Why is my new hub motor making my controller super hot?

ClintBX

1 kW
Joined
Mar 6, 2014
Messages
419
Hi EPFMs,

I just received a new hub motor for m ebike and after connecting it and taking it out on m commute today, I noticed that my controller is getting pretty hot.

The new hub motor is a 1200 watt.
The controller is a 36/48v 30A.

I don't know an other specs for the controller like how many fets but I know that when I got it, the ad said it was rated for 500/600 watts.

The controller gets piping hot. I have it in a black seat post bag. Its one of those rosswheel bags. It even smells hot.

Is this happening because of the obvious? Do I need to upgrade to a controller that is rated for a 1200 watt? Or is it something else?


Clinton
 
It could be a false positive match on the Hall wires... It will run, but usually buzzy and slow accelerating, and crazy hot.

And 48v times 30a equals 1440ish watts at peak, so shouldn't​ heat up too fast on the flats anyway.

And ventilated is always better for controllers. The air cooling fins on the shell are there for a reason... I used an old soldering iron to burn vent holes in my bag on a stealth build. Then the holes don't fray but you get some air in there.

Hope you get it running better and enjoy that new motor!
 
I'm not sure about that false positive with the hall sensor. The acceleration is pretty good, I'm getting nearly 10 kms/h more compared to my other motor and I was surprised at the amount of partial throttling I had to do most of the time.

It does make a buzzing sound though which turns into a high pitch type of whistle at high speed. Just thought it was the nature of the beast.

My old hub, I have no idea exactly what the wattage it was exactly as I bought it second hand from a guy who bought second hand and had no idea about specs or anything technical about ebikes. It could be a 500w or it could be 1000w. Judging by the performance jump I'm getting from this motor, I'm guessing its closer to that 500w.
 
Main reason it's getting hot is because you have it enclosed in a bag. Mount it where it gets direct airflow and not in a bag. And if you're running above 15A continuous, you're going over the max watt limit. It will overheat in time.
 
Re the false positive, theres no guarantee the color codes match up on different replacement parts. My first hookup it ran backwards, lurching out of the stand and friction burning a hole in the carpet. I had to swap the yellow and blue on the phases and Halls to get it going right. You can measure the amps with the wheel in the air as one test... no load spinning should only be a few amps.
Maybe its just the bag, but maybe you will find you do have one, then you will really like the acceleration :D .
 
I think I might just upgrade to a higher watt rating controller. I don't want to chance it and I don't really want to mess up my controller pouch or the bolting to the frame unbagged idea.

How long can a controller last before it today ?

I'm also open to other ideas for cooling It as much as possible though.
 
ClintBX said:
I'm getting nearly 10 kms/h more compared to my other motor.
This could have something to do with it. It sounds like your new motor has a faster winding than the old one and this will work the controller harder.
 
I have a few burning questions regarding the amp ratings and watt ratings on controllers.

If a higher motor has a higher continuous amp draw on a controller, why is it that you can get controllers with a lower amp rating but a higher watt rating?

My active controller is a 30A 500/600w but I have another controller (I plan to swap with and see if it remedies it) its a 26A 700-900w.

My second question is about amp draw with partial throttle versus wide open throttle.

If you twist the throttle just a little bit to cruize at lower speeds for a while, does this save energy by drawing less amps or is this counterintuitive and draws more amps?

And thirdly, does an upgrade in amp rating mean an increase in consumption?

The vendor of my motor has offered me a compatible controller thats 35A and I'm concerned that the 5A increase might turn my machine into a vampire.
 
Well, for starters, some vendors sell controllers by listing whatever wattage rating will sell em. This is like ebay, etc I mean. They just say whatever it takes to move product.

So your amp rating is the more reliable number, if they give one.

You can take your amps x volts and get your max watts number, but that will never be your max continuous watts. As you get motor rpm up to an efficient speed, it can't draw max amps anymore. Depending on that and many other variables, your pull at cruise speed will be less. So some of that variation of the wattage ratings is vendors selling by max amps, or a more reasonable, likely to happen in reality, continuous watts at such and such speed.

