Cooling fans inside Hub motors

[Offroader]

From what I found the cromotor sheds heat very quickly from its side covers. My testing method is not great, but if I let the motor sit for a minute when it's over 100c, it gets very hot and stays hot for a long time. When I start driving for a very small distance, like 15-30 seconds, I touch the motor cover and it's warm to the touch. I can't believe just how fast it cooled down it almost seems as if spinning the motor stops the heat from getting to the side covers.


From what I can determine it seems to shed heat very quickly when moving, the problem is the heat takes too long to get to the side covers. This makes me believe oil will work very well if it speeds up the process of getting heat to the side covers.

I find it interesting that other people with oil had issues with not cooling quick enough. From what I have read most people were very happy with how well oil cooling works. It's messy but it worked.

 

[John in CR]

I find it interesting that other people with oil had issues with not cooling quick enough. From what I have read most people were very happy with how well oil cooling works. It's messy but it worked.

Moderate power only. With GCinDC's motor temp continuing to climb running just 4kw it's painfully obvious why you don't see anyone running the high power you talk about who put ATF in the motor and brag later about how well it works.

Of course the side covers reject more heat while moving, because flowing air creates forced air convection that works much better than natural convection just sitting there. That's why cars have more overheating problems in heavy traffic (less air flow), and why radiators have fans (for more air flow), and even why blowing on your hot coffee or soup cools it faster than letting it sit. While moving the heat will also transfer faster (no slower) from the stator to the covers too for the same reasons, and higher rpms does it better both inside and out, which is part of why smaller wheels are better. Interior blades even with a sealed motor will increase the transfer of heat from the stator to the inside surface of the covers.

Justin did some very controlled testing and shared tons of great info with basic holes in the covers, even with graphs for those lacking the attention span for good reading comprehension. It demonstrated that holes improve cooling, getting the exhaust holes out to the perimeter improved cooling, and that higher rpm improved heat transfer. It also demonstrated that while better than sealed, simply putting holes in side covers regardless of their location doesn't move all that much heat. We're talking hundreds of watts of additional cooling, not the thousands that would be required running at 10kw.

If you're really interested in transferring heat away from your motor, and not being just another holes in a pizza pan guy by copying something that works a little, then it behooves you get a decent understanding of the processes. Start with looking up the heat capacity of air, and the mass of air. Then ballpark how much heat you want the air to take with it and the difference in temperature of the ambient air flowing in and the air coming out of the exhaust, so you can calculate the air flow rate required. It's a real eye opener. Compare it to how much air mr chickenduck can possibly push through the motor even with the most optimum estimates. Those little fans are designed to help get rid of maybe 100W of heat with lots of help of good conduction and good heat sinks and close proximate. Of course they help cool a motor that is stopped faster. Raise your hand if you needed a graph to understand that?

I don't post the power I run and the load my motor pushes along with the max temperatures my motor hits to brag about anything. It's to help people understand how much heat is being created in my system, and rejected with my approach to cooling. It's not magic, and heat can't disappear, and even though my motors are high efficiency, they still produce plenty of heat at the power I run, thousands of watts of heat. That also means no graph is necessary to know with certainty that thousands of watts of heat are being rejected by the air flow design. Hell HillsofValp suffered with heat problems for a year and recently burned up some wiring with a nearly identical design simply because he was too hard headed to listen and simply install the exterior fan blades his motor was all set up and ready to bolt them on.

John

 

[Offroader]

Thanks john,

I looked more into GCinDC's posts about his oil cooling and it didn't look as good as I would have expected. I do know that climbing can be really tough on a motor, really steep stuff seems to raise temp by like 1 deg c per second.

I don't think you can ever cool these small hub motors down during heavy duty climbing or heavy load with any type of cooling, when I say climbing, I'm not talking about a steep street. I'm talking about hard off-roading with steep hill climbs in the dirt.


The way to go may be air cooling with at least one blower in the motor for some heavy duty cooling. I think you're always going to have to take breaks while doing hard off-roading for cool-downs, but at least you can cool the motor down like 10 times faster with fans and a blower.

