View Full Version : How to bench test a motor with load?

16 May 2011, 1807
So, one of the problems i'm currently having is that my controller cuts power because it thinks that the motor is over temperature. I'm not sure it is or not, but i can only see this situation when i'm riding the bike for a while. it never happens on a bench test.

I'm assuming it's because i'm not stressing the system enough when there's no load.

So, do you guys have any idea about what i can do to load the system up? I tried applying the brake, but i think that actually be part of the problem...since i have a hub motor, applying the brake makes the rotor heat up, and since the rotor is bolted to the motor, that's just adding heat.

Any other ideas?

16 May 2011, 2141
If the brakes aren't dragging when bench testing, why would they drag when you ride?

16 May 2011, 2151
Just do a sustained burnout!

Honestly, I believe your motor is rated for 10 kW continuous. It's going to be hard to place that much load on it without a dyno (or something a lot like a dyno).

Maybe we can help you diagnose it under the conditions where it happens, when your driving. Do you have a reading of voltage and current, and how long you have to sustain those levels for the controller to cut out?

When it cuts out and you touch the motor, how hot does it feel? I don't know about the Enertrac, but most motors use class H insulation which should handle 350 deg F. If you touch it when it's that hot, it should burn you.

16 May 2011, 2207
Assuming that the motor can handle a gentle ride then I'd unhook the temperature sensor and take the bike for a gentle cruise. If it still cuts out then it's not the temperature sensor. Ask Mark what the sensor is supposed to read and see if you can hook up a meter to read its resistance or voltage or whatever it puts out, (it might be a simple make or break switch at a given temperature). If it's the controller overheating then see if you can get some kind of data log from it.

Nuts & Volts
16 May 2011, 2212
Try pulling something heavy. This should draw a large amount of current and heat your motor up quickly. You can then check the temp with some sort of thermostat or a laser thermometer from harbor freight. Draw enough current until the controller trips the temp fault, then quickly read the motors temperature

Or make you wheel a lot heavier, but putting link-chains or something on it

You could also set the temp limit very low and see if you controller loses power the same way it is now. Then gradually increase the temp setting to determine if it is always the temp setting that is tripping the fault. Can you ride without the sensor as well?

17 May 2011, 0313
I've had a couple of projects where I wanted to test a motor under load... if that's still the subject. :D

One was to test how fast it would accelerate, and the idea I had was to have it drive a large, rotating mass, like a grinding wheel or heavier. It would apply the load in the same way that you'd get it on a full-throttle acceleration, not counting drag. Huge load at first, then dropping as it gathered speed.

The other test I wanted was the opposite, load growing as the RPMs increased, and the easiest way to do that, I think, is with either a wind-load or water-load tester of some sort, similar to your stationary bike load fans. With the motor hooked up to a water pump you could control the load with a simple valve, I think, and actually dial it in for whatever you wanted to apply... not quite sure how that would work, now that I'm actually talking about it.

A friction-load bed of some sort would apply a constant load, either simple pressure on a drive wheel, as in brakes, or dragging a weight across a smooth surface, as in a tractor-pull sort of affair.

I really think at some point, somehow, some of us need to get together with all the motors that Noah graphed, and do this type of testing. I'm honestly not sure how a dynamometer works, but I'm guessing you can apply all three of these types of load... anybody know for sure?

(googles it. http://en.wikipedia.org/wiki/Dynamometer. Yeah, ok, there are various types of dynos that do measure these...)

17 May 2011, 0313
Hey Mark,

well, i have the new kelly controller hooked up, though, i'm not actually sure if the problem still exists or not. It seems to be better, but i've had an occassional dropout, but it seems somewhat different than before.

One other issue i'm having, and the one i'm writing about is that after riding a while (10 minutes), I start getting motor over temperature warnings. i'm also having some rear brake sticking issues (i had a shop fix this, but aparently it's not totally fixed), so that could be adding to the heat issue.

However, I was wondering what is a normal range for the motor temperature? I don't recall offhand what i have the controller temperature warning set to, but i think it's the stock setting. Also, when i got home, i used an infrared thermometer and read that it was 170F when pointing at the grills along the side of the motor.



