Cut-Off Problem due to bad mosfet? but measures ok

[marce002]

Hello

A motorcycle VOTOL EM50, was fine and ok for 3 months 70kmh ok... then suddenly have : soft start barelly keep going and sometimes CUT OFF happens,,, fast start ends up in cut off..... no battery problem, swapped controlled to stock one, and all is fine, so the issue is inside the votol controller, my question is :

since all seems to be crystal clear no signs of damage at all, can it be the mosfets despite they are all measuring ok with diode mode multimeter, (in circuit test, i did not take them appart).... was using 96V max volt specs though..... caps seems to be ok too..... what else to llok??? should I take mosfets appart to be sure thay are all ok?? what is the chance some mosfets have ruined even in diode mode and resistance look are identical (neg and positive group of course)

 

[Dui ni shuo de dui]

In my opinion if you had defective mosfets the controller wouldn't work at all.

The Votol hardware is actually very solid and pretty well built, so it is not very likely to be the culprit.

What is extremely bad with votol is the software, so make sure the parameters you entered are 100% correct. Also make sure you properly followed the motor angle identification procedure and don't hesitate to redo it several times until it works. That is likely to be your issue.

When the cut off happen the controller should also display some kid of error description, I suggest you look for it so we could better know what is going on.

 

[Gruesome]

May I also add that while it is certainly possible to take Mosfets apart, it is near impossible to put them back together...

 

[marce002]

ok make sense... i will check my params.. maybe they got changed randomly or in the many on off cycles when i was programning my android app connecting it to an hc05 bt module.... the unit was working fine for 3 months with correct params ... angle was set -60 in my case all was correct except flux that i did not use yet (because still no boost sport wire connected).

regarding the mosfet replace and puting back i wonder why you think would be imposible? if cut legs of existing and solder new ones cutting legs?? or why you said??? in case of them to be bad i wantef to buy more powerful ones like irfb4115 they are 150v and 100A instead of 100v

 

[marce002]

ok the parameters are ok, they were not modified or altered a alla and reflects the version i was using for 3 months, and is attached as image.....
Still wonder what can be wrong... if mosfets that measures perfectly well can be damaged.... or caps o what to look internally, the scooter works perfectly fine with stock controller (sunra hawk 3000watt)

 

[fechter]

When mosfets fail, they almost always go shorted. Sometimes they short, then explode and go open.

I'm not really familiar with the Votol, but I would look for the current measuring shunt (or sensor). This is typically between the battery negative and the bus feeding the mosfets. If the shunt went open or degraded for some reason, the controller will sense abnormally high current.

 

[marce002]

I can tell you they seem to be ok then... also because all tested with multmeter to be fine...
caps cant tell because they are 5 not 2 per phase..... the whole circuit board and compomemts look perfect and new... i cant understand why cut off is still happening... around 7 8amps... no error codes in votol software..... no weird noises,,,, tested with another bat pack....it is not the battery.any other help involving circuitry?

 

[Gruesome]

regarding the mosfet replace and puting back i wonder why you think would be imposible? if cut legs of existing and solder new ones cutting legs?? or why you said??? in case of them to be bad i wantef to buy more powerful ones like irfb4115 they are 150v and 100A instead of 100v

You said 'take apart', not 'replace'.

 

[Dui ni shuo de dui]

I can tell you they seem to be ok then... also because all tested with multmeter to be fine...
caps cant tell because they are 5 not 2 per phase..... the whole circuit board and compomemts look perfect and new... i cant understand why cut off is still happening... around 7 8amps... no error codes in votol software..... no weird noises,,,, tested with another bat pack....it is not the battery.any other help involving circuitry?

Again, forget about a hardware issue, it is a software issue.
The votols are a huge pain to get to work sometimes, I spent days getting mine to work with my motor even though it was working just fine with a similar motor just seconds ago.

I can see some of your settings that don't seem correct to me, like HDC is still enabled, which usually is not desirable (it's probably not what causes your issue now, but still). You can try different phase wire combinations and different motor angles, like -60 degree instead of 60 degree, for some reason some work better than other. Always do that with the wheel off the ground, because in some cases the motor can do very weird stuff.
There is one combination of parameters that will work for your motor, you just need to find it. Don't ask me why the votols are such a mess, I don't know. But unfortunately that's just how it is. Don't waste your time opening it, you aren't an electronics engineer and if you were you'd see that the votols are very well designed and pretty much bulletproof.