Really, there isn't that much I can find about people oil cooling there motors and their results.

Good tip about learning how heat transfer works. I will probably eventually dive into that area soon enough, there is just so much to learn. You would think that the cromotor builders would have motors built with proper water cooling or air cooling. Like they would take the time to experiment and create a better motor.

 

[John in CR]

To go really hard under heavy loads, especially accelerations from slow to up to half of top speed on the flats, makes lots of heat because it requires real power at the wheel and efficiency isn't good.

From a dead stop every take-off starts at 0% efficiency. That never really sunk in for me until a ride with the first 10 miles was normal and a few hundred meters of net climb, and that got the motor up to a healthy but fully warm through and through. I just happened to check the non-ventilated motor for heat just before downtown, because it was one of the earliest rides with that rig, and the temp was fine. Then I got caught in downtown, narrow street, nowhere to go bumper to bumper start-stop traffic with no way to bail out of it. I just used the throttle easy and only accelerated to a few mph mostly. I was stuck in that traffic that made me hate using the car for 10-15 minutes, but going so slow at so little power I wasn't heat worries didn't cross my mind. I got to my destination and the motor was scorching hot to the touch. I didn't want to just park it, so I lifted the wheel and spun it at moderate throttle for a couple of minutes that helped enough that I went ahead and sprayed it with water which immediately evaporated but no sizzle. The shell would cool quickly with water, but quickly heat back up when the water was off, because the 12lb stator held so much heat, so I came back a few more times and watered it down. To make a long story short (which I always fail miserably at) the motor was fine and only the slightest change in motor wire color, but I came dangerously close to burning up a perfectly good motor at low speed and low power due to stop and go traffic.

For off-road riding the real answer is using the motor as a mid-drive and add a high power blower like Toolman2's leaf blower and force crazy air through. Hopefully we can get sensorless operation truly effective so they can build motors made for oil inside the shell with no leaking, and nothing to damage with piping to bring the heated oil out and pump to AL tubing welded to the frame. Currently motors designed for liquid cooling don't get the liquid in direct contact with the copper and the heat migration through the stator steel to the metal containing the coolant is too slow. That's why the Zero emoto that won the Pikes Peak climb last year had a high power blower instead of liquid. An oil fill of our hubbies is the first part of really effective cooling, but there's stuff inside not designed for it...and it leaks some too...and they way people do it doesn't increase the surface area, which is really all a radiator does. The surface area of smooth covered hubbies could easily be 4X or more surface area, and that means 4X the heat dissipation.

Yeah, moving heat is easy in some respects but reducing it to the numbers can be complex. I was lucky and my hobby before ebikes was designing hot air engines, ECE's instead of ICE's...E for external, and I tried my hand at low temp solar powered engines too, but getting meaningful power that way using the sun is really hard. Did you ever see one of those engines that could run just from the heat of your hand?....Fascinating and elegantly simple, but designing and building your own is HARD. I did get one running off of the sun though, my daughter and I did for a science fair project, but sure enough it ended up overcast that day so she had to use an electric heat source to get it running.

 

[Cowardlyduck]

Compare it to how much air mr chickenduck can possibly push through the motor even with the most optimum estimates. Those little fans are designed to help get rid of maybe 100W of heat with lots of help of good conduction and good heat sinks and close proximate. Of course they help cool a motor that is stopped faster. Raise your hand if you needed a graph to understand that?

Seriously John....re-read my previous post and think again about how you respond. I wasn't criticising you in any way, and in fact supporting your attempts at approaching this problem. You can mock all you like, fact is, the fan approach works, and works well. Sure so does your approach, but so what, until you show others with a test or similar they are less likely to follow suit.
What your missing is any kind of evidence to support your claims. No matter how logical your claims may be, you still need to prove them. That's just the way things work.
You can continue wasting your time writing essay's justifying your approach (and others time reading it), or achieve the same intended outcome (assuming you care about us all believing you) by using the same time taken in just demonstrating your approach through some documented tests. Doesn't have to be a graph if you don't like them, but I would suggest one as a good start.

