I read this post in a another forum where a guy claims 50Hz is insufficient PWM cycle for accomplishing quadcopter stability. Is there a truth in this ? What is the cycle of the ArduPilot PWM outputs that control the motor ESCs?
There's a lot of "mine's faster than your's" bluster online about this, mostly by people who haven't really done the analysis. But ArduCopter drives the stabilization loop at 100Hz, which is as fast as most commercial ESCs can go. We also have code for I2C ESCs, which can go faster, although we don't see good reasons why that is necessary yet.
Don't confuse PWM for ESCs with PWM for servos on planes. Regular servos can only go 50hz, so fixed-wing autopilots usually run their inner loop at that speed.
I guess the idea would be that a shorter PWM period would have ESCs react quicker to a value change.
One thing is the stabilisation loop freq, if the PWM cycle would still be say 50 Hz could this affect response time?. Well too late at night for me to think clear anyway.
50Hz response is too slow for a quad (because it can correct it's attitude only in 50Hz)
Arducopter uses 200Hz (not 100Hz), I got a more stable flight with I2C ESCs (quicker reaction on command change, but at 200Hz, too)