BLDC Motor shaft heating up

jskounakis

1 mW
Joined
Jan 25, 2022
Messages
17
Hello,

I just got a ME1803 electric motor from kit-elec-shop, and unfortunately, noticed some wear and tear (like rust on the shaft). While that on its own didn't bother me much, when I run some tests the shaft got relatively hot starting from the bottom near the motor body and gradually increasing, especially at high RPM.
**I also noticed there is a seal at the shaft base, could it be the source of the issue? Some sort of friction?

That got me really worried since the shaft is supposed to stay cool (doesn't have any wiring, thus only heat should come from the windings via the air gaps. The sound at the end of the video linked was also alarming.

Here is some technical info regarding setup and testing, here is some visual help.

Running FOC with a VESC 75/300 controller hooked to a 52V Li-Ion battery.
The test was no load, didn't have water cooling hooked up (but again, the internal thermistor showed 18C and the body was cool)
The problem starts at around 30-50% duty cycle (should be around 2-3K RPM)
Measured battery current at around 5-7A, phase current at 15-20A max.

Has anyone faced similar issues/or has any idea how to handle them?
 
Does it spin freely by hand?
A seal should have only minimal drag. Should not heat up the shaft.
If it has tapered bearings they will need some running time to break in (and will drop bearing drag maybe 20%) but still they shouldn't have enough drag to heat it up that quickly.
 
Jrbe said:
Does it spin freely by hand?
A seal should have only minimal drag. Should not heat up the shaft.
If it has tapered bearings they will need some running time to break in (and will drop bearing drag maybe 20%) but still they shouldn't have enough drag to heat it up that quickly.

Well it depends on how you define freely spinning. It does need some force to get it started, but after that it does spin, and you can feel the magnets. But overall it does oppose your movement.
 
Shamelessly bumping my own thread, hoping it's allowed.

I removed the seal and it's getting less hot, BUT, after 15mins running, the shaft still gets pretty hot to touch, around 50+C and motor body is 35C. Still sounds strange, and I've only been feeding it around 200W.

Please please any answer would be very welcome, I'm pretty sure a BLDC motor is not supposed to be hotter on the shaft than on the body.
 
50c isn't that hot. Have you added water yet?

I meant to check that the bearings don't feel crunchy or have tight spots.

It doesn't sound bad to me. Sounds like it braked / regened at the end.
 
Jrbe said:
50c isn't that hot. Have you added water yet?

I meant to check that the bearings don't feel crunchy or have tight spots.

It doesn't sound bad to me. Sounds like it braked / regened at the end.

It just caught me off guard since it's on the shaft, and cooling won't do much since it's focused on the windings. Bearings got changed and it spins much more easily by hand.

I added water but only with a big bucket and pump, water ended up hitting 33C which might also be an issue, ill focus on cooling more.
 
If the internal temperature sensor (in the windings?) indicates only 18c, but the shaft is 50C and the casing is 30C, it means the shaft itself, or rather it's bearings or seals or anything else rubbing on the shaft, is generating most of the heat, and transferring it to the casing via the bearing/case interface.

If the water is heating to 33C, but does not have contact with the shaft, yet the casing (before water cooling) did not reach even 30C, it may indicate the internal sensor is reading much lower than reality; the water couldn't get hotter than the casing does if there was not a source of greater heat inside the motor (other than the shaft), where the water courses thru, and that is typically only the windings and/or stator laminations.


The casing can be cooler because it has a large surface area vs the shaft, and being aluminum (vs the shaft's steel) also has a higher thermal conductivity. But the shaft is on the rotor (which if it's a brushless motor has no direct connection to the active part of the motor that generates heat.



Something to note with FOC is that typically a controller needs to be tuned to the specific motor it is operating for it to work correctly and efficiently. If that tuning hasn't been done yet, it would generate more heat, but this should happen inside the windings, not the shaft, so it's unlikely to be the source of heat, assuming that the internal temperature sensor reading is correct.
 
amberwolf said:
If the internal temperature sensor (in the windings?) indicates only 18c, but the shaft is 50C and the casing is 30C, it means the shaft itself, or rather it's bearings or seals or anything else rubbing on the shaft, is generating most of the heat, and transferring it to the casing via the bearing/case interface.

If the water is heating to 33C, but does not have contact with the shaft, yet the casing (before water cooling) did not reach even 30C, it may indicate the internal sensor is reading much lower than reality; the water couldn't get hotter than the casing does if there was not a source of greater heat inside the motor (other than the shaft), where the water courses thru, and that is typically only the windings and/or stator laminations.


The casing can be cooler because it has a large surface area vs the shaft, and being aluminum (vs the shaft's steel) also has a higher thermal conductivity. But the shaft is on the rotor (which if it's a brushless motor has no direct connection to the active part of the motor that generates heat.



Something to note with FOC is that typically a controller needs to be tuned to the specific motor it is operating for it to work correctly and efficiently. If that tuning hasn't been done yet, it would generate more heat, but this should happen inside the windings, not the shaft, so it's unlikely to be the source of heat, assuming that the internal temperature sensor reading is correct.
Thanks for your input!
To clarify, the last test I run was on a hot day, internal thermistor reads 35, water is around 33, casing was the same. Other tests I previously run were on colder days, but still thermistor-water-casing are approximately same temps. Shaft on the other hand gets hot to touch.

Just to have a point of reference since I am worried about demagnetization on the long run, from your experience, at what temperatures are BLDC motor shafts usually running?
 
