What can be done about motor overheating?

[Lowell]

Stripping the black paint off might help with heat transfer.

Not really sure how effective these are, but I use them on pistons. Thermal barrier on the crown, dry film on the skirt.
http://www.techlinecoatings.com/introduction.html#thermaldispersant

 

[Nimbuzz]

I think you can just give it a coat of white or silver. Stripping the black can be a messy job.

 

[xyster]

C-Lyte 96F
TF 107F
Volvo Roof 153

Two hours later
C-Lyte 99
TF 124
Volvo 153

Your temperature results seem pretty significantly different to me, Al, especially considering the motors weren't also being heated from the inside. The case of the TF motor is a different material, right? It looks plastic. Two x5's painted black and white might demonstrate an even more dramatic difference.

 

[Nimbuzz]

it's an aluminum cover, standard hub motor design. Plastic doesn't heat up like metal and wouldn't be a relevant test.

Sitting in the direct noonday sun, aimed up at the sun for an hour = 11 degrees difference. From this 11 degree difference I don't see how a hub could get heated much from the sun when at an upright angle and the wind blowing by it big time. But go ahead and paint it white, it will be an interesting experiment.

I'll bet there's something wrong in the hub. But I've been wrong 50 million times. It would be bogus to paint the hubs black if it caused overheating. I would have called the dealer to discuss it right away to get his take on it as he sees the black hubs and the silver hubs in action all the time.

A

 

[CGameProgrammer]

Well I painted it white but it still got pretty hot by the end of an 8-mile ride today; weather was sunny and 87 degrees. It was a long shot anyway.

Could a faulty connection cause heat somehow? And if it is bearing-related, how do I check them?

 

[Lowell]

Start drilling cooling holes. I'll post a pic later tonight of what I did to my X503 recently. One thing to keep in mind as well, is that the X503 is not the same as an X5303.

 

[xyster]

... One thing to keep in mind as well, is that the X503 is not the same as an X5303.

I'd not heard that before. How are they different? The crystalyte site refers only to the "5000 series".
http://www.crystalyte.com/xlyteupgrade.htm
http://www.crystalyte.com

I understand that a couple years ago 5mm was shaved off the width of the x5. Is this what you're referring to?

 

[Nimbuzz]

Game, for the bearings you want to see how well the wheel spins -- you can test it on the bike. The front should spin very free and the rear will have cogging effect from the motor but should spin fairly free. When I hold the rear 5304 off the ground and my strong girl spins the wheel as hard as she can is spins for 3-4 seconds with the brake completely released. There is a buzz and this short time from the 'cogging effect' of the motor. If yours has more resistance than this I'd first check the dropouts to see if both sides are seated equally and that nothing is touching the wheel. If nothing shows up I'd take the wheel off and the cover off to see what's inside. Take pics and post them here. The bearing should be in the cover (I've taken many motors apart but not a 5304. Spin the bearing with your fingers and it should feel perfectly smooth. The other side someone else may have to tell you how to get to it. If the center hub pulls out (does a lock ring or anything hold it guys?) you can feel the other side bearing the same way. Inspect the coils and post pics.

If this doesn't work the only cure is to paint a perfect rendition of Van Gogh's 'Starry Night' on the right side hub and Dali's 'Last Supper' on the left -- this works every time!

We'll figure this puppy out!

 

[fechter]

The only kind of electrical problems that would cause more heating would seriously affect the performance at the same time. One test would be to lift the wheel and measure the no-load current at full throttle.
Amps = heat.

 

[CGameProgrammer]

I don't know how to measure current. I haven't seen any multimeter that handles more than a fraction of an amp.

 

[fechter]

The more expensive models can usually handle 10 amps. Otherwise you can use a shunt and measure the voltage across the shunt. AC/DC clamp-on probes are very expensive, but handy.

 

[CGameProgrammer]

Can someone tell me how to remove the motor's side covers? I removed the eight screws but can't figure out how to pry off the side plate; I tried jabbing a knife and flathead screwdriver around the edge but it all seems like it's a solid piece of metal.

EDIT: I then tried putting a screw in half-way, then grabbing the head with a pair of pliers and pulling hard. The plate did not budge.

What am I doing wrong? I sold a similar broken motor to fechter; fechter, have you been able to remove its side cover? How did you do it?

 

[Rassy]

I asked a similar question after trying to "pry" open a WE hub motor. See "Replacing Hub Motor Brushes"

http://endless-sphere.com/forums/viewtopic.php?t=999


The following site has pictures, uses a puller, and includes some cautions:

http://www.7gen.com/book/my-ev-projects/cleaning-evt-hub-motor/701

 

[Nimbuzz]

... One thing to keep in mind as well, is that the X503 is not the same as an X5303.