We can fairly accurately predict watts by speed for a typical bike. (flat, no wind) 350-450w gets you 20 mph, takes 1000w to go 30 mph, and 25 mph is never less than 600w.

A larger controller, 1000w or more, will get pretty warm in normal use. You can't stuff it in a bag. It's got to get out in the wind or it will just get even hotter, shortening the lifespan of it.
 
Thanks dogman. So, I don't have to worry about extra drainage if I'm doing a lot of stopping and going?

dogman dan said:
A larger controller, 1000w or more, will get pretty warm in normal use. You can't stuff it in a bag. It's got to get out in the wind or it will just get even hotter, shortening the lifespan of it.

My controller is only 600 watts. Are you saying that a 1200 watt controller will still get equally toasty?
 
Don't put the controller in a bag.

No matter what electronic item you have, if it can produce heat, don't put it in a bag.

If it can produce heat, put it in the airflow where it can be cooled off.


If you put something that produces heat in a bag, the heat stays in the bag with it, and the something that produces the heat gets overheated and damaged or destroyed.



So don't put heat-producing things in a bag, put them in the open air.
 
If you do more stop/start riding, or have hills, then you'll also use more power, more battery capacity, and generate more heat (in both the motor and the controller (and wiring/battery/etc)) than if you do just a continous no-stops run of the same distance, on the flats.


If you then also increase the power capability of the system (and use that capability), you will use even more power/battery capacity and generate even more heat with the same stop/start riding path/conditions.
 
Alright guys. I see theres a strong consensus against storing controllers in a bag. How exactly do you fasten a controller to a bike without a bag so its not exposed to the elements? Am I supposed to bolt it to the frame? I'm especially concerned about the connectors.

Alternatively, are there any mesh pouches out there designed for this so it can get maximum airflow?
 
A quick forum search for the phrase "mount controller*" just in the title, finds a number of threads with ideas

https://endless-sphere.com/forums/search.php?keywords=mount+controller*&terms=all&author=&sc=1&sf=titleonly&sk=t&sd=d&sr=topics&st=0&ch=300&t=0&submit=Search


I"m sure there's more if searching by individual posts, or in the text rather than titles, but it'll probably take longer to find relevant results in there.
 
Thanks for that link amberwolf. Some interesting approaches there.

From some of what I read, mounting the controller to the frame can make it vulnerable to shock.
 
I'm attracted to the idea of creating ventilation holes in the controller bag. Has any done this approach successfully? Can it work?
 
It would be cool to make an air duct that ventilates the controller more the faster you go.

Anyway, here are your options, as I see them, if you don't want to do that:

1) Put it in the open air (great idea).
2) Replace the case, or side of the case that bolts to the mosfets, with a really big ass heatsink or heat pipe.
3) Get a much bigger controller, and run it derated.
 
option 3 only works if the heat input is very short infrequent bursts, and there is sufficient heatsinking material to keep absorbing those heat bursts and keep the component temperatures down (especially electrolytic capacitors as the liquid in those is heat sensitive, and most of the common ones I've seen used aren't rated for higher temps, and even the ones marked as for that don't even survive normal temps in some cases so are probably counterfeit or marked with fake specs).

Important to keep the FETs cool too, but they're usually directly connected to heatsinks so tha'ts easier, while caps and other internal parts (MCU, etc) that might be heatsensitive (even if they don't generate their own) are just in the middle of the little controller oven box, so any heat that doestn' get sucked out of the case by the heatsinks ends up cooking those internal parts.

Once the caps begin to degrade, voltage spikes and sags are more severe, whcih causes more heat, which degrades the caps further, which causes more spikes/sags, and eventually it can reach a point where the FETs or other parts simply can't tolerate this, and fail.



Something to think about when designing your cooling solutions.