Take the temperatures after the motor has sat for a few hours. I'm guessing you'll find the internal temp sensor is off by 10° or more. Figure out why it's off if it is.
You should be able to set a temp limit to keep things safe. You can add a temperature indicating sticker on the shaft if there's room to keep an eye on it.

There's no way seal drag is making that much heat, it would have melted.

Bearings might be tight to allow for break in. Did you ask the company you bought the motor from if your experience is abnormal for this motor?

Set some safety parameters, get the water cooling setup, add a temp sticker / paint to the shaft, check the temp every few minutes, and gradually push it harder as you check it and are more comfortable with it.
 
jskounakis said:
Hello,

I just got a ME1803 electric motor from kit-elec-shop, and unfortunately, noticed some wear and tear (like rust on the shaft). While that on its own didn't bother me much, when I run some tests the shaft got relatively hot starting from the bottom near the motor body and gradually increasing, especially at high RPM.
**I also noticed there is a seal at the shaft base, could it be the source of the issue? Some sort of friction?

That got me really worried since the shaft is supposed to stay cool (doesn't have any wiring, thus only heat should come from the windings via the air gaps. The sound at the end of the video linked was also alarming.

Here is some technical info regarding setup and testing, here is some visual help.

Running FOC with a VESC 75/300 controller hooked to a 52V Li-Ion battery.
The test was no load, didn't have water cooling hooked up (but again, the internal thermistor showed 18C and the body was cool)
The problem starts at around 30-50% duty cycle (should be around 2-3K RPM)
Measured battery current at around 5-7A, phase current at 15-20A max.

Has anyone faced similar issues/or has any idea how to handle them?

Are you sure about the measurements? Any type of losses would show on the noload power needed and i don’t think the current seems high. What’s the kV of the motor? You might ask kit-elec shop about the noload power values, if they seem normal.

Normally when running at noload the current goes (slightly) down for the first minutes or so because the bearing grease heats up and the viscosity decreases. If your current stays the same or even increases then i think there might be an issue with motor bearings.

They don’t sound too good in the video but it’s always hard to tell from a video depending on the microphone. Bearings might also still be fine but with too high preload from poor tolerances, misaligned installation etc - i think they are the issue, not the seal.
 
larsb said:
jskounakis said:
Hello,

I just got a ME1803 electric motor from kit-elec-shop, and unfortunately, noticed some wear and tear (like rust on the shaft). While that on its own didn't bother me much, when I run some tests the shaft got relatively hot starting from the bottom near the motor body and gradually increasing, especially at high RPM.
**I also noticed there is a seal at the shaft base, could it be the source of the issue? Some sort of friction?

That got me really worried since the shaft is supposed to stay cool (doesn't have any wiring, thus only heat should come from the windings via the air gaps. The sound at the end of the video linked was also alarming.

Here is some technical info regarding setup and testing, here is some visual help.

Running FOC with a VESC 75/300 controller hooked to a 52V Li-Ion battery.
The test was no load, didn't have water cooling hooked up (but again, the internal thermistor showed 18C and the body was cool)
The problem starts at around 30-50% duty cycle (should be around 2-3K RPM)
Measured battery current at around 5-7A, phase current at 15-20A max.

Has anyone faced similar issues/or has any idea how to handle them?

Are you sure about the measurements? Any type of losses would show on the noload power needed and i don’t think the current seems high. What’s the kV of the motor? You might ask kit-elec shop about the noload power values, if they seem normal.

Normally when running at noload the current goes (slightly) down for the first minutes or so because the bearing grease heats up and the viscosity decreases. If your current stays the same or even increases then i think there might be an issue with motor bearings.

They don’t sound too good in the video but it’s always hard to tell from a video depending on the microphone. Bearings might also still be fine but with too high preload from poor tolerances, misaligned installation etc - i think they are the issue, not the seal.

Someone on our team changed the bearings and I removed the seal, temperature increase is much much slower, but it still seems that it's going hot, I'll run some more tests just to be sure.

Unfortunately kit-elec-shop has given no replies about the state of the motor, why is it rusty, etc, it's a bit fishy.

Thanks for the input guys, I'll check the current and if it's getting higher, and try to find a temperature sticker as Jrbe suggested. Maybe even try to open it up after the test to check where the temperature increase originates from.
 
jskounakis said:
Thanks for your input!
To clarify, the last test I run was on a hot day, internal thermistor reads 35, water is around 33, casing was the same. Other tests I previously run were on colder days, but still thermistor-water-casing are approximately same temps. Shaft on the other hand gets hot to touch.

Just to have a point of reference since I am worried about demagnetization on the long run, from your experience, at what temperatures are BLDC motor shafts usually running?
I use hubmotors myself, so it's not the same situation as yours (where the shaft is part of the rotor, rather than part of the stator like mine), but most of the time the axles of mine are near ambient, until I've been running high loads for a while, and then they can warm up a dozen or two degrees above that. (but the rest of the motor is at least twice *that* hot, just on the outside, and much hotter inside).


If the shaft is getting that hot, but the rest of the system is not, and you are sure the temeprature readings (internal and external) are accurate (for instance, measured with at least two known-good direct-contact (not "IR gun" non-contact type) devices to be sure you have consistent readings), then it really sounds like the bearings and/or seals, or something else that causes friction with the shaft, is heating the shaft itself up.

In this case, it will also heat the area of the casing around the source of this friction, so a "real" FLIR unit will show you the source of the heat (it will be the hottest thing there is during heat generation) if it is visible from the outside.
 
Back
Top