Cheers

 

[Offroader]

John, I know what your saying that the small fans only push away so much heat, say 100watts. But those ratings are probably based on heat sinks which are in open air already, and the air surrounding the heat sinks is already cool. Think like in an RC car or a computer case, which has relatively cool airflow inside.

You see, the fans in our motors are replacing that hot air with cooler air, and also help blowing air through the windings helping to remove heat. So in a way they are probably removing a lot more heat than simply testing what a fan can do on a computer heatsink. I'm sure even the small fans are always going to have the airflow being a lot cooler than it is in a sealed motor.
So the stator and windings are able to cool off in a nice 40c temp rather than a 100c temp, than of course you have the extra air flow blowing through the windings.

Of course you may be able to get the same effect with proper cooling fins, but a combination of fan and cooling fins are obviously going to be best. I'm thinking the air duct method may work very well. The fins could probably move a lot of airflow also, I wonder how they would compare to just fans. The thing is the fans are very useful when stopping and letting the bike cool down. I have to take about 3 ten minute stops to constantly let my motor cool off while offroading, and it only cools to about 105c to 80c, and quickly heats back up again.

I just can't believe that our hub motors are sealed up the way they are, every electric motor has tremendous air flow, even electric drills have a high amount of air flow. I can still drive relatively hard with my sealed up cromotor, and all that heat and air is tightly contained. I can just imagine a motor that is open and with fans making sure the air inside the motor is always cool fresh air.

 

[Offroader]

Found these super fast fans when searching for something unrelated.

They offer 25mm and 30mm fans, I'm thinking cromotor can use 30mm's. These look to be pretty powerful, but of course not as powerful as those blowers. I think 1 high speed blower when you really need the air flow and a few of these would probably do a lot.

http://www.wildturbofan.com/#!product-new/ce6q

[youtube]RZNN3SrGxl8[/youtube]

 

[madin88]

Nice find on this Fans, but i believe those small RC impellers will move more air and have better efficiency.

new sketch with radial blower wheel (green part) inside the motor bolted to the sidecover:

With a bit of luck we should be able to find a buyable blower wheel which fits good.

the red lines show how i imagine the wheel must be modificated to make it fit. I hope you can follow

 

[Offroader]

I'm having a hard time seeing how the radial blower will help in that location.

Wouldn't it make more sense to put it on the other side, at least when you don't have the fans running it will be sucking cool air through?

 

[John in CR]

For interior blades my thought has been to push air away from the side covers at the stator. I doubt they move much air, but instead help ensure the air inside is spinning at near wheel rpm, make it difficult to take the easy route and flow along the smooth cover, and last but not least pass within a millimeter or 2 of the stator and end windings to create more velocity and turbulence there where it will do the most good. Nice and simple to make and bend, then a little epoxy and a nut and bolt to ensure they can't move or break loose and destroy the motor by damaging the copper. Here are some I've done.

View attachment 2
View attachment 1



I do higher rpms using small wheels and street riding, so I don't need fans, but adding fans to between the stator and axle to get more flow in at low speeds and when stopped with exhaust holes at the perimeter along with blades like these should work well without getting too complicated. You guys are looking at fans much more powerful than Zappy uses and he pumps over 10kw into a MagPie motor. Note that none of mine in the pics have the much more numerous small intake holes or slots cut with a dremel cutoff discs that spread out much further from the center, which is the approach I use now that I copied from Zappy.

Sorry, still no graphs for the duckman, but I could do one for a 5km climb with about 500m vertical where I pumped 700wh into the motor with the covers in the bottom pic (inside blades on both, intake on wire side, exhaust on wire side, no outside blades). The graph of motor temp would be 2 points, starting at ambient of 20°C going to 37°C at the top when we stopped. Due to the altitude and wind coming from over the higher nearby ridge ambient at the top of the mountain was a degree or 2 cooler, which helped I'm sure, but this was riding so hard that I got 5% regen on the way up braking for the many curves.

 

[Allex]

Madin starting to get serious. Who said that the EV supposed to be simple?