There have been various threads where people have added cooling to their bags or boxes containing controllers or wiring or batteries, including methods that try to maintain weather resistance of the bags/boxes/etc. I don't have any links but I'm sure a good enough search could find them. :(



FWIW, if vibration from frame moutning a controller causes it to fail, it wasn't well-designed mechanically to start with. ;) Potting controllers with thermally-conductive epoxy or silicones helps with both that and heat issues, as well as weatherproofing (see the Phaserunner from Grin for an example). But it makes them unserviceable for the average person if they do fail. Not a problem for most people as they'd just toss it and get a new one regardless, but for those DIYers among us..... ;)

Gluing down components like big caps, resistors, and wires will help with vibration too, and still leave the controller serviceable, if desired. There's a number of threads about using PSUs as onboard chargers that discuss this, including methods of gluing and kinds of glue.
 
Mount the controller with the wires entering from the low end and downwind. Water doesn't flow uphill and into the wind. More weather-proofing is needed at the connections.

Partial throttle pulses create more heat in the controller due to sharper higher spikes in the phase current...even and especially at low speed where the motor isn't creating enough BEMF to help self regulate the current spike. eg On my 60mph capable cargo bike with well over 1000 miles of riding in that specific configuration, I blew the controller on a 1/4 mile stretch of bad road that forced me to ride at less than 5mph. The little bumps created tiny throttle pulses to maintain that speed, and I unknowingly overheated the controller, so as soon as I got to smooth road and hit the throttle normally, it popped.

It's also probable that your new motor has lower internal resistance, and despite the current limiting features inside the controller, connecting it to a lower resistance load will cause it to pass more average current and create more heat.

Since it's a new combination, as stated by other above, you should verify that you have the correct wiring combination. Most motor and controller combos will spin the wheel with some incorrect wiring configs, but the resulting advanced timing is extremely stressful on the system, which has burned up countless controllers and motors. The only way to be certain is to (with the wheel off the ground and chain off so a reverse doesn't create pedal havoc) is to try all 6 possible combinations of either the hall or phase wires (never switch both at the same time). Every hall combo has one valid phase combo, and every phase combo has one valid hall combo, so swap only the set that is easier. Use small short duration throttle pulses to check each combo. I use alligator clip jumper wires to make swapping simple and only swap 2 wires at a time to be systematic without notes. The thin jumpers also act like fuses to make sure high current can't pass and damage something. The valid combo will have the smoothest most silent start, and it will have the lowest current. If the valid combo spins the wheel backward, then swap 2, any 2, only 2 of the wires you were keeping static, and go back to finding the new valid combo which will be forward. Note that false positives often spin the wheel up to unusually high speed, and they have higher no-load current.
 
John in CR said:
. Note that false positives often spin the wheel up to unusually high speed, and they have higher no-load current.

Could this be what's happening here? I know that my older motor, I had to flip the configuration of the phases (after upgrading to a 48V controller). Are false positives harmful?

I guess I will test out the config of the new motor's phases. I don't want to touch the hall sensors. 5 tiny wires in a box connector. I had to re-crimp halls in my last motor and still had issues when throttling.

I do notice that there's a slight stall when taking off from a dead start and in fact, I feel I have to always take off gently with the throttle. (though that could be a habit from my old hub)
 
ClintBX said:
Could this be what's happening here? I know that my older motor, I had to flip the configuration of the phases (after upgrading to a 48V controller). Are false positives harmful?
Yes. The first reply to your thread here tried to advise you of this.
 
amberwolf said:
ClintBX said:
Could this be what's happening here? I know that my older motor, I had to flip the configuration of the phases (after upgrading to a 48V controller). Are false positives harmful?
Yes. The first reply to your thread here tried to advise you of this.

I shrugged it off because I didn't know that a false positive can result in a motor that runs faster with more acceleration. This seems odd. Im familiar with the backward spins and the grindy false positives. Why can't they just standardize the wires so they match every time?! I really hope I don't have to mess around with hall sensor wires.
 
Back
Top