Cooling fans inside Hub motors

[Offroader]

OK I see.

I think you may be better off drilling only one side and allowing the air to be sucked around the magnets and windings. Yes a lot of air will be recirculated but some will leave and the air would constantly be cooling the coils. If you have holes drilled near the center and then some up near but just past the coils you would have constant air over the windings and magnet gap where it is hottest, and then some air would be pushed out and new air into the system.

I was looking at the water cooling setups people have done and what is interesting is that they only put some water blocks on the inside of the stator. So all the cooling is from pulling heat from the stator. I wonder how that would be compared to my active cooling with high air flow. A water cooling setup in this way still relies on the heat exchange from the windings to the stator.

My setup if I ever get it running relies on high air flow through the windings and blowing the hot air out of the motor and having fresh air entering. The 30mm EDF is so powerful that the airflow is very strong. I'm thinking it may be more effective then the water cooling method others have used which they claim allows them to run the motor as hard as they want. Like you mentioned, your fan setup doesn't slow down the temp rise, it only allows you to cool it down faster to allow for the next temp rise. My setup hopefully will stop the temp rise. I also don't see the water cooling method from the stator stopping the temp rise also. It takes too long to get that heat from the windings into the stator and into the water block.

 

[Cowardlyduck]

I'll have to try and get some better photo's at some stage as I'm not sure you fully get how my setup works Offroader.

The holes in my side covers are as close to the perimeter as possible. Looking directly into them, only a small part (maybe 1/8th) is not over the windings.
The air doesn't come in the intake holes and turn 90 degrees. It has to flow over the windings first, and I would bet some of it also makes it's way up into the magnet gap.

I do hear you on the intake near the axle and doing it all on one side, but as you mentioned, the majority of the air would recirculate in that case. I'm not convinced this would be any better than how I have it now.

As for your setup removing the most heat and prevent it from being created as much...I can totally see your point. That's why I was saying, your approach x6 EDF's would make a killer drag setup.
Doctor Bass...if your listening...do this to your NYX motor's...you will be able to pull 16KW+ per motor without insane heat.

Cheers

 

[newbiebiker]

In SUV type cars we have 12V portable refrigerator for keeping our beer cans cool, i think that consume very less power (no sure) like that we can try something, because cooling with refrigerant means, it can keep the constant temperature.

http://www.alibaba.com/product-detail/12v-dc-fridge-compressor_60138319054.html
http://www.alibaba.com/product-detail/Restaurant-diabetic-portable-refrigerator-AC-DC_391377614.html

 

[kiwiev]

The sooner someone makes a A.C. drive hub the better then all you have to cool is the controller

Cheers Kiwi

 

[Offroader]

Just an update, ran into some trouble with filling in the gap on top of the air duct. It seems that if there is any gap there most of the air will recirculate and it won't offer much cooling. You'll just recirculate the hot air inside the motor. It may offer a small amount of better cooling but nothing that will make any difference as far as I can tell.

I'm on my 3rd attempt now to fill in this gap and think I finally got it this time. I'm shooting for about .3mm spacing in this gap as it seems you need a little bit of space because there are tolerance issues in the motor or it will rip off the air duct.

I may also have to try and do something to redirect the exhaust air so that it doesn't get pulled into the intake holes as it is on the same side.

From what I can tell is that these motors get really hot once the heat has heated up all the aluminum and steel inside the motor. It may be harder to shed this heat then I originally expected. But I am saying this before I have filled in the air gap for proper hot air exhaust.

With a proper filled in gap the exhaust air is pretty strong, without it filled in or only about 2-3mm of space in the gap the air will recirculate. I guess the fan sucking will draw air easily from places where the air is being pushed from the fan, so any small opening will cause lots of air recirculation and the gaps really needs to be closed up.

It may take better air flow directing inside the motor also. I'm thinking water cooling may be the way to go and is easy to implement if a side cover and axle are custom made to fit the water pipes as some sell kits for certain motors already.

 

[Cowardlyduck]

It sucks your still having issues Offroader. Hope you manage to get them all sorted and end up with something functional.

