One s2740vc loses sync

[CarlOlsson]

We have two s2740vc ESC:s that spins motors, however one has performance issues
Preparation

• Erased flash on both ESC:s
• Flashed the bootloader in 68e34de6_gcc4_8_4_Ben to both ESC:s
• Current consumption of both ESC:s connected to a motor is ~47mA
• Flashed the fw in 68e34de6_gcc4_8_4_Ben to both ESC:s

Logs from the ESC which performs worse:
Setpoint 10, motor first don't start so I press stop then start again

Setpoint 20, motor spins:

Setpoint 30, motor starts and stops:

Setpoint 30, strange sounds:

Sweep, loses sync:

Logs from the ESC which performs good, same motor, same fw:
Setpoint 25

Sweep:

[bendyer]

The second (good) ESC is getting acceptable resistance measurements, but the first is way off, and varies significantly from plot to plot. That's causing sync issues similar to what happens with that motor and the PX4ESC 1.6.

I'm concerned about the sweep plot with the second ESC. Your bus voltage is exceeding the abs max rating of the device during braking—those spikes are well above 27 V for a significant fraction of a second and will be damaging components on the ESC (and even your power supply if it's not rated for 30+ V). If you've run heavy braking tests with the other ESCs, I would say that could easily be a factor in the failure you saw, maybe also in the poor measurements from the first ESC.

If you don't have a beefier supply available, I would suggest using a much smaller motor, or reducing the braking parameter to a value of 0.01–0.1.

Now that I think about it, I should really have the software sense voltage spikes caused by regenerative braking and have it reduce the maximum allowable braking current dynamically.

[LorenzMeier]

@bendyer Sounds like there is no way around fixing the braking, as its unexpected from a user perspective. Is this easy to do?