9c Hub Motor Temperature Limits

[Erogo]

There's no way you can get enough volume of air in through the size of tube you could drill down an axle, to do any sort of cooling worth bothering with.

But, as has been discussed in other threads on this topic, squirting water through that tube, and allowing it to exit as steam, would carry a lot of heat away.

Your objections:
a) water inside my motor is a bad idea for the electricities, and
b) water inside my motor is a bad idea for the limescale build up

are
a) invalid in a brushless motor, which would run ok completely underwater, and
b) valid, but solved by using distilled water.

E

 

[sn0wchyld]

There's no way you can get enough volume of air in through the size of tube you could drill down an axle, to do any sort of cooling worth bothering with.

But, as has been discussed in other threads on this topic, squirting water through that tube, and allowing it to exit as steam, would carry a lot of heat away.

Your objections:
a) water inside my motor is a bad idea for the electricities, and
b) water inside my motor is a bad idea for the limescale build up

are
a) invalid in a brushless motor, which would run ok completely underwater, and
b) valid, but solved by using distilled water.

E

How often does the air in a motor need to be replaced to make a notable difference? I would have thought that any forced air would be greater than passive cooling via holes allone, where people have reported worthwile results?

 

[Erogo]

Yes but those are big holes. The hole you can drill through an axle is a small hole. Regardless of the size of your hole water will remove more heat than air will. Allowing the water to boil is key too. It also ensures positive pressure out of any vent holes.

 

[itchynackers]

I believe the only way to know for sure is to empirically test it on an actual motor being used in the field, not on a bench. If you have baseline temps without a pump, and the temps go down with a pump, then it works. We can guesstimate all day long. It may just be a case of diminished return on investment though.

 

[Erogo]

Sure. Go for it!

 

[itchynackers]

I did some temp testing tonight.
25.5C ambient.
Current limited to 15A with cycle analyst (100V)
12% slope road from 1400 feet (documented)
No pedal speed = 12.8mph
Mild pedaling = 14mph
Max temp observed 86.4C

The max temp seems to occur after I stop. The temp climbs for about a minute, then drops. Not sure if this is due to my sensor location, or if the windings are actually getting hotter.

 

[dozentrio]

Temperature does not have "inertial" effects (doesn't keep going after you stop pushing it). It is sensor lag

 

[itchynackers]

How can I "improve" this? If I coated the entire sensor in silicone sealant, and it was sealed right to the windings, I may get more accurate readings. I tried finding a sensor that was smaller, and could fit inside the windings. That didn't work out, so I have a metal cased sensor that is 3/4" long and is now mounted about a millimeter from the windings. I'm hesitant to let it touch the windings, since it is metal.

Edit: Also the sensor is exposed to the outside air, since I have cooling holes that pass right over it (and the windings).

 

[dozentrio]

Well, the metal casing on your sensor will delay heat transfer a little. Also the gap between it and the windings. While it might be metal, the coating on the wire will insulate it from shorting out against the sensor. If the coating melts/dies your motor is done anyway. Perhaps use some of that high temp string they like to use in motors and tie it firmly to the winding. With a dab of epoxy to keep it from rattling around. Don't know.

 

[evo1087]

I find I have little to zero problems if I keep a motor's temperature under 200F, measured inside, 150f measured outside. Those are the limits I shoot for when reliability is needed.

Hey where do you have your probe mounted externally? Would a non contact IR type be a good idea attached to the swing arm pointed straight at the windings with a few mm gap?

 

[John in CR]

Hey where do you have your probe mounted externally? Would a non contact IR type be a good idea attached to the swing arm pointed straight at the windings with a few mm gap?

If you're limited to measuring temp from the outside using a sensor, the place to do it would be the motor axle. Though somewhat slow due to the long pathway, it does have a direct heat path through metal from the copper windings all the way to the axle ends.

 

[evo1087]

If you're limited to measuring temp from the outside using a sensor, the place to do it would be the motor axle. Though somewhat slow due to the long pathway, it does have a direct heat path through metal from the copper windings all the way to the axle ends.

I never thought of that next time I ride I'll check that out. I've been measuring temperature at the phase wires near the controller and I stay bellow 50 Celsius. You think I can push it up to 60 Celsius if my hub motor is still under 65cel/150F?

 

[John in CR]

If you're limited to measuring temp from the outside using a sensor, the place to do it would be the motor axle. Though somewhat slow due to the long pathway, it does have a direct heat path through metal from the copper windings all the way to the axle ends.

I never thought of that next time I ride I'll check that out. I've been measuring temperature at the phase wires near the controller and I stay bellow 50 Celsius. You think I can push it up to 60 Celsius if my hub motor is still under 65cel/150F?

I run significantly larger phase wires shortly after coming out of the axle, so temp sensing would be less predictable because the copper will wick away heat from the motor so quickly. One good thing about measuring there is that heat from the windings will get there more quickly than to the axle. Before I used temp sensors, I used to get a good feel for motor temps using my hand on the side cover near the perimeter.