My issues haven't been related to my cooling fan setup, but have still prevented me from testing it properly.

Today I re-wired my old 18Fet Infineon back onto my bike so I can continue to use it while my Adaptto is back in Russia for warranty repairs. The problem is, in order for my fans to run at the right voltage, I need to have enough of them in series...removing my controller with 3 fans attached has bypassed 1/3 of my voltage absorption. It also means I have no display since my cycle analyst is on my commuter recumbent presently. So I have no means of knowing power usage, or battery voltage, although I've programed some limits into the controller to keep it safe.

For the remaining 6 fans in the motor it's only a problem when the voltage exceeds about 52V (which happens to also be the nominal voltage of my LiFeP04 pack) or 8.6V per fan since they are rated up to 7.4V each and I'm fairly confident they can handle an extra volt. My long wiring, button, and multiple connectors would also be absorbing some voltage I imagine. My old stock Stealth Fighter pack is 16S LiFeP04 so 58.4V fully charged, but it's so old now the voltage drops off very quickly.
I went for a quick ride without the fans running and got the motor up to about 60C. Once stopped I tried (very briefly since I didn't know the battery voltage) running the fans and the rate at which the temp dropped seemed to almost triple! It looked like it went from dropping 0.1-0.2C every second to dropping 0.4-0.6C every second!
I'll do more definitive testing once I've got everything up and running properly with the Adaptto, but initial results seem promising, and better than my last attempt.

Cheers

 

[kiwiev]

Awesome work Blake

PM me I can lend you a V2?

what amps are you pulling?

Cheers Kiwi

 

[Cowardlyduck]

PM me I can lend you a V2?

what amps are you pulling?

Don't worry about the CA...I'm only going to be using this setup till my Adaptto is fixed...a couple of months max I imagine.

I've limited the controller to 50A for now...I might drop it even further if I find the BMS shutting down in the old Stealth pack, or my range is too low with it.

Something else I've done that's a bit unrelated, but still worth mentioning is I've setup my main battery plugs (XT150's) external to the frame. This is so I can switch over to a LiPo backpack battery consisting of 12S3P Zippy Compact 3700mah cells, so about 400Wh usable. I'll have to run buzzers on it till I get my Adaptto with display back so I don't over discharge it.
The good thing about the 12S LiPo is it's max voltage is 50.4V, so I'll be able to run my fans any time when hooked up to that battery.

Cheers

 

[litespeed]

Well I'm on board for trying this! I went for a hard run today, some 26 miles worth and rode it like a spanked ape for the last couple miles. When I got home and parked it I could feel the temp going up. Went up some 10 degrees in like 5 minutes. I grabbed a desk fan and turned it on pointing at a forward angle and in like ten minutes it dropped like 20 degrees......that was when I had one of those "Duh!" Moments. I said to my self, "self, if this cools good with air blowing on the out side imagine how well it would do if cool air was blowing on the inside???"

I'll report back once I figure out what size fans I need, where to get them and have it all ready to go.

Thanks for the awesome idea CowardlyDuck.

Tom

 

[Cowardlyduck]

Thanks for the awesome idea CowardlyDuck.

No worries Tom. I'm glad more and more people are going to give this a go as it will only improve everyone's approach in the end.

Cheers

 

[kiwiev]

Hey Tom is that e Vector Ekross?

what are you running? do you have a thread on it?

Cheers Kiwi

 

[litespeed]

Mine is the EVelbike bike frame and is the same as the Vector. Mine is mostly hand me down parts for builds past trying to decide what new parts I want to purchase.

In a nut shell and so I don't stray far from topic mine is running a Lyen 18 FET on a home built spot welded ( built a spot welder from some boards I purchased!) 24s 15p Koinion pack through a CA V2.3 pumping power into a MXUS 6T with upgraded wires and replace halls in a 17" Roadrash rim/tire combo. The whole setup is great minus a little bit lower top speed than I thought I wanted and it runs hotter then I wanted. I will say it sips power though.....I can average 20mph on a 26 mile trip and use just 9 amps of the available 22.5! That is really riding pretty aggressively on the street peaking at 38 mph many times but always peddling.