EnerTrac Corp. to Alex

show details May 13 (3 days ago)

the Motor temp should be set between 125 and 135C
More important is the temperature of the Rotor The outside Shell should not be over 70C
I don't think there is a heating problem maybe the wiring isn't right you have the 1K resistor done correctly (sorry I have to Ask)?
What are the conditions when it shuts down?

Best Regards
EnerTrac Corp
www.enertrac.net (http://www.enertrac.net)

EnerTrac Corp. to Alex

show details May 13 (3 days ago)

Hi Alex

One last thing keep a log of your watt-hr/mile off of the CA unit basically if you keep it under 130 watt-hr/mile you should not overheat ever, 150 is kinda a limit 170 for any length of time and she will overheat. Make sure your shunt is calibrated and these numbers will guarantee the motor will be fine.

Great. Though unfortunately, i don't have a CA. I have a setup where i have 6 CellLog 8S, so i can tell voltages of the individual batteries (though, they're not all hooked up at this time). This is similar to what you had on yours when i rode it a year and a half ago.

This setup doesn't allow me to tell the total voltage or Wh. though i guess i could set that up....or get a CA. Actually, i was thinking of getting a CA (which one i'm not sure...), or something like this: http://minibms.mybigcommerce.com/products/EV-Display.html, do you have any insight as to which would be better?

As for your original email about wiring the 1K resistor properly, i'll definitely check on that. I'm taking the bike to the TTXGP tomorrow, but since i'm not really riding it much there, things should be ok. i'm not quite sure what you're asking when you ask for the conditions when it shuts down, i have the wiring of the controller such that when there's a fault in the controller, the main contactor disengages. When the controller senses an over temp, it goes into fault and bye bye throttle.

Thanks as always.


PS, since i'm taking the bike to show, i have all of the fairings painted and on, i'll send you some pics shortly for your site. :)
- Show quoted text -

Reply to all
EnerTrac Corp. to Alex

show details May 14 (3 days ago)

I just mean are you accelerating going steady at what speed that sort of thing

You can get a estimate of power used per mile if you know the AH either used or put back in by the charger.
Cell voltage will give you no idea how much power you have used or have left in the tank so to speak. Only a AH gauge of some kind can do this.

I use cell voltage to give me an idea of cell health not to determine range. The voltage falls off the cliff when the cells go empty with no usable range, i've gotten stuck a few times looking at cell voltage - 3 volts - then half a mile later I'm at 2.5 volts resting voltage and stuck Only counting AH or amp-hr/mile can give one any feel for range without being surprised with a dead battery.

On long rides I'll baby the throttle and try to maintain 80 w-h/m and get 30 or 40% more range or beat on the bike and get 150

it's a great feedback parameter.

Anyway the LED has an error code for temperature; is this what you are getting you can stop the contactor from dropping out if need be to read the code

Have fun at the track

17 May 2011, 0610
If the brakes aren't dragging when bench testing, why would they drag when you ride?

The brakes are dragging while bench testing, but they're not fully clamped down. One of the possible suspects is the brakes dragging some (adding heat) + load on the motor (generating heat).

17 May 2011, 0620
Thanks for all the replies, I was hoping to not need a dyno, and with some ideas here, maybe i can pull something off. As mark has shown, i've been in touch with him, and he has been (as always), super helpful. As per his suggestion, i'm going to check on the wiring of the thermistor to make sure i didn't screw that up, but am also looking for other ideas in case that's not it.

One reason i'm asking about ways to do this on the bench is that I don't have a CA so can't log voltage or current draw, hence I need to be able to look at measurements for voltage and current draw while it's running (seemingly with load), and it's hard to do that when riding.

In any case, this is an interesting discussion even outside of the issue i'm currently facing.


17 May 2011, 0837
When you say "brakes dragging" are you talking about the rear, front or both?
If just the rear brake is the heat source you are concerned about, why not just yank the calipers off and go for a test ride?

17 May 2011, 0842
when you say dragging, what do you mean? Does it actually stop the wheel immediately if you let off throttle?

Brakes are supposed to be in constant contact with the brake rotor. is it possible it's adjusted too heavily to one side?