anywhere near the air flow of OffRoaders EDF, but as I've found, even a small amount of air moving through the stator has a large impact on cooling.
Maybe this can be version 12.0. :)

Cheers
 
When I ran my EDF in the closed motor I actually overheated my EDF and it broke the impeller on it. Lost $26 dollars as i have to replace it. Oh well, not happy about it.

I knew this would be an issue running the fans in a sealed motor because I assume the heat was around 115C being sucked through the fan. When I turned it on Max it the force was enough to break it because the plastic was heated very hot.

I also had high temp melt glue that had also started to run inside the motor. The heat it was sucking around was that hot inside to melt this.

The funny part was even though I was circulating this heat so quickly around the motor the exterior of the motor was not getting hot very quickly. This is why I question how good FF + hubsinks can be. I've asked numerous people to test the hubshell temp when the motor is spinning as I've found that mine sheds heat so quickly that I qusetion the usefulness of hubsinks unless the motor is stopped.

It seems most people don't understand that yes FF could work, but not nearly as good as passing air through the motor and vented out.

 
The biggest draw back for fans I can see is dust build up. Add to that a bit of moisture and then you have mud gunk.

I have run quite large holes venting my old clyte a while back. Every ride was as can be expected a total covering of every part of my bike in dust and or mud if moist.

Having large holes with the circumference at gap meant I could use air pressure to blow out from one to the other, I'd guess about 80%. But when I'd open up the motor there was always super fine dust in between magnets.

With the one sided design in the diagram I can only imagine that there will be a good build up of material on one side in particular as particles get bounced around not being able to follow air so well.

I often get prooved wrong,

good luck.
 
Your not wrong John, dust build up is the biggest problem with active fan cooling, however it's not as big of a deal as you might think, even riding in/through very dirty/dusty environments with the fans running the build up takes some time.
I've found it takes at least 2000km of use before the build up is bad enough to warrant a decent cleanout and even then it's not that bad.

I've never let it go much longer than about 2000km before cleaning at least a small amount. The build up does not seem to effect cooling performance much, and the bearings are fairly shielded, so theoretically I think you could run cooling fans for 10000km+ before the dirt/dust build up would actually cause any problems.

Cheers
 
Cowardlyduck said:
Your not wrong John, dust build up is the biggest problem with active fan cooling, however it's not as big of a deal as you might think, even riding in/through very dirty/dusty environments with the fans running the build up takes some time.
I've found it takes at least 2000km of use before the build up is bad enough to warrant a decent cleanout and even then it's not that bad.

I've never let it go much longer than about 2000km before cleaning at least a small amount. The build up does not seem to effect cooling performance much, and the bearings are fairly shielded, so theoretically I think you could run cooling fans for 10000km+ before the dirt/dust build up would actually cause any problems.

Cheers
Click to expand...

I thought with the fans running it would self clean to some degree ?

Cheers Kiwi
 
kiwiev said:
I thought with the fans running it would self clean to some degree ?

Cheers Kiwi
Click to expand...
Yeah they do to some degree.
But similarly to radiators and computer fans, it's the super fine dust that still sticks and builds up over time. Doesn't matter how fast the air flow is, there will always be a level of dust build up with any active air cooling method unless you put some kind of air filter in, but then you would need far more powerful fans to pull air through the filter at a decent rate adding lots of complexity. Basically it's more hassle than it's worth, and I would rather just periodically clean out the motor.

Cheers
 
How would you clean the windings after lots of dust buildup? I'm thinking just running water through it and using compressed air to blow it dry.
 
Offroader said:
How would you clean the windings after lots of dust buildup? I'm thinking just running water through it and using compressed air to blow it dry.
Click to expand...
Haven't tried that, but not sure if I want to. Any remaining water could accelerate rust, and with my recent experiments with FF stripping away my insulating varnish layer, I'm not keen to try using water.

I actually use baby wipes. They capture dirt/dust super well and are easy enough to get into most hard to reach places. I don't bother with cleaning in the tiny nooks and crannies, and so far that hasn't been a problem.

Cheers
 
Just ride it with water in it, will dry it out for sure.

This is my mate riding through with a vented hub. The steam was sensational.

https://youtu.be/bJzsa_9QeWE?t=1m25s

Not sure about your fans though. Another mate just hosed his out too and then ran it to dry it out.
 
