Strange controller problem

[LHelge]

I need some help with a problem om a street legal bike I'm building.

I've been trying to run the bike on a 36 V battery, but the BMS cuts off as soon as i connect the battery. Yesterday I tried using a 35 V, 10 A power supply instead. When I turn on the PSU and gradually increase the voltage the controller starts working when the voltage is above 20 V then it spins faster and faster when I increase the voltage drawing a couple of amps until the voltage exceeds 30 V. The the current meter peaks for an instant and then the PSU overcurrent protection turns it of.

My first guess is that the controller have 30 V MOSFETs but I have opened it up and the MOSFETs say S140NF75 which are 75 V. I've attached a photo of the controller

 

[wesnewell]

Well, since the controller is marked as a 24V controller with a 21V LVC, I'd guess it's a 24V controller with a HVC somewhere around 30V. So if you want to run a 36V battery on it, either mod it for 36V or get a 36v controller.

 

[LHelge]

I was told this is a 24 V - 36 V controller but maybe I'm misinformed, as you can see on the attached image the 36 V box is checked as well. Mi first guess what would happen, if you apply higher voltage than the controller can handle, is that the LM317 goes up in flames. How would you explain that the controller shorts the battery over 30 V? Is this some feature built in to the infineon chip?

Maybe the checked 36 V box means that the hardware can witstand 36 V but I need to reprogram the LVC & HVC for that to work?

 

[fechter]

It seems like that controller should be OK for 36v.
If there is a sudden spike in current, that would indicate some other problem.
There are a number of possible things that could do that, but I can't think of a reason it would spike the current when you slowly exceed a certain voltage threshold.
You might try it without the motor phase wires connected and see if you get the same spike. If so, then the problem is in the controller. If not, there could be a problem with the motor halls or motor wiring.

 

[LHelge]

I think the phase and hall wiring is correct since the wheel turns with good torque as long as the voltage is below 30 V. Maybe I shouldn't say spike, the controller shorts the battery when the voltage exceeds 30 V. I havn't tried it without fuse/BMS yet but I'm pretty sure that the magic smoke would escape pretty quick if I did!

 

[d8veh]

It sounds to me like your battery is faulty. See if you can get any current out of it with any other device like three 12v bulbs in series. Check the main fuse is making good contact and check that the battery connections are making god contact. Could also be a faulty BMS. I've had a battery that measured 41v with a voltmeter at it's connector to the controller, but it absolutely refused to give any current.

 

[LHelge]

That was my first thought as well, the battery has already been replaced and I've tried 3 different batteries.

But as I wrote before, it has the same problem when I run it on a powersupply while the wheel rotates freely. It draws around 1 amp between 20 V and 30 V and when i turn the voltage up to 30 V it instantly spikes to 10+ amps which is the maximum for the power supply and the overcurrent protection turns it of.

 

[fechter]

That IS very strange. I've never seen anything like that. It sounds like the controller is allowing shoot through (both high and low side switches on). Why it would do this when the voltage goes over 30, no clue.
It seems unlikely the controller has some kind of overvoltage protection circuit that works like that, but I suppose it's possible. More likely there is some kind of fault in the gate driver circuit.

If you had a way to hold, say, 5 amps steady on it after it shorts, you could see which parts are getting hot inside the controller.