There is a very long thread on the forum about votol controllers, I suggest you take a look at it if not done already.

Once these controllers work, they do work great.
Good luck and patience!

 

[marce002]

ok ok sounds reasonable... only thing is 3 months working perfect with that parameters in the pocture .... same battery nothing changed .... that is the point.... i wonder if my 96v ruined something (96v but no more than 50A and most of time under 30A) so do not think so.... again components looks perfect.... the phase angle was adquired automatically like i yave read in the thread you mentioned... using a Port and turning on and off several times intiñ i saw that 60 was correct... will try -60 of course... o feel that you are right does not seem to be a hardware issue but.. ..so sad and weird.
will try with another motor.... a 120 degree ebike hub motor to check if it is working fine storing current parameters and will report back... then will try -60 and hdc settings.. thanks!!

 

[marce002]

basic hub motor from ebike test.... to help diagnose my problem instead of using the motorcycle...
. im having trouble to make it work too.... i have loaded the stock firmware for the votol and does not start.... tried 120 angle and also not spining...I have connected hall..power lock...power and throtle
im using my bench power supply with 42v undervolt ok...can you please help guess pole pairs.... shift angle (120 says the controller that came with this motor) and other params that need to change from stock firmw?

 

[fechter]

Try checking the 5v line going to the halls (or throttle) and make sure it's around 5v. It's possible something in the voltage regulator failed.

 

[marce002]

yes it is 5.04V ... i think this unit is almost dead because now the symptom us that can save parameters, it connects and retrieves phase current at 0 and also overvoltaje at 0, no matter what i write to them.... always retrieven that fields at 0, maybe others params too...trying to write stock factory default.ini when i retrieve values they also ar in 0 for those fields.....
sad

 

[abcaste1006]

1) YES you are using flux weakening and also is a little high 4000. For Hub motors you are supposed to set Max 1200. If u exceed that value your motor can demagnetized.
2) You said there was a -60 angle at begin. The does it magical changes to 60??
3) overcurrent often means a short circuit mosfet.

 

[Dui ni shuo de dui]

ok ok sounds reasonable... only thing is 3 months working perfect with that parameters in the pocture ....

Ah sorry I did not understand that before, I thought you were using a different controller for a few months then swapped it with this new one that one didn't work. My bad, maybe my english isn't good enough, sorry about that.

If it was working fine then suddently doesn't work anymore then yes you are right, this might indeed be a hardware issue.
Again, all my apologies for not paying enough attention.

Do you remember what you were doing when it went from working fine to suddently not working? Was it under acceleration? Under braking/regen?

 

[marce002]

ok, i was using a 96v battery, (yes i know a bit high, but a lot of sag so with load was 86v average), doing 8km road to work and 8km back home every day.. speed average 60kmh.... internal temp was 25 initial and ending at 37deg... those values were monitored with my own android app (see picture) ... all of a sudden, the bike started to fail turning off the motor but not the display...... then i though mosfets would habe been damaged and took it apart.... from my small knowledge they seem to be ok, but not sure because "in circuit" they all are measuring the same in diode mode.... also phases between grond seem perfect.... caps i do not know, also in circuit would not give correct values i think.....
when I started to try a new motor from my ebike, to prepare a desktop laboratory test with ebike front wheel hub motor, i have changed angle to 120 and some other parameters, like ports settings to learn the new configuration.... the ebike testing motor never worked and then could not recover the factory settings anymore... because bringing any of the configuration files and factory ini files BRING OVERVOLTAGE AND PHASE CURRENT AT "0" !!! .... so my question was if MOSFETS might be bada even if i think they loog goog at multmeter tests, or if someone can indicate of any sense resistor that can drive me to a failed phase current and or overvoltage issue.... note: i have reflashed also the firmware from a BIN file in this thread for my EM50SP and did not change this problem... so now I have this bricked that need to recover because should be a simple smd resistor or something replaceable!!!! do not think it is a firmware or setting issue...... thanks for any clue~~


i ve seen this video with similar problem but more advanced model, and he said something of Debug mode typing ~~ in ERROR COUNT filed... will this work for em50sp ???