Did my first testing of a small piece of silicone that will be used as an air duct to control the flow of air in the hub motor. Placed a little bit on the motor to see how well it will stick. The silicone holds on very strong, I would really have to pull hard to rip it off, even on the windings

The big piece of silicone is to seal to the motor side cover and stop the air from circulating inside the motor, it allows fresh air in and hot air exhausted out.

What is modified compared to my last cooling setup, is the following. I will put silicone on the center of the windings to stop the air flow from flowing above the windings like it has in my Cromotor. The space was so large above the windings on the cromotor, as large as a finger, and the air would just flow on top instead of through the windings.

The silicone is just placed on the top and not pushed into the windings so it should not impact the air flow through the windings. It will seal against the motor case above it, forcing all air below it and through the hot copper windings.

It should increase the air cooling by a lot. This may also make the airflow more turbulent at that location to also better cool the windings.





 
Just wanted to update everyone on what could be the last motor to be modified with a cooling fan. Most everyone is switching over to Ferro Fluid and mostly everyone is against air cooling now.

Maybe it works just as good or better, I'll have to eventually switch over and compare the results. My air cooling simply works and eliminates overheating, will FF do just as well?

I will easily put my motor up against any FF motor and I would expect to blow a FF cooled motor away.

Right now, I have completed my air cooling, except for the exterior air duct which will stop a lot of the air from recirculating back into the motor. So my setup will only get better when I make this air duct.

I did a test today in 65F / 18C weather and everything worked great, I could not overheat the motor on any of my on-road or off-road riding. I was riding the bike as hard as I could with constant full throttle acclerations. This was with a slow fan speed of about 25%. The most severe full throttle hill climb (basically as steep a dirt hill as you can ride up) I could get the motor to about 105C, but it would cool down to 75C within 30 seconds after the climb, this was also with a slow fan speed of 25%.

This is the 2nd motor I modified for air cooling, and both motors stopped overheating. FF supposedly stops overheating also. The question is will it stop overheating for my harsh riding and hill climbing that I do? Will it deal with constant stops and then full accelerations? This is why I will have to test it personally because most people just don't push their bike like I push mine.

The last thing I want to deal with is after riding up a steep hill, have to wait for the thermal transfer out of the windings and into the stator, then through the FF and then through the magnet ring. Basically, there could easily be bottlenecks in the heat transfer when using FF. With my setup, I'm blowing cool air directly on the windings and blowing this hot air directly out of the motor.

My setup is race ready, basically, ride as hard as you like and never have to wait for cooling or deal with overheating.

My cooling setup pushes out a constant stream of very hot air, it pushes it straight off the windings and out of the motor. It also will cool the stator, if that gets hot. So compared to FF, it has the advantage of cooling both the windings and the stator, and the windings are much more important to cool during those peak load times when you are really pushing your bike to the max.

If FF will work to stop overheating, then it makes sense to just use FF instead of drilling open the motor.





 
i likeit mate... i think part of the reason ff is popular is it keespyour motor sealed... not a major issue but better sealed than not when it comes to ingress of mud etc. its also simpler... drill a small hole to add the ff, ad the hubsinks, done.
that said, i tend to agree in principal - if both methods move the same amount of air over the same surface area... then yours will win hands down simply because its moving air directly over the heat source, not requiring it to go through 5-6 other mediums before it gets to open air. those 5-6 mediums may have a low r value bit its still a resistance >0. added to this your moving air even at standstill... so better low speed dissipation (assuming all else equal)

for interest, what phase currents are you running, and whats teh phase resistance of your motor, just to give an idea of your peak heat production?
also appologies if i missed it, but why th/e 2 circles of (what ooks like) silicon seals? (nvm... missed the preceeding post) - how much air flow actuall goes through the windings? id have thought the gaps would be too small to allow any notable airflow...
 
Offroader, is that second silicon ring on top of the copper windings really necessary? Is that to make sure that the air is guided between the stator teeth and the magnets perhaps? Anyway, good work and I really like the idea of your air cooling mod.
 
Thanks for the responses.

The second silicone ring was to push more of the airflow through the windings. What happens is that most of the air comes through the stator/magnet gap and will directly flow right to the exhaust air holes in a nice direct path, probably mostly blowing above the windings.

This silicone ring will force that air downwards, into the windings,. The issue here was that I thought the gap between windings and side cover was larger so I used too much silicone and it came wider than I wanted it to, covering more of the windings. I tried to cut away as much excess as I could, but wish it was not as wide to cover less of the windings. I don't think it matters all that much as the the air flow will hit most of the windings under the silicone.

