Kepler
10 MW
Taken to PM so as not hijack the thread.
I love that there are those of us just unhindered by what everyone else tells you. It is how we move ahead in all things. Great job ac!ac246 said:Succesfully repaired my blown controller. It was just clicking when pressing the throttle, or when turning the pedals, or when pressing and holding the walk assist. I removed the controller and removed all of the rubber around it. Then identified the faulty FETs and ordered the exact same ones. P75NF75 and soldered these in their place and now it works brilliantly. Have reprogrammed it from the computer so now it starts gently and also limited the controller to 18A.
I spoke with a friend who is an electronics engineer and he said that one of the FETs will be shorted and causing the overload circuit to turn off the power. Which is why there was an initial click.
Once I took them out I noticed it was only one of the fets on the blue phase that was blown but without taking them out I would not have been able to tell. So I replaced all three. Cost me about £3 to order them on ebay but they would have taken 30 days from china so I also ordered some off of rapid electronics here in England for £15 with next day postage for 9 of them.
Hopefully this will help someone else repair their burnt out controller.
neptronix said:Are these new 9FET controllers that are failing?
I've got an old 6FET controller and bought the drive with the intent on running an external controller to remove a heat source from near the motor and increase the total thermal dissipation capacity of the drive ( therefore, i can either take increased efficiency, or just pump more power in - similar to running a magic pie with an external controller ).
If these are 9FET controllers popping, we may be in the same boat with internal controller hub motors. There is an issue with reliability even within stock limits.
Maybe you guys should start thinking about external controllers as well. Not just for reliability, but for increased power or efficiency as well.
ac246 said:The one I have fixed is a 9 fet controller.
fechter said:Nice repair job!
If the FETs blow, I'd suggest replacing all of them with better ones. It sounds like the underlying issue is the software allows destructively high currents under some conditions. Using FETs with a lower Ron will reduce the heating somewhat. There are some pretty good choices out there these days.
gorndog said:I've had 2 controllers blown on me already. The second was a warranty replacement from EV3MV. I thought I'd give the repair a go myself and it's worked! Total time was about 1.5 hours.
I ripped out the putting compound with a screw driver. It takes about 20 mins to get it all out. You have to be careful to not scrape the circuit board. I then used a multi meter set to 'Diode' / Audio mode which beeps when it detects a short. The first FET on the right hand side looking down on the controller had blown so I chopped it off and individually de-soldered each legs. I then re-soldered a replacement FET, fired it up and now it's works great. I also added some more thermal grease between the aluminium bar and the casing. I'm not going to replace the putting compound incase it blows again, just going to seal the outer case with black sugru and pop some silica gel into the half empty casing. The motor is now a bit lighter too. I've removed the throttle as I suspect this was causing the issue. I've ordered another identical one from Ali Express which should be here in a month or three.
Will be creating a build thread soon with pics of the bike etc.
fechter said:A fuse of the right rating might work, but I've generally found the FET will blow faster than the fuse, so the fuse only prevents it from burning after it blows.
It definitely sounds like some kind of design issue causing the problem. It may be difficult to tell exactly what without some fancy test equipment. Common fail modes include shoot-through, where the high side and low side of the same phase are on at the same time, ringing in part of the circuit, causing improper gate triggering or transient voltage exceeding the rating, overdissipation caused by excessively long on-time at low motor speed, and plain old overheating from excessive resistance losses.
Better caps on the power rails usually helps lots of things. These could even be mounted externally on the battery wires.
If the gate driver is sort of lame, it can cause problems. It's possible to damage the driver when the FET blows also.
If the pulse timing is off due to a software glitch, then all bets are off. Nothing in the hardware will fix that.
neptronix said:2 controllers dead? was one of them a 9fet? both?
Yes, both 9 FET. I also repaired the 1st 9FET and it blew again too after about 5 mins. I want to love this BBS system, I really do, but getting really frustrated with the amount of time and money I'm having to put into it. I was told that this system will run happily at 14s and it did for a while. I'm starting to think that It can't take that high a voltage, even though the FET's are rated to 75V.
All time one spends one the forum reading about other peoples experience means little until one tries it themselves.
I'm wondering if we can start to compile a list of system voltages and circumstances in which the BBS-0x system failed.