Bafang Hydro Brake Switches Issue

[TNC]

I'm loving my BBSHD on my Santa Cruz Nomad MTB. I've had one issue that befuddles me that involves the hydraulic brake lever sensors. I don't see a lot on the forums about these sensors besides people having questions or issues about where/how to mount them. With only a little head scratching I was able to figure out a solid way to position and fix the magnet and the sensor that appeared to work perfectly. The error code symbol comes on at a perfect time and goes off just right when the levers are engaged and disengaged.
Now the problem. At first, the motor stopped exactly as it should with the engagement of the brakes and came back on when the levers were released while riding. After just a few miles, however, the motor quit running for no apparent reason. No error codes and no error code indicating the brake sensors are triggered which would kill the motor as required. The display looks fine, the motor is not hot, and after quite a bit of poking around and inspection, nothing appears to be amiss. For no apparent reason, the motor would come back on occasionally, and the bike/motor would behave as expected...then it would quit again. Believe me, I've disconnected, reconnected, rechecked, and even reprogrammed back to OEM specs all elements of this package...except the brake switch cables.
All obvious and critical components that are required operational appear to be functioning. I am new to ebikes and the Banfang mid-drives, but I'm a bike mechanic and automotive/motorcycle mechanic, so this stuff doesn't seem completely alien to me...though the way the programming elements all come together to provide a smooth power delivery that suits the rider and conditions are somewhat interesting and apparently not always crystal clear to even the highly exposed...LOL! Still, not crazy mysterious from what I've seen reprogramming this package. In fact using much of the excellent direction from many on the forums and internet in general, I've come to a program that I'm extremely happy with for real trail riding.
So, today I was at it again with a dead motor and noticed the brake sensor error light that indicates the levers are killing the motor as required seemed quite slow to go off after the lever was released. This was not normal. I totally understand that as long as this brake switch error code is triggered that the motor will stop. This slow response of the brake error code to go off got me suspicious, so I disconnected the brake sensor cables. The motor immediately came to life. I rode the bike around quite a bit, and the bike/motor all functioned flawlessly. The odd thing to me about this deal is that the motor was dead even when the brake switch error light was off. In other words there was no indication the brake switch was killing the motor...but...it certainly appears to be the case.
I'm at a loss to figure out why the motor is acting like the brake switch(s) are triggered even when the display doesn't show it. The display is a 500C. I'd tend to say the display is working as it should in this regard. I really appreciate those brake sensors when mountain biking as doing a back pedal in most instances isn't always ideal to kill the motor. Now, I will say my latest programming effort has the motor stop quickly when I stop pedaling, but as most of you know it's not an immediate stop like you'd want when riding technical terrain. The brake switches provide that and also let you get back to pedaling quicker without jerkiness.
Anyway, anyone run into this or found a good solution for living with the Bafang brake switches in a less problematic fashion?

 

[Gionnirocket]

The only thing I can think of is that the display and controller act in parallel with the signal from the brake lever... Meaning that they process the same signal independently.
I think I would start by bench testing both brake levers and see what the signal voltage is at rest and activated.
Perhaps one of the levers is leaking signal that is enough to satisfy the controller but not the display.
As far as the display having a delay and extending the brake indicator, I think this is normal as my 860C does this and I believe that they are made by the same 3rd party manufacturer. The delay is 1 second or less.

 

[TNC]

Rocket, I think you may have hit on something. I just came back from a ride, and while out there I did some experimentation. With just one brake switch cable plugged in, I didn't get any unwanted motor shutdown. It didn't seem to matter which brake switch was plugged in as long as it was just one of them. Plug them both in, and the bike runs fine for a bit, but then the brake switch error pops on and the motor predictably quits. I'm still not quite understanding why the motor was dead without any error light showing in the display earlier. I would think the light would be more predictable.

This is hardly the confirmed long term proof of the issue, but based on its behavior before, it should have shut down during the time of this ride. And the fact that it periodically goes back to killing the motor when you plug both back in seems to indicate some kind of "signal bleed" or such perhaps like you suggest. The good thing is that I can ride with one brake switch which is what I was doing originally. This didn't start until I hooked up both brake switches. I only really need the rear lever switch. Obviously I seriously brake with the front brake for the hard stuff, but I always engage the rear a little or a lot depending on the situation. I may do the voltage check you mention just for more info. Thanks for this insight. You may have hit it.

BTW, my observation on the watt meter bar graph on another thread was apparently incorrect. I've been riding on a trail for the most part and apparently didn't focus on the display as much as I have while doing the diagnosis of this brake switch issue. Also I now notice that bar graph doesn't really "jump up there" while pedaling as much as it does on the throttle. The few times I punched the throttle on a trail, I definitely kept my eyes on the trail...LOL! The bar graph seems to be working as advertised.

 

[Gionnirocket]

Did you happen to notice with only one plugged in if you used the same dongle of the main harness?
Even bench testing might not be crystal clear as it may take some vibration to recreate the issue... But it'll still be more information.
Anywho.. Probably not the controller as it stands now and I'd try swapping brake levers with main harness dongles and if the bench test looks good I'd try a new main harness. If it's a new kit, maybe the seller can send it to you under warranty.

On the bar graph... I'm not familiar with yours in particular, but that's good that it's working. Mine has a yellow circular pattern that is pretty easy to see even when drawing under 100w...and yes a throttle slam will light it up fully.

 

[TNC]

Good grief, man...you're reading my mind...LOL! I'm going for a ride today later, and I was going to try switching the dongles to check that very thing. In the house last night, that very thought crossed my mind about whether the problem was mainly in one cable or the other, because I realized I didn't pay attention to which cable I used during experimentation. They're both fastened close together with a velcro tie, so I didn't notice. I will today. I decided to use some nifty velcro ties instead of the more traditional nylon zips, because I wasn't sure how robust these cables are. They look quite good, but...

Yes, the kit is new...last month from Bafang USA. I tried to get the whole kit from Luna and checked a couple of other sources, but the current bike component rage didn't allow that. I got a lot of the components from Luna, but I doubt getting the motor and basic essentials from BUSA is a problem...correct? The good news is that there is apparently no issue with the motor, controller, and display, and I think there's where the heartburn would have existed. You've been a big help...thanks.