I was planning to do the other side of the windings also, but I'm not sure if I should or it is worth it. I have to be more careful because of the way the air flows on the other side. The silicone would have to be placed on the very outside of the windings and care has to used to make sure I don't block any holes where the windings to through the stator.

A good amount of airflow does get blown through the stator where the windings are. While not as much as through the stator / magnet gap, it still gets a good flow. In the picture below you can see that plastic is wrapped around the windings. There is a lot of air gaps around this, the air will blow through here.
The fan puts a lot of pressure through all these areas and the air is forced through at a decent speed through all those gaps.

With one side cover off and the stator pressed against once side blocking the stator/magnet gap, I can feel the strong air flow coming through those stator winding holes.


I'm pushing 7500 peak watts on my MXUS turbo. Phase amps are maxed out on my max-e, and are set at around 360 phase amps. I'm not sure the resistance on my 4T MXUS turbo motor.

I really punish my motors, with constant stop and then full throttle accelerations. Lots of really steep hill climbs, where I'm full throttle at very slow speeds. The only way to really heat this motors up is through full throttle at low speeds or steep hill climbs. You don't really overheat it at all on the street. You wold need a highway to try and overheat it by using OVS, and would be doing highway speeds for a long time.

I still need to design the external air duct so that the hot exhaust air does not get sucked back into the motor. I'm going to 3d print this. Lots of the hot exhaust air gets sucked right back in, because the holes are so close and on the same side. This duct should really help lower the intake air temps by a lot.

 
Just wanted to add.

I had long thoughts over many weeks on if I should go FF + hubsinks or another air cooling fan mod.

I know from my last air cooling modification that it simply works and is extremly reliable, never had a failure after my first week of building it (I have used it for 2 summer riding seaons already without an issue), and my new motor I just modified was made better with a bigger fan and will use better ducting and an external duct.

I just could not convince myself FF + hubsinks would meet my requirements and the problem is I would have a motor full of FF. Since I use silicone to make the air ducts, it is questionable if I could get all that FF oil off of the windings. Any Oil stuck inside the windings would attract dust and dirt also. I didn't want to try FF first and then not be happy with it and deal with a mess.

Before my cooling modification on my first motor, I could not ride in the summer at all. After only 8 miles on my cromotor, it would overheat and even if I let the motor cool off for 30-45 minutes, I would overheat it within 2 miles of riding. A sealed stock motor is so bad shedding heat.


This also really scared me, here is a post in another thread of someone who oil filled their motor and added his own heat sinks on the exterior. He said that he still has to let his motor cool on hill climbs. This really scared me because he was also using an MXUS motor ( MXUS motors have really bad thermal transfer inside compared to a Cromotor). Me having to stop to wait for a motor to cool is basically out of the question with my setup. I don't believe he was talking about climbing dirt hills either where the hill is so steep you can only ride about 10 MPH at full throttle.

Don't get me wrong, if I ride up a very steep hill at full throttle, where the hill is so steep I can only ride about 10MPH, instead of 50MPH my top speed, I'll overheat my motor eventually. No cooling setup is going to stop that, but that is not something I do, there are no hills that big I can ride that long.

Here is the post below that I didn't like from an oil filled + heat sink user.

https://endless-sphere.com/forums/viewtopic.php?p=1246285#p1246285

Im still on the mxus v2 you sold me :) it gets nice and toasty on my rides, hubsinks and trans oil still can't remove heat on hill climbing. It helps and extends my climbs but I still have to pull over and let her cool down ferro and water cooling would be a good match , bring heat up into the cooling pipes.
Click to expand...

So this postand thread kind of scared me with the whole oil + hubsinks. While FF helps, I don't believe it can do what my cooling setup can. I can also always turn up my fans to a higher speed if necessary.

FF + hubsinks will work well to remove heat over time, I don't think it will work well at those peak heat times. There is no way it is going to transfer the heat quickly enough when your climbing that steep hill.

My bike is built for racing and extreme riding, and that is how I ride it. I need the next level of cooling, and believe I found it with my setup.

Most riders on this forum don't ride their bike like it was a race, for them, FF is the way to go because it will work and is easier.
 
Think I'm still missing something here.

won't there always be a gap between the stationary silicon and spinning side covers even if you have it actually rubbing? a gap which I'm guessing is comparably tight as the windings....
 
John Bozi said:
Think I'm still missing something here.

won't there always be a gap between the stationary silicon and spinning side covers even if you have it actually rubbing? a gap which I'm guessing is comparably tight as the windings....
Click to expand...