 

[madin88]

I'm having a hard time seeing how the radial blower will help in that location.

Wouldn't it make more sense to put it on the other side, at least when you don't have the fans running it will be sucking cool air through?

Sure you have right, but on this side there are phase and hall sensor wires passing the stator and i guess it will not be easy to ensure the blower wheel does not chafe against them.
It still will suck air through the motor, but not much through the windings or stator theeth. The biggest part will flow through the cooling fins on the stator and over end turns on exhaust side. Best would be radial blower wheels on both sides

 

[John Bozi]

If I understand your graph correctly, it takes a minute to lose 5 degrees C from 50 to 40.

Could you do it from 90-80 drop off as I have already done two tests at that temperature with vented holes.

I had a drop of 10 degrees in 1 minute with no power or movement,

or 10 degrees in 3 minutes with low amps and slow pedal cadence.

ambient temperature would have been around 20 degrees.

thanks looks very interesting but rather complex with a lot more things to mount to the bike and motor, I wonder if it is worth just rewinding your motor if the fans will only be efficient for low speeds.

 

[Cowardlyduck]

I plan to do more tests. Next test will be a regular ride on some single track, with/without the fans to show the difference.

thanks looks very interesting but rather complex with a lot more things to mount to the bike and motor, I wonder if it is worth just rewinding your motor if the fans will only be efficient for low speeds.

It's really not that complex. All I did was epoxy 6 fans over the pre-existing holes, run some additional wires, and hook it up to a DC-DC converter. You could skip the DC-DC converter if you ran enough fans in series directly off your battery. For example, if I was using 75v as I used to on this motor, I could run the 6 fans in series directly off the battery for 12.5v each...perfect.

Cheers

 

[John Bozi]

water if there is a splash of water in the hub and or just one fan failed does the whole system stop if its all in series?

did you use epoxy paste? suppose to be hard as steel the one I bought is high temp but it feels likes plastic... and I can scratch it off pvc which it should bind to. Do all those fans and or additional required parts add up to another kg?

Adding more weight to the unsprung weight...

I am interested or I wouldnt have followed this up just ask a ton of annoying questions before entertaining going this route... I'm currently entertaining buying a burnt out 40 motor and rewinding for the same reasons you do fans. I reckon rewinding would probably suck more than your fans by a bit...

 

[Cowardlyduck]

water if there is a splash of water in the hub and or just one fan failed does the whole system stop if its all in series?

Mine are not wired in series, but yes if one connection failed in a series connected scenario, they would all stop.
Water seems to not be an issue, see this post: http://endless-sphere.com/forums/viewtopic.php?f=30&t=56965&start=25#p944665
It probably helps that I periodically spray WD40 into the spinning fans also.

did you use epoxy paste? suppose to be hard as steel the one I bought is high temp but it feels likes plastic... and I can scratch it off pvc which it should bind to.

Yes, just standard epoxy was used. No problems with it so far.

Do all those fans and or additional required parts add up to another kg? Adding more weight to the unsprung weight...

No way it added that much weight. Each of the fans weigh no more than 50 grams + the wires, solder and epoxy all up might be another 100g max. So an addition of maybe 400 grams at most...probably less.

I am interested or I wouldnt have followed this up just ask a ton of annoying questions before entertaining going this route... I'm currently entertaining buying a burnt out 40 motor and rewinding for the same reasons you do fans. I reckon rewinding would probably suck more than your fans by a bit...

Rewinding would be lots more work that this was. Rewinding also doesn't guarantee increased power capacity or a cooler running motor. I've seen a few rewinding attempts end in approximately the same copper fill as a standard wound motor meaning the same current carrying capacity, and same cooling/heating capacity.

The great thing about this mod is it doesn't lock you in like other mods. Water cooling requires tones of difficult to reverse physical mods, and Oil cooling leaves messy oil everywhere. If you do this mod and it doesn't work out for you, you've not lost much, and it's pretty easy to rip the fans back out if needed.

Cheers

 

[Cowardlyduck]

Well it's only fair that I also post about the down sides of this mod.

