Power cutting out on BBSHD

[trevor brooker]

today the power cut out on mile 10 of 15. On return journey sometimes it only lasted seconds. Suggestions please to allow me to identify the problem.
On the way back I identified the problem could be
1- Battery (cells or BMS)
2-wiring from battery to controller
3-Controller
4-wiring from controller to display/handlebar switch.
I then tried to work out what was causing the power loss.
Potholes/bumps in the road - no problems, it even cut off on smooth tarmac just when I was going uphill & stopping on the path to remove a bramble at head height. So I'm guessing not a loose connection in the wiring.
Next guess was battery, but after the power cut off the handlebar switch would not turn it back on, until I had flipped the battery switch.
If it was battery cells, then would they be able to sustain power just a few seconds later when I switched the battery on/off.
So I'm guessing turning the power on/off is resetting the computer in either the BMS or Controller.
I then tried turning the power level to 1 & riding faster than the assistance speed - result no power loss until I came to the next hill & Watts started being shown again on the display.
Total Watts per mile used was similar to average for that journey.
I'm coming down to the idea that I need to replace the Controller, but before I do so have I missed anything obvious?

 

[peter.c]

What voltage doe the battery show on the display ?
Do you have a multi meter to check the volts on a the battery when fully charged ,and then recheck when it cuts out

 

[trevor brooker]

when I left it was 58.8v it had dropped to 54.5v after 24.5miles.
I was watching the display when the power cut off, it showed the same voltage before & after I switched it back on, no voltage sag.

 

[peter.c]

Sounds more like a battery problem to me, if the battery switches off and cuts all power
But switching on and off clears the problem could be the bms is anything getting warm? how old is the battery and what volts does it normally cut out at when flat

 

[trevor brooker]

battery purchased January 2018. Number of charge cycles (counting four charges from 75% to 100% as I cycle) is 59.
Battery 52v17.5ah using Sanyo NCR18650GA 3.7v 3450mAh cells 14S5P & 30A BMS Note March 2018 battery discharged to 43.9v in error
since then its been charging to either 90% or 100%, before it stops charging.
I have not felt the battery to see if it got hot, but the range was normal so I assume that power not lost as heat.
I normally charge mid afternoon & ride it the next morning. Battery voltage does not change overnight.
Normal usage is 30/40% so I charge again before using. The controller LVC is set to 43v. The lowest I have gone since March 2018 is 48.3v, so it has never cut out.

 

[peter.c]

Hard to long distance fault find you seem to have covered all the obvious usual suspects
Have you recharged since the cut out

 

[trevor brooker]

Not yet, I was going to try that tomorrow.
I was hoping to limit the things to check out

 

[Nealh]

Check operation of brake lever if cut outs are fitted/used.

 

[trevor brooker]

Brake cut outs on both brakes, when I was coasting home & used the brakes the display showed the brake symbol correctly, but I will unplug & refit & check that display shows they are working

 

[vfr400]

If the LCD went off when the cut-out occurred, the battery supply to it has been interrupted.

If you have to operate the battery switch to get it back on, then the BMS switched off the power because it detected a fault.

 

[trevor brooker]

Ok - yes the display went off & would only turn back on when I turned the battery off/on & then the handlebar on switch.
But does not the battery wiring go into the controller then the display, so I'm no further forward in narrowing down the fault

 

[vfr400]

The display takes a feed directly from where the battery wires go into the controller, so the controller has no effect on them. Normally, your controller could be completely smoked, but the LCD would still work.

Charge up the battery and measure its voltage hot off the charger.

 

[trevor brooker]

Ok- thanks that is something I never knew.

 

[trevor brooker]

Well this morning I thought to check if the fault had cleared itself overnight, before disturbing anything.
Voltage - same as last night
Brake cut offs work & shown on display
So took it for a ride, throttle only, to maximise load, got 100 feet before power cut off.
Returned home, switch on handlebar had no effect.
Took battery off (with battery switch still on) & used voltage meter on output terminals - no voltage
Flipped battery on/off switch - full voltage now showing
So as per vfr400 I can now be certain that its not a controller fault
So battery fault, with the BMS shutting done to prevent a problem.
If my logic is wrong would someone please say, otherwise I'm looking at replacing the battery
thanks

 

[peter.c]

That's a bms fault you need to check the volts when fully charged
Can you open the battery pack and take a picture might help and check the voltage of each of the 5 rows see if one string is lower than the others

 

[Nealh]

Rule out a switch problem first as it could be an easy fix, check solder contacts and switch continuation with the dvm whilst switching to see if there is an intermittent fault with said switch.

 

[trevor brooker]

I checked the battery using the probe on the sense black wire & each white wires.
So cumulative voltage
3.9v 3.9v
7.8v 3.9v
11.7v 3.9v
15.6v 3.9v
19.5v 3.9v
23.3v 3.8v
27.2v 3.9v
31.1v 3.9v
35.0v 3.9v
38.9v 3.9v
42.8v 3.9v
46.7v 3.9v
50.6v 3.9v
54.4v 3.8v

So difference between cell groups is a maximum of 0.1v for two groups

 

[trevor brooker]

Rule out a switch problem first as it could be an easy fix, check solder contacts and switch continuation with the dvm whilst switching to see if there is an intermittent fault with said switch.

Solder contacts feel firm
where should I place the meter probes, on the thick black & yellow wires as the solder joints are exposed on the BMS & flip the on/off switch to see if the voltage changes?

 

[Nealh]

Cell groups are near perfick not much issue there with imbalance unless that last one collapses, try and apply a load and monitor it to see what happens but would have expect a greater voltage variation of maybe more then 0.3v.

Look at the actual switch it self place your probes on the actual switch contacts, in continuation mode toggle the switch on/off several times to see if any fault occurs. A beep should be audible each time the switch is on, if need be try wiggling the switch wires.

Controller wise you need to feel if it gets hot when the cut outs occur on the road, a hot controller means the fets not coping with the current and getting to hot. They will temporary short/thermal cut out until they cool then work again before again shorting.
Check the ohm value of the fets to see if any one is bad.
Using 200 ohm scale test V- to each phase wire all readings should be the same (9k to 15k is normalish) also V+ to each phase reading may just be infinity so 1 if 0 no good. The important bit is each set of three should read the same but each set will have a different value.

 

[Nealh]

An outside chance of a fault is the actual cell contact with nickel.
Peeling back the red protective cell covering to look at the connections to see if any welds have separated, not always evident use a pointed wooden kebab skewer to place gentle lift pressure between cell and nickel. Any poor contact will cause power shut down as a cell/cells in a group can't cope.