I used a piece of tape to give a small gap between the silicone and side cover so there would be no rubbing. It isn't exactly perfectly air tight.

But it doesn't matter. The gap is so tiny about .001 mm , the silicone is about 9 mm, Very little air flow will go around this gap and have no effect.

The bigger issue is outside of the motor where a lot more air is recirculating back into the motor because the holes for the exhaust are right near the holes for the air intake. That is why I intend to design by a 3d printer a duct to push the exhaust air far away from the intake holes.
 
If you're moving, I'd expect that normal airflow (even turbulent) will prevent that re-ingress of old air back in.

When sitting still, it might be an issue, if the intake and exhaust are on the same side near each other.
 
Offroader said:
John Bozi said:
Think I'm still missing something here.

won't there always be a gap between the stationary silicon and spinning side covers even if you have it actually rubbing? a gap which I'm guessing is comparably tight as the windings....
Click to expand...

I used a piece of tape to give a small gap between the silicone and side cover so there would be no rubbing. It isn't exactly perfectly air tight.

But it doesn't matter. The gap is so tiny about .001 mm , the silicone is about 9 mm, Very little air flow will go around this gap and have no effect.

The bigger issue is outside of the motor where a lot more air is recirculating back into the motor because the holes for the exhaust are right near the holes for the air intake. That is why I intend to design by a 3d printer a duct to push the exhaust air far away from the intake holes.
Click to expand...

i wouldn't be too concerned about re-circulation unless your stationary, and even then its only going to be form part of the motor (ie the hot air will rise, so you'll only pull in part of it...)

judging by the numbers you gave, your generating 7.5kw of heat (peak) in the motor (your peak motor amps are around 260A, as you dont have the battery wattage to push more than that given the Turbo has a phase R of about 0.1). this should be dropping off exponentially (1st order) with your speed, up to ~0w at no load... so you can get a pretty good idea of how much heat your generating based on your speed (at full throttle). So if your no load is 70km'h, and your going up a hill at full throttle at 35km'h, then your generating about 1.8kw of heat. your motor at this point is about 75% efficient... just rough numbers but it gives some idea of the peak dissipation this setup can manage. if you want to work it out for any speed:

(1-(current speed/full speed))*260 = phase amps at your current speed (at full throttle, lets call this PhA/v)
((PhA/v)^2)x0.1 = heat generated, approximately as it doens't account for iron loss, but that should be negligible at this power level. 0.1 is your phase resistance from a quick google search.

so using 25km'h (assuming the same 70km'h top speed) we have:
25/70=0.35
1-0.35=0.65
0.65*260=170A (ie pulling 170pha amps at 25km'h at full throttle)

170^2 * .01 = 2.8kw of heat being generated.

it'll all be an inexact science as i doubt you can find a hill that will let you pull 100% throttle at a consistent speed for long enough, but it gets you in the ballpark. if you have a grunty enough power supply you could pump heat directly into the motor and test it that way, as with the fan i dont think spinning it will make a big difference. just set a fan blowing on the motor to simulate some level of speed.

sorry for the rambling post, just some thoughts i had before my morning coffee!
 
The RC EDF I'm using is actually a very strong fan. You can't compare this with a computer fan, it is so much more powerful.

With the bike stationary, if I put on the fan, I actually will feel the air flow only at the very center of the motor when far away from the motor. This is because the intake is actually pulling the exhaust air towards the center. Lots of air is also getting sucked back in and recirculating.

Basically, the exhaust air completely surrounds all the intake holes in the center.

I somewhat agree it isn't all that important because with the way it is now, I don't have any overheating issues. It is also true when moving that not much air will be sucked back in.

However, most of the times I'm overheating is for really slow climbs, where I'm not moving all that fast and there is a higher chance of sucking in hot air.

Since I can easily design on freeCAD and 3d print it, an exterior duct to push that air far away, I see no reason to add it. I already started to design one.


In the image below you can see the center holes are the intake, and they are surrounded by the exhaust holes. When sucking in air, it will pull in a lot of the exhaust air.

I can basically feel the hot air being sucked right back into the motor.

 
Offroader said:
The RC EDF I'm using is actually a very strong fan. You can't compare this with a computer fan, it is so much more powerful.

With the bike stationary, if I put on the fan, I actually will feel the air flow only at the very center of the motor when far away from the motor. This is because the intake is actually pulling the exhaust air towards the center. Lots of air is also getting sucked back in and recirculating.

Basically, the exhaust air completely surrounds all the intake holes in the center.