Yesterday I was attempting to do that real world testing I've been promising, and my motor died.
I had noticed a decent amount of dirt/ sludge build up on the windings. I would say this is primarily from the fans sucking in dirty air as it's worse on the intake side.
I attempted to clean the windings by jamming a soft wet wipe into one of my vent holes up against the windings, then rotating the motor so as to forcefully wipe off the dirt.
It seemed to work, and the wipe came out with a bunch of dirt on it from both sides of the motor. What I didn't realize was that in doing so, it had snagged and ripped a bunch of copper strands off one of my phases, leaving only a few hanging on.
So I went for a ride (to do my testing as above). The motor worked fine at low power...but after stopping on a hill, then dumping full power (4000W) into the motor to get going up the hill, the remaining few strands on the one phase couldn't take it and blew apart taking some other windings with it. The phase wires then rested on the windings creating a weak short...needless to say, the ride home (5+km) was painful!

Anyway, the pictures below show the dirt build up, and damage.

I think I may be able to salvage this motor...again! Today I'm going to clean it up, and attempt to drop some solder onto the phase wire stubs, then reconnect that phase. I'll have to put some insulation (probably silicone rubber wrap) up against the nearby windings so it doesn't short. Ill also have to reconnect the damaged windings that are separated now. If it works, hopefully I'll be back in business soon. This motor is turning into a bit of a Frankenstein motor...but who cares as long as it still works. I probably shouldn't put as much power through it if I get it working again though. I would say no more than 2500w (50A) should ensure it doesn't die again.

Cheers

 

[Cowardlyduck]

Well that didn't work out as I was hoping.

Turns out the phase was somehow internally shorted to the stator. I thought the phase might have ripped off in between the laminations, so I managed to pull it through to the other side, but the short is still there. I can't see anyway of pinpointing it further at this stage.

I think this motor is now toast.

John Bozi if you want a decent motor for rewinding, this might be it. It's a HS4065 Stealth build, so upgraded axle, and quality 24" rim with 10g spokes. It's got the fans as shown still in place ready to go and a temp sensor mounted also. I might even chuck in my DC-DC converter/ multi display package also if your interested. PM me if you want it.

Anyway...on to the next target, my other motor also Stealth built but this time a HS4080. I like this motor better on 12S LiPo as it tops out at 65kph and pulls harder for longer. That does also mean it draws more amps though, so will heat up more...meaning it should benefit even more from fans, especially over the coming summer months.

This time I think I'll switch the intake - outtake sides, so in from the disc side, out on the freewheel side. I will also use better fans, probably the Yeah racing ones, and possibly add radial fins like Jonh in CR suggests to the inside cover of the outtake side to aid in getting air out.

Cheers

 

[John Bozi]

Is this

because of the water you added to clean it or were you riding through mud?

I can't work out if you killed the motor before, after or before and after it got caked in mud,,,

 

[madin88]

it looks like all the dust in there got wet. did it?
that pictures scare me to also build a fan cooled motor.

btw: why you want to swap intake and outtake side on the next motor? i guess if you place the intake on brake side, the fans will suck in lots of dust which was rubbed off the pads.

 

[Cowardlyduck]

Is this
because of the water you added to clean it or were you riding through mud?
I can't work out if you killed the motor before, after or before and after it got caked in mud,,,

I did ride through mud...plenty of times, but that's not what made the motor look like that. As mentioned I periodically sprayed WD40 into the motor...I only just did it a few day's ago. The WD40 combined with the dirt/dust build up just makes it look like mud. The motor died previously...see here for details, but basically only 2 strands of one phase shorted, so removed them and did the fan mod at the same time. It worked mostly fine for the last 6 months or so.
This time it's totally toast. I think the bending of the previously shorted phase connection when I re-soldered it weakened the strands enough so that it only took a wet wipe to snap most of them off.

it looks like all the dust in there got wet. did it?
that pictures scare me to also build a fan cooled motor.

As above, WD40 + dirt build up. My thinking was that the WD40 would at least aid thermal conduction through the layer of dirt for the fans to be more effective.

btw: why you want to swap intake and outtake side on the next motor? i guess if you place the intake on brake side, the fans will suck in lots of dust which was rubbed off the pads.

