Magic Pie Burned Out?

[Ugrd]

So I guess il start with a story since its most helpful in these cases. So Ive had my magic pie for quite some time (a year) and ive put 2000 miles on it. After 500 miles I decided to up how much power it got, I got a Lyen 12fet Controller and bumped the voltage of the battery up to a 72v lipo. I replaced the hal sensors in the magic pie with S41's. All went well and got a top speed of 38 mph with it, pulling ~3500 watts on full throttle. Went another 1500 till last week, I was out riding and was going full throttle and then my bike started shaking, So I pulled over and inspected it, it seemed like a phase wire came out or something. I went home and tested the wires and everything seemed fine. It was still ridable it just vibrated badly and pulled twice the amps it normally did, it also didn't accelerate half of what it used to before, the throttle would also cut out randomly. The motor got really hot (it didnt before), and when it got to a point it would stutter REALLY badly, to the point where I couldn't go forward, but if I let it to cool it wouldn't stutter anymore, it just went back to vibrating. I think its the hall sensors because hall sensors hate heat but I don't know. I rode it like this for the past 5 days and it was fine besides vibrating. Yesterday I went out for a ride and went 500 feet and it just died. No throttle, not even stuttering. Just dead. So I don't know whats wrong. Could it be a burned motor or the hall sensors? I pulled apart the controller too and one of the capacitors got really hot when the battery was plugged in but with no throttle. I hope its the motor and not the controller, because I wanted to rewind it anyway... Can anyone tell me anything? Here are some pictures...

The capacitor that got super hot is the left one in that cluster.

Thanks,
Ugrd

 

[Bill Nye the science guy]

can you run it sensorless with that controller? is so try to do that and see if it works

check and double check connectors and solder joints, had a car with a broken solder connection in heat shrink, fine when cold but once the heat shrink would heat up conection would break.

check ohm reading between phases of isolated motor conections. should all be close in ohms

check throttle voltage output. had a problem with a throttle that caused a very high pitched whine and a hot cap.

 

[dnmun]

if the input caps are getting hot that is not good. can you point to what you think a capacitor is?

 

[dogman dan]

My guess would be the halls died. Amazing it took so long, you must ride pretty smart, or have shorter trips to make.

 

[dnmun]

that motor does not look cooked. he needs to test the halls and the mosfets first but i would like him to show the capacitor that is overheating.

do you have a voltmeter to test the controller and the hall sensors?

 

[Ugrd]

The impedance of all combinations of the phase wires is 0.03 ohms so that's fine.

But when I went to test the hall sensors. I tested the ground and power first. And it read 0.15 volts. I tested each of the hall sensors and they came up as 0.9 volts. Which doesnt make any sense to me because hall sensors need 5v to operate. Is the 5v regulator messed up or something?

Also I circled the capacitor that gets REALLY hot.

 

[dnmun]

if it is getting hot then it is shorted and i am certain that is the 12V rail. so there is no 12V on the output of your regulator because the output is shorted by that capacitor. the regulator is the little IC to the left of it that says 78L15 on the top. you need to replace that cap.

 

[Ugrd]

So I ordered a 220uf 25v capacitor off ebay since my radio shack was out of stock. Hopefully the capacitor is the only thing shorting. And then hopefully the hal sensors are ok. We will see...

 

[mlt34]

Hopefully swapping out that cap will do it for you. The good news is that if you have to replace the controller, its a lot cheaper than burning through phases in the motor or something like that.

Hopefully the halls didn't get fried in the process of the cap failure. You can check them with the diode setting of your multimeter. Put the positive probe on the black lead of the motor (when its unplugged from the controller). Next measure the yellow, blue and green leads with the negative probe. They should all read a nearly identical number, something in the 0.500-0.700 range. If so, they are all fine. If you get one or two giving an OL reading, you know they are fried.

 

[dnmun]

you don't understand.

the 15V regulator provides the voltage to drive the transistors that turn the mosfets on and off. the 5V that is needed to supply current to the halls, the throttle and the microprocessor is supplied by the 5V regulator from the 15V rail that is now shorted by the dead capacitor.

with no 15V then there is no 5V. so if you had tested the controller properly you would have measured the throttle voltage to see if there was 5V there at the throttle or the halls.

there is nothing wrong with the halls so try to avoid letting people talk you into thinking they are dead unless you actually test them and find they actually are dead. there is no reason to think the hall sensors are dead.

 

[Ugrd]

So I replaced that cap and now its not getting hot but now I have TWO MORE things that are getting ridiculously hot...

In the red, I think this is a voltage regulator but I could be completely wrong. But this thing gets REALLY hot, well over 300 degrees im sure. Im surprised it didn't melt yet...

Also this green resistor gets REALLY hot, and you can see brownness on the outside of it, so im guessing it got REALLY HOT, also the red wires around it were parity melted, but not showing the actual wire...

So I dont know if this is some type of short in my halls or something (I already checked shorts) but do you think I need a new controller or should I contact Lyen?

 

[dnmun]

did you measure the voltages on the 15V rail and the 5V rail?

that green resistor is the input voltage resistor is part of the regulator i suspect and the three pin regulator should have something like 7805 printed on it but sounds like it is having to handle a lot of current if it is so hot.

but obviously it is not your hall sensors that failed. or does the heat stop when the hall sensors are disconnected?

something is shorting the controller and i am surprised that the cap you replaced is still hot.