The fact you havent just said fuck it, im selling these anyway is admirable.
9 Likes
i am looking for 100% reliability.
for the watchdog reset bug, i guess the problem might be the oscillation of current loop. cause the sampled current has too many ripples.
7 Likes
Wait, which firmware version are you guys running? I don’t think the older firmware versions have this problem. I’m running 3.65 on my flipsky ESCs. I can give you guys the vesc_tool 1.29 executable.
The firmware has gotten really bloated, including inside the all important foc update interrupt, since versions 4 and 5 and I suspect this is contributing to the problem. If you don’t understand what I mean, sample in v0 and v7 at 25khz doesn’t work anymore.
2 Likes
The fact that this doesn’t work with the official VESC firmware is the only thing stopping me from using this on my upcoming 13a build.
I guess it’s time to spin a custom, better performing firmware? :P.
Setup doesn’t have to be painful.
5 Likes
fw5.1. this version works well on my v7h.
2 Likes
I’d recommend trying it on older firmware to see if you are able to reproduce the issue.
5 Likes
I’d recommend sending one to @Gamer43
Yeah has to be 5.1 for hfi using it on a focbox v1.6 no problems
2 Likes
While I would greatly appreciate such a gesture, I don’t have the time to combo through the codebase to figure out why the watchdog isn’t being tickled, and what hardware situation causes code to hang in such a way.
3 Likes
Ok nevermind, quick search reveals that a timer thread in main, the canbus read thread, and the FOC update thread need to tickle the watchdog within 12ms.
FOC has highest priority and will always tickle the watchdog.
So something is causing a hang up that prevents the lowest priority of those three threads from executing once in 12ms.
4 Likes
In a perfect world that would have been a progressive step, I like the ESC and the attitude. Ps: can your A/S do 18s?
We’re wishing you success @zhanzhan
4 Likes
It’s literally a MOSFET swap.
I can probably get to the bottom of this, but then the question is, is the root cause hardware or software?
Since @NuRxG reported this happening on his stormcore.
2 Likes
I had the booting from watchdog reset fault show up the other day on my vesc 6+ on my ebike. I’m hoping it is something software related and not hardware. Running 5.1
1 Like
Sounds like a firmware issue. I’m pretty sure the lowest priority thread is not getting any scheduling time from the chibiOS scheduler.
I tried running an encoder with the vesc6 mk3 a while back and it completely stalled because the MCU was missing the index interrupt when the FOC loop was running. Ended up missing enough pulses to offset by 90°.
For me, it was only happening on hard braking.
At high speed or low speed?
It’s been happening to me anytime I apply throttle after coasting. Today on my ride (after resetting FW to 5.1) it started happening every 30 or so feet an hour into my ride. I ended up riding home on one motor.
1 Like
I have had similar issues with watchdog resets. Different HW. Cant remember what fm i used at the time. Could have been 5.1 or latest 4.xx
It happened always to the same esc in my dual setup. About one time every 15min at worst. Randomly.
As an attempt to fault trace I switched positions of the two escs and the one which previously had the vx1 receicer now has the NRF51822 and vice versa. I might have routed the wires for the receiver slightly differently as well.
Never had a watchdog reset since.
Assume my case was related to disturbances 
1 Like
Low speed.
Sounds like the watchdog timeout is triggering during a sensor transition then
Which sensors are you running?
1 Like