I'll do a build thread after I get this thing cooling properly.

Tom

 

[Cowardlyduck]

I'm surprised you need any cooling with a 6T motor!

Are you sure you have the right phase/halls combo? You know there are some combo's that will 'seem' to work, but they are still not the correct combo and chew more power/generate more heat than the correct combo.
If you are sure you have the right combo, then you must be hitting some serious hills or doing a LOT of stop/start to be over heating it.
And if that's the case, you definitely won't need to do much more than what I've done with the Yeah Racing fans to keep it cool.

Cheers

 

[litespeed]

That was what I was thinking! I for sure have the right wiring but it does heat up. Not 200 degrees or anything like that but first time it hit 150 ish it smoked the crappy stock halls and I replaced them with the ones from Digikey....the good stuff.
I'm a long time R/C guy and like a cool running set up. It's the way my brain works so the time I get to 130 degrees I'm looking for ways to cool it down. I'm sponsored by Scorpion (R/C helicopter.) and even though they are good to 200 plus degrees ask me how hot mine have ever gotten.....not even close is what I'll tell you.

Thanks again.

Tom

 

[Offroader]

I just keep having bad luck.

Finally got the air duct set and heated the motor up to 200F / 93C with an 8-10 mile ride in the hot summer heat.

I came back home to quickly hook up the power to the electronics because I didn't get to installing it on my bike. I started the fan at full speed and the exhaust air out of the exhaust holes all around the motor was pushing out air so hot it would burn my hand if I held it in front of the exhaust holes, and it was staying hot like this. Seemed to be working well.

I then went to get my father to show him how it was working and when I started the fan again the propeller came off the EDF fan. So now I have to open up the motor yet again to fix it. This will be the 5th time I installed the hubmotor and had to rip it apart.


Now the problem may come to the reliability of these cheap China EDF fans. Unfortunately I didn't have to tools to check the propeller tightness which was recommended I think in the manual. I'll have to figure this all out.

 

[Cowardlyduck]

That sucks man...or as we say here in Australia...Bugger!

TBH I'm surprised I haven't seen more fan failures myself...I mean I rode through a freaking flooded creak/river with them running and they still kept going.
Maybe EDF's are less tolerant to imbalance or heat?

Cheers

 

[Offroader]

That sucks man...or as we say here in Australia...Bugger!

TBH I'm surprised I haven't seen more fan failures myself...I mean I rode through a freaking flooded creak/river with them running and they still kept going.
Maybe EDF's are less tolerant to imbalance or heat?

Cheers

The thing is the EDF spins at like 70-80,000 RPM or something like that.

It seems that the propeller is only press fit onto the shaft, so a plastic pressfit onto metal, so no wonder it fell off. From my research it seems that people use some glue to keep them on, so i'll reattach it tomorrow with some glue.

 

[Offroader]

Cowardlyduck, I'm thinking about doing something like your chart. Why did you choose 53C as your temp? I normally have my motor pushed to 90C

What did you just write down every so many seconds the temp drop?

 

[Cowardlyduck]

Cowardlyduck, I'm thinking about doing something like your chart. Why did you choose 53C as your temp? I normally have my motor pushed to 90C

What did you just write down every so many seconds the temp drop?

Hey cool, please do, this is the only way to really show/prove the effectiveness of this approach. I need to do it again with my new setup at some stage also.

I didn't choose 53C as the temp...it just happened to be the temp I reached by the end of that particular hill. The key is repeatability for these kind of tests, so I choose that hill as I knew I could go up/down it numerous times without interruption and it wasn't so long as to make it take ages to do the testing.

I made the graph in excel based off the video below.
[youtube]P-VXS-aZZkU[/youtube]
I mounted my camera on the front of my seat (in my crouch :lol: ) and reviewed the footage afterwards pausing at 10 second intervals from when I started riding to record the temperature in each frame. Some temperature readings were pretty hard to see and some had to be guessed based on what I could see in the frames around the interval point when I couldn't see it properly, but I would say still pretty accurate.

