I’m having studdering at light accelerations from stand still, like if the sensors were not synced correctly, this might be the same thing you are feeling at light inclines.
I don’t think it’s the sensors because at that point it’s running pure sensorless because it’s way past the sensorless erpm
Got the fault, the dreaded overcurrent fault again 
Fault : FAULT_CODE_ABS_OVER_CURRENT Current : 318.3 Current filtered : 153.8 Voltage : 24.93 Duty : 0.950 RPM : 27241.8 Tacho : 16706 Cycles running : 70844 TIM duty : 7980 TIM val samp : 2 TIM current samp : 4200 TIM top : 8400 Comm step : 0 Temperature : 29.52
Going to try using “Stator Saturation Compensation” as that sounds like it might help
That’s a crazy amount of current it read…
Yea thats pretty much been all my OC faults… still unsure as to why, but always related to a sudden increase in acceleration.
welp that didn’t work, fuck it i’m going back to BLDC
(╯°□°）╯︵ ┻━┻
2 Likes
After seeing this message in the vesc-tool I decided to try doubling my Observer Gain and… my problems might be fixed!! I haven’t gotten OC faults either on the bench or while riding, but only time will tell!
EDIT: Confirmed, this fixed my overcurrent problems! Doubling it worked great, halving it also worked, however caused some weird acceleration behavior intermittently when doing full throttle from standstill.
2 Likes
What the hell does that setting do…why are 2x or .5 suggested alternate values…lol Itll be crazy if thats what’s causing the myriad of oc faults
Yea I tried googling it, but it just made me feel dumb. So far doubling has produced no OC faults, but… it might’ve gotten louder? Hard to tell.
I hate this blindly changing settings to fix one issue not knowing what other issues it could cause. Like I expect “detected values” to be correct…if not, why not
I think FOC is too complex to have a “catch all solution” especially with the wide variety of motors, but perhaps there could be something set in place that tests the motors are varying currents and duty to see if there are any faults and/or cogging.
I think some kind of understanding of what values a specific motor should have to be able to know if detected values are correct would help. Also I think anybody selling esk8 motors should provide a datasheet with “expected, valid results with certain battery configurations” that would help immensely
1 Like
I agree, but the esk8 scene (especially DIY esk8 scene) isn’t mature enough yet. I mean, we only have 1 semi-dedicated ESC right now. Hopefully the scene will evolve over the next couple years and give us more options, standardization, competition, and stability.
3 Likes
It’s been good! I had a wipe out that gave me a concussion, but that turned out to be a mixture of a shorted motor and a bad VESC. After replacing both motor and VESC, I’ve been cautious while riding, but haven’t encountered anything weird yet and no faults have been thrown.
The official Vedder f/w 3.3 has this stated in the change log.
=== FW 3.30 ===
- Activated iterative observer for better operation at high ERPM.
- Check for NAN and truncate some FOC variables.
- Speed controller windup protection improvement
Not sure if this means that I no longer have to double the observer gain, but I’ll hold off on updating for now.
I heared about your accident 
Hope it is all coming together well. 
I am using 2.54 FW in watt mode and BLDC now because I had problems with sensored FOC a while back. I found your thread recently and decided to revisit the FOC for the quiteness mainly 
I am having the same issue that in BLDC the motor starts fine but stutters in FOC at low speeds. I am going to try sensorless but I think I did that already with no change so I dont blame the sensors for now. I will try messing with the observer gain and see what happens now that I know it has worked for you!
Just FYI I had issues with FOC in 2.xx as well where it would stutter (at high RPM though), but FOC runs a LOT better in 3.xx f/w