Sometime ago I got rid of the front wheel and NiMH battery. I moved the battery circuit boards to a protective enclosure and connected LiCoMn tool batteries to provide 36V21AH. Those batteries are charged off the bike on Makita rapid charger. When it worked, it was an amazing ride, but it was plagued by seemingly random #1(display/battery) and #2(display/motor) errors. In trying to track down possible cable or display problems, the display stopped turning on at all. I gave up.
I decided to take one more try. Fortunately, I had a 2nd top battery circuit board. That board fixed the display not turning on. I rebuilt the housing for the battery board so no connection could possibly loosen and nothing physically touching any part of the boards, except the mount screws. I replace the cables between the display, motor and battery boards with a better shielded cable.
I don't want to jinks it, but it seems to be fixed.
A couple of things unique to using batteries not charged through E+'s charge board circuit. The replacement upper charge board came from a bike that had the NiMH batteries charged normally. I showed 80% SOC and 4 bars, even though the batteries mounted were full at 41 volts. After about 8 miles, no bars even though the 21AH battery had discharged less than 25%. On subsequent rides, still no bars. Apparently, the E+ gets it's battery capacity data from the actually AH used to charge the battery. So if the battery is not charged through the battery boards the battery state of charge display doesn't work. Fortunately everything else works and a simple watt meter is a good substitute.
One other issue that may or may not prove to be a problem. In the original E+ design, the battery board always had voltage (battery connected). With my system, the battery is removed for charging and off the bike until it is next used. I seems to me the delay between the display lighting and power to the wheel is longer than I remember with the front hub battery. I have also had at least 1 start up error but unplugging and re-inserting the com cable on the display fixes that the first time.
of course you are aware of 10 min. timeout.
Powell I haven't heard about that. Is it a delay after LVC or temp problem or something else?