There are 2 main reasons for this:
One, I've found the exhaust onto the disc brake makes it pretty weak. My rear brake is already pretty weak...adding heat to it doesn't help.

Two, inside the motor it's only really possible to make radial fins on the side without wires, etc. If I mount all the fans to the wire side as I did with this motor that leaves the freewheel side free for radial fins.
As well as that, I'm not sure, but some of the 'dirt' inside the motor above could have been oil/grease from my chain/freewheel since it was sucking in on that side.

I'm not too concerned about brake dust...I mean look at the motor above! The build up was a little bad, but I don't think it effected the performance in any way. My bearings are still buttery smooth, and what you saw in the picture was the worst of it that took 6 months of almost daily off-road use to build up to that level.

Cheers

 

[madin88]

what the hell has prompt you to spray WD 40 into the motor? thermal conduction through the layer of dirt must be a joke lool
No wonder all the dust will turn into mud grease inside the motor. no way to get this out anymore. dust alone maybe can be blown out with an air compressor.

 

[John Bozi]

I understand why you spray wd40 = lubricates and decreases rust....

but you have added oil which permeates into every bit of surface area, an area which is exposed to even higher levels of super fine ground down dust/dirt.

basically caking every part of the motor with more insulation instead of air ..... so what levels of improved efficiency the fans may have produced were being cancelled out by extra layering, on one side of the motor anyway...

[youtube]CMThz7eQ6K0[/youtube]

And in reply to an early comment, I agree and I doubt I could wind the motor better....

I've been thinking of carrying one these in my backpack for a trial for times I don't feel like waiting.
http://www.officeworks.com.au/shop/officeworks/insystem-pressurised-air-duster-le50015
$10 is not cheap though... considering overheating say 5 times up mountain paths.

 

[Cowardlyduck]

what the hell has prompt you to spray WD 40 into the motor? thermal conduction through the layer of dirt must be a joke lool
No wonder all the dust will turn into mud grease inside the motor. no way to get this out anymore. dust alone maybe can be blown out with an air compressor.

Yeah, your probably right, but I'm the kinda guy that has to try things first. :lol:
Anyway, lesson learned...I won't be spraying it into the next motor. At least not regularly...maybe once every 6 months to prevent rust.

I understand why you spray wd40 = lubricates and decreases rust....

but you have added oil which permeates into every bit of surface area, an area which is exposed to even higher levels of super fine ground down dust/dirt.

basically caking every part of the motor with more insulation instead of air ..... so what levels of improved efficiency the fans may have produced were being cancelled out by extra layering, on one side of the motor anyway...

Yes, so as above, I probably won't do that again...but never would have thought it would be quite so bad had I not actually tried it. Live and learn.

I've been thinking of carrying one these in my backpack for a trial for times I don't feel like waiting.
<span><a href="http://www.officeworks.com.au/shop/officeworks/insystem-pressurised-air-duster-le50015" class="smarterwiki-linkify">http://www.officeworks.com.au/shop/officeworks/insystem-pressurised-air-duster-le50015</a></span>
$10 is not cheap though... considering overheating say 5 times up mountain paths.

Actually a much more compact, cheap and effective solution is those little C02 canisters used for inflating flat tires. I've used them numerous times in the past to cool an overheating motor before I ran fans.

Based on my observations, one cartridge can take 15-20C off the motor instantly. That's from about 100C. I think dumping one of these into the motor windings periodically will also help keep it clean.

Cheers

 

[Offroader]

cowardlyduck, was the fan noise very noticeable while riding or just noticeable if stopped? Was the fan noise annoying in anyway?

I am trying to determine what fan configuration to use in my cromotor.

I can fit up to 30mm fans in my cromotor, and have room for 6 of them.

I will use one slot for a high speed blower, but I want to fill some of the other spots with fans to have some options when driving.

I may use some quiet 30mm fans for when I want silent operation but want some airflow going.

What fan configurations would you recommend?

Thanks