I somewhat agree it isn't all that important because with the way it is now, I don't have any overheating issues. It is also true when moving that not much air will be sucked back in.

However, most of the times I'm overheating is for really slow climbs, where I'm not moving all that fast and there is a higher chance of sucking in hot air.

Since I can easily design on freeCAD and 3d print it, an exterior duct to push that air far away, I see no reason to add it. I already started to design one.


In the image below you can see the center holes are the intake, and they are surrounded by the exhaust holes. When sucking in air, it will pull in a lot of the exhaust air.

I can basically feel the hot air being sucked right back into the motor.

Click to expand...


if you get some incense sticks you can see the airflow quite well generally...
I do get what you mean by it circling back in, but i suspect when its particularly hot air, allot of it is going to rise away form the motor, and a notable amount of fresh air will still get sucked in. something to keep the hot air away will help, but I wouldn't be surprised if its hard to notice next to how effective your system already is.

Dont forget too that at slow speeds your going to be generating far more heat... the current through your motor is linearly proportional to the speed your going (at full throttle) however the heat generated is exponential, ie you'll generate 4x the heat at 20km'h as you will at 40 (again, at full throttle). You could try dropping the max phase current to ~200A or so, you may find that makes a significant difference to the heat generated, without a big impact on torque (and then only in the bottom 25% of your RPM range).
 
I'm going to have to do more testing. But one thing I noticed is that the stator does act as a good heat buffer for the windings.

I did only 2 runs of a similar climb where I heated my motor up to about 100-105 C.

What surprised me is that after the run, I tested how long it took for the temp to fall from 105 to 70 C.

What I noticed was that the motor does fall quickly from 105 to 70 C in under a minute on a fully heated motor, in both instances where the fan was left running after the climb and the fan turned off after the climb. I had the fan running during the climb to keep things consistent.

What was surprising was that I expected the windings to cool much faster with the fans on. While they did cool faster with the fans on, we are talking about maybe 30 seconds vs 45 seconds. This is probably because the stator was cool in both runs and could easily soak up the winding heat.

What this means is that Cooling the stator is important to act as a buffer during peak climbs.

However, the drop from 105 to 70C, is not super fast, still takes about 30-45 seconds when stopped. This means that there is a bottleneck in the heat transfer from the windings to the stator.

This means that it is still important to cool down the windings during peak motor heating. Ferro Fluid will not cool the windings down during peak heating, like when you are climbing a hill. Ferro Fluid will only slowly cool the stator over time, and then the stator will act as a heat buffer for the windings. Ferro Fluid will probably do very little during those peak motor heating.


This is where air cooling may have the advantage over Ferro Fluid. Air cooling will do both, keep the stator cool to act as a buffer, but during those peak heating times, will help cool the windings also. I'm only running my fans at about 1/4 speed, so I could easily have the fans go to max cooling during peak loads also, which would be a massive amount more air flow over the windings.


I still need to do a lot of testing, I just ran this one quick test and didn't even have a timer. It is still winter here and everything is wet and cold, so I can't ride very often.

I will be running more tests with the fans on and off, different fan speeds, etc.

One of those runs I ran the fan at a slightly lower speed and noticed that I hit a higher peak temp at the end of my run. This tells me that Fan speed may matter a lot in those peak heating times.
 
Another thing I want to add is that I can design a circuit for my fan that when I hit over a certain winding temperature the fans go to maximum speed.

For example, As soon as I hit 90C, fans turn on at max speed, and then when the temp falls below 80C turn fans back to normal speed. This would make it a lot easier than having to manually turn them up or down.

What I also notice is that if I leave my fan running full time, I really only need to keep it at about 10% fan speed to eliminate overheating for all except those few times I'm really climbing hard with the bike.

It doesn't take much air flow to keep the motor cool, you just need some airflow to constantly be flowing through the motor.
 
You must log in or register to reply here.

Similar threads

D
E-BikeKit.com Front Hub Motor Bearing Replacement?
Replies
10
Views
2,199
amberwolf
amberwolf
ecat
Catrike with Grin All Axle front hub motors
2
Replies
31
Views
9,755
eMULE2020
E
E
Latest air cooling mods for ebike hub motor- with drawnings
Replies
1
Views
1,255
ebikedelight
E
LawrenceEaden
Single Gear Stealth Build - Battery in Frame Tubes
2 3 4
Replies
80
Views
14,199
memberseven
M
T
Donor bike for new build\rear hub wheel recommendation.
Replies
6
Views
910
dogman dan
dogman dan
Back
Top