Next time I will also try and go for a higher temp...maybe around 80-90C. When I originally did this graph, I expected I would be doing another better test soon afterwards, but never got around to it.
In fact, the time I went out to do my next test + video + graph was when that motor (HS4065) died from a ripped off phase wire (not related to the cooling fans).

Cheers

 

[kiwiev]

Good work Blake

How many amps were you pulling?

I assume that was your 50volt pack?

Cheers Kiwi

 

[Cowardlyduck]

That video was ages ago...don't even remember what voltage pack I was running at the time...could have been the 18S LiPo.

Amps would have been limited...I think to 40 or 50, but the draw going up that hill would not have reached that for long if at all.

Next time I will try for more amps, but none of my current batteries can really handle much.

Cheers

 

[Cowardlyduck]

I took the bike for it's first decent spin today. It was a quick ride around the local lake...about 12.5km total.
I didn't run the fans until I hit about half way and about 60C. The temp didn't seem to drop much, and then I started to hear some mild rattling from the motor.
My controller (old 18Fet infineon) then did it's hall sensor thing...(whole other story, but basically the controller drops a hall sensor randomly, for random periods then comes good)... that then sent the temp soaring as I pulled WOT repeatedly for a few seconds at a time to keep up my speed and check if controller would come good...and yes I know/understand the risk to the controller FET's of doing this.
The motor temp peaked at about 90C once or twice, but the fans seemed to keep it from rising any higher, and it mostly hovered around 60-70C. I thought that's pretty good considering the motor is essentially fighting itself going WOT with a dropped hall sensor.

Anyway, once home I tried to have a look inside the motor (while it was still on the bike) through one of the holes when I noticed I could see straight through (diagonally) to a side cover hole on the other side!
So I disassembled everything, took the wheel off the bike, and opened the motor to find this:

I think at least one of the fans had fallen off some time ago as the feel of the air flow exhaust was never what I expected. The other 2 fans might have come off during the ride, but I really don't know.

Anyway the wires are all still good, the fans all still work, and the epoxy is still firmly attached on 3 of the fans (the one's in the bottom of the shot above).
So I'm going to try again, this time with a few changes. I've scratched up the surface around the holes.

This should help the epoxy bond better than previously.
I've also got some different epoxy.
Previously I used this stuff:

This time I will use this stuff:

Apparently it's got the strength of a light weld!

I've also got this glue which can handle high temps and is flexible.

I might add a good about of this around the existing fans so I don't have to rip them off and remount them.

Anyone have any advice...I'll probably be doing this tonight so chances are I might not see the advice before I go ahead with what I thinks best anyway, but that doesn't mean I can't learn after the fact.

Cheers

 

[kiwiev]

Bugger Blake

I would drill 4 small holes around the fan holes this will really give the epoxy something to grip to.

I have some oilfield weld 2 part glue the glue pipes together if you want to try.


Cheers Conrad

 

[Cowardlyduck]

Thanks for the suggestion Conrad.
I didn't want to have to remove the other side cover from the motor as it's the wire side and usually involves painful extraction of all the wires through the axle slot and under the side cover. It often results in ripped insulation or broken wires if not careful, so I try and avoid it where possible and as a result didn't end up doing any drilling as it could have left metal shards inside the motor if I didn't remove the other side cover to ensure there was nothing in there.

I did however go nuts with the epoxy and glue:

New epoxy is light grey, old is dark grey, and the glue is yellow/clear.

End result is I'm pretty sure these fans aren't coming loose again!
The combination of a much stronger epoxy and strong semi flexible super glue over the top should definitely hold. I ended up using up that whole tube of glue since it started leaking everywhere and I thought I may as well put it to good use. :lol:

I only hope I never have to remove or work on parts that have been glued down as I will be a huge PIA to remove all that glue now.

I'm flat out for the next few weeks, but may be able to do some more testing in a month or so.

Cheers

 

[kiwiev]

We will have to meet up in August I think a ride on each other's local trail is in order we can really do some testing.

Btw had 4 inches of snow on the ground at home woke in this morning and gone rain and blowing a gale.

Cheers Conrad