Cyclone kit suddenly failed? Throttle or controller..?

[le15otl]

It was all working perfectly running smoothly. I had tested it well on full throttle up a steep hill in a high gear (worst conditions for the controller to burn) Then randomly going at low speed it stopped running, the lights on the throttle stayed on but nothing happens when I twist it.

I can hear a very faint click from the controller when I twist the throttle but that's all. I checked the hall throttle the signal varies from 0.8v to about 3.7v seem to be ok?

All the FETs look fine I cannot see anything damaged in the controller, but when I measure from the phase wires to the main + and - wires the resistance on some of them is very high (much more than 10k) but, I don't understand how the fet's could have blown when I was hardly using any power and they also look fine.

(I'm almost 100% sure the motor is not the problem and the batteries are at ~48v which is well above the cut-off voltage.)
This is the one, i have the 36-60v controller http://www.turbobikekit.com/fast-cyclone-electric-bike-kit/cyclone-electric-bike-kit/cyclone-waterproof-geared-electric-ebike-kit-1920watt-3000watt-36v-60volt

Any help would be greatly appreciated.

 

[amberwolf]

batteries are at ~48v which is well above the cut-off voltage.)

Is that under load? If not, measure that under load and see if it drops below LVC. If it does, then check the cells individually and see which are lower than others.

 

[le15otl]

Under load it goes down to about 46v, not under LVC. Using a 20ah batt so there's little voltage sag.
I'm thinking that somehow one of the FETs is blown, because from the yellow phase wire to ground input the resistance is under 10ohms.
I just don't understand how and why because it never had any problems going up some really steep hills on high gears.

Thanks

 

[amberwolf]

Yeah, that sounds like a controller problem.

Is the controller inside the motor? (some of these types are, some arent')

Maybe heat-soak eventually got the FET(s) to a point where they just couldn't handle anything more, and finally popped at the next use while still hot (assuming the low-speed stuff was right after the heat-up test).

Probably will never know exactly what did it, unfortunately.

(apparently random controller failures have happened to a number people over the years I've been here on ES, and there are probalby many more that don't go on the internet for forums and whatnot, so never posted their problem, and some more that didnt' even try, just replaced the part or the bike or simply gave up and tossed the whole thing).

Good news is, you're here on ES, and there are a number of good controller troubleshooting/repair threads with tips on testing and fixing, if you need them. Some may be linked in the "index" sticky thread at the top of the various forums.

 

[le15otl]

It's an external controller (40A continue, 100 burst). The low speed was after 4 hours of no riding. I had just gone about 200m and it cut out.

This is the one: http://www.turbobikekit.com/electric-electronic-parts-voltage/motor/cyclone-controller-ecs-throttle-for-geared-6000-8000-watt-motors


I will try to follow the repair threads, not sure how well I'll be able to replace the FETs can't even find them online.

Do you know of a way to test individual fets to see exactly which one is burnt? Also I expected that there would still be a little bit of movment in the motor just stuttering and humming etc, but there's nothing? Is that normal symptom of fried FETs?

Many thanks amber

 

[Voltron]

It took me a while to get it...But the FETs are working the hardest during low speed pulling out, and that's when controller failures usually happen, not high speed, high back EMF cruising.

 

[le15otl]

I understand what you mean, but I had been going at slow speed numerous times before, accelerating up hills (while towing another bike!) and really pushing the controller. :?

 

[DAND214]

It just happens.
I blew one on the kickstand. Worked one time next think it didn't?

Dan

 

[le15otl]

Dammmn... I would still like to know if there's a way to test individual fets to see exactly which one/ones is burnt?

And I am interested because I expected that there would still be a little bit of movement in the motor - just stuttering and humming etc, but there's nothing? Is that a normal symptom of fried FETs?

Thanks again.

 

[amberwolf]

There's a lot of controller test and repair threads by various people with broken controllers that shoudl have the info you're after, but you'd have to look around for them.

There might still be an article on teh wiki on how to do this.

There also might still be an article on ebikes.ca too.


As for symptoms, it depends on the failure mode, and whether other circuitry was taken out when FETs failed.

 

[slowhands]

Dammmn... I would still like to know if there's a way to test individual fets to see exactly which one/ones is burnt?

And I am interested because I expected that there would still be a little bit of movement in the motor - just stuttering and humming etc, but there's nothing? Is that a normal symptom of fried FETs?

Thanks again.

You could have wiring or connection problems, that is more likely, double check them all. But let's assume it's a bad FET.

FETs are not easy to fully test on a DMM, and you are unlikely to have a more advanced semiconductor tester. You can make simple tests for shorts and opens. You need to have the FET out of the circuit, desoldered gate and source at least. I assume it would fail source to drain. Assuming they are N channel, if you connect gate to source you should have infinite resistance source to drain, that is, OFF. Put gate at say +8v, and you should have near zero resistance, milliohms, that is, ON. Merely putting the positive test lead on the gate may turn it on too. and a good FET will store the charge long enough to test Drain to Source. This will help you find a bad FET if it is a simple failure,, or it took out other components, or other components caused the failure, you have to go on testing.

These are crude tests, just a quick go/no-go to find a dead FET. Sometimes it is more expedient just to swap in a known good FET, or entire controller, to quickly find where the problem is. It may not be worth repairing a cheap controller, considering new ones cost under $50. If you are not an experienced tech, you could do more harm than good.

 

[le15otl]

I found the wiki, I think I know which FETs are bad, at least the 4 which might be. New controller is on the way in case I can't repair it. I will remove the suspect FETs later and try to find exactly which ones are dead.

Thing is I can't find them online, the serial code doesn't come up with anything. Will any FET which I can fit in place and has appropriate current/voltage ratings be ok? Or will that cause some issues.

Cheers

 

[le15otl]

f*** me the other controller I ordered just failed too, in the same way. Just a faint clicking from the controller, exactly the same symptom. How enraging...

Not gonna buy another cyclone controller, any other relatively cheap but RELIABLE controllers? (which will work with a cyclone ORO throttle)