I updated AM32 to 2.20 due to some issues with desync in 2.19 just to test, and seems arcucopter 4.6.3 is unable to connect to ESC properly. FC: MicoAir743v2-AIO-45A
I get ESC connecting beeping (3 tones _ - “) and ESC OK (short beeps ) in an endless loop - https://youtu.be/9n-fRi8SdV8
AM32 2.18 works fine.
Sigh. I don’t currently have my AM32 vehicles to test anymore. I’ll see if Huibean’s ESC show’s the issue …
Did you revert back to 2.18? When you’re on 2.20 can you still access the configurator?
I don’t have any solutions, I just have the same controller on the shelf waiting for a build.
Yes, I can access configurator and revert it back to 2.18.
I flashed AM32 2.20 to some other ESC, and that seems to be working fine, so I guess this is somehow limited to AIO type, or even this one AIO itself. It had some issues, as not long ago it went of with flames and smoke…
Woah! What happened to that board? I’ve ruined my share of parts but that is proper cooked!
I have no idea. It has happened during autotune flight in a bit gusty weather. Surprisingly it was trying to keep the altitude to the very end.
I do even have flight log recovered (large file as gyro data for filtering tests also there) 
Motors seem to be fine, I just replaced wires and they work OK.
Panasonic EEUFR1V471L - 470uF/35V LowESR cap. But… now that You asked… this failed once when there was stock 220uF from micoair and I replaced it with panasonic one (and panasonic cap was intact after this… malfunction).
Current spikes are probably due to pitch/roll autotune moves, but voltage looks good I would say. At first I thought those spikes are voltage, but it is current.
Any ideas for a root cause are welcomed, as I am going to build it again . Would be good to know what potential issue I need to fix/avoid to skip the fireworks and “magic smoke escape” flight mode 
Anyway… there had to be some pre-existing issue in it, as with automatic timing and pmw via RPMs it behaved badly on one motor. Video shows motor test from mission planner.
And I had to disable automatic timing - set to 15deg, and pwm 24-48kHz instead of “by RPM” to eliminate this behavior.
I found the newest am32 firmware needs lots of capacitors.
What do You mean by that? Is there some kind of bug/(feature?) generating more voltage spikes than usual?
I don’t know the technical details, I just know that my bird flew fine on the older firmware, when I updated it it kept crashing. I had a suggestion from a few on this forum to add capacitors and that fixed the issues.
How much capacitance are you using per esc?
AM32 can ramp up the power much faster than BLHeli32. So it’ll find the weak spots in a build.
I was told 100uf per motor should be enough, so a 470 on an AIO should cover it. Might also help to turn down the ramp rate.
By add, You mean add more to existing one, or add where there were none?
I use 470uF as minimum.
I added total 500uf per motor.
Does That depend on where they are installed?
Yes, 100%. In one of my builds I ended up going to 500/motor because I couldn’t get the capacitor right on the ESC. Longer wire runs will increase the problem. In the case of an AIO the cap is pretty much as close as it can get on the board, so lower values will work.
That’s what I was thinking, thanks!
But… has that changed with 2.20? I had no issues with 470uF and 2.18
Any reason why that would became a problem with 2.20?