How are they different?

I understand that a couple years ago 5mm was shaved off the width of the x5. Is this what you're referring to?

I had an early 503 and it wouldn't dish all the way to center. The spokes kept loosening, Kenny told me the newer 06+ 5304 etc had the hub moved over 9mm on the axle. I exchanged the old for the new -- the new centers perfectly but I'm still going through some spoke tightening cycles but expect it to settle in eventually.

 

[xyster]

Thanks for the explan, Al.

 

[Ypedal]

Getting the motor cover off can be a chore....

First, if the cover was painted onto the ring.. you should try to take a sharp knife and cut the paint away all around the hub cover.

Then if you have a 3 jaw puller

http://crystalyte-europe.com/

look for " Open motor hub " on the left.. same principle as the 4 series

 

[CGameProgrammer]

Start drilling cooling holes. I'll post a pic later tonight of what I did to my X503 recently. One thing to keep in mind as well, is that the X503 is not the same as an X5303.

Please post this. I spoke with the vendor, ElectricRider, and they feel drilling holes is foolish as it could invite dust and dirt into the motor, and a mesh can't protect against that (not without blocking most of the air anyway).

Maybe I should leave out one of the screws and insert a temperature probe into its hole instead, and attempt to record internal temperature. Unfortunately I'm an electronics newbie so what would I need to purchase / wire to see temperature readings on a digital or analog display based off a single probe?

 

[xyster]

Is your hubmotor still under warranty? If so, I'd exchange it before possibly damaging it.

 

[CGameProgrammer]

It's not damaged (performance is still good, 42 mph on the flat with 72V 35A lead-acid) and it's not under warranty anymore anyway.

 

[Nimbuzz]

1) Well what the heck did Electric Rider say about the overheating??!?

2) Here is a $37 infared thermometer I bought that seems to work perfectly -- and it's pocketsized.

http://www.aerostich.com/catalog/US/Thermo-Hawk-Touchless-Thermometer-p-19146.html

OR
http://tinyurl.com/2pmmgj

3) game, have you tried any more night or morning rides? Is this heating intermittent or variable? Brng out all the details so we can get to the bottom of this -- tire pressure? Electric Rider's comments??

 

[CGameProgrammer]

ElectricRider at first suggested making sure the connections weren't loose and thus hot. I don't think anything was but I added some hot-melt glue to the Andersons used for the phase wires to make sure they stay stuck (and I ordered the plastic snap-on piece Anderson makes for just this purpose).

ElectricRider felt the black paint was best for dissipating heat built up in the motor, which is true if you don't count external sources of heat, but the outside temperature appears to be significant so I'm still not sure if it would really have been best leaving the black paint on.

Mostly they were against the idea of "overvolting" (as they called it) at all, not that that will stop me.

The motor still gets warm at night, but I'll have to do a scientific test, riding an identical route in the day and then at night. I did take a 9-mile ride of hard riding today in bright sunlight and the motor was merely very warm but not hot, and the outside temperature was only in the high 70s instead of the high 80s.

 

[Ypedal]

A question for fechter or the like, does voltage sag have any measureable effect on a motor's operating temps under load ?

ie:

72v of NIMH that sags 10v at 20 amps ( 84v peak charge operating at 70v under load )

vs

72v of LI that sags 2v at 20 amps ( 84v peak charger operating at 82v under load )

I realize that the NIMH battery itself will heat up significantly, vs the Lithium who stays cool, but what effect does this have on a motor, if any ?

 

[Nimbuzz]

game, In the middle of a hard ride feel the andersons, and all connectors for warmth. an Anderson can appear to be plugged in all the way but the internal contacts can get pulled back and not be in full contact. I would pull all the adnersons apart and look into the contact and make sure all look identical with the metal contact coming out to the front edge. The removeable contact snaps over a premanent metal strip that is the same color. Make sure all the contact pieces have come forward to snap into place over the front edge of the strip. If you can pull on the wire from behind and it pulls the contact out then it is not snapped on but can be pushed on throuhg the back of the connector housing where the wire goes in with a small screwdriver.

I'm wondering if the phase connectors are making intermittent contact as that could be the heat problem. What about the 5 pin hall connection -- what type of connector is that?

 

[CGameProgrammer]

The hall wires are soldered together. I might as well solder the phase wires too since the controller's effectively permanently on anyway (I detach it from the bike when removing the wheel so it can stay connected to it). The Anderson connectors used for the phase wires don't really snap like the big dual 50-amp one used for the battery->controller wires so it doesn't inspire confidence.

I'll report any possible difference.

 

[xyster]

How can a poor external connector connection cause the motor itself to heat more?