Re: [Paparazzi-devel] Preventing Servos from Cycling at Boot

[antoine drouin]

Hi Luke

On the twog, I believe sw/airborne/arch/lpc21/servos_4017_hw.c is used.
On initialization, it starts by settings all servo to 1.5ms
It has a couple hard coded special cases when  SERVO_MOTOR,
SERVO_MOTOR_LEFT, SERVO_MOTOR_RIGHT or SERVO_HATCH is defined. It then
outputs the specific SERVO_MOTOR_NEUTRAL etc... values.

On stm32 which uses
./sw/airborne/firmwares/rotorcraft/actuators/arch/stm32/actuators_pwm_arch.c
, the behaviour is different and I believe servos are not commanded
(no PWM is output)  until the autopilot runs and makes the first call
to  actuators_commit

I'm not sure what the right thing to do is, as actuators are fed with
commands produced by the autopilot. There is nevertheless a failsafe
value defined for those commands ( it is used on failure of the
autopilot process - which only make sense on systems that have a level
of protection between the autopilot and fbw processes). Maybe a thing
to do would be to command servos using values computed from the
failsafe commands, but that would mean those values would have to be
computed before the actuators hardware initialization.

hth

Poine


On Sat, May 21, 2011 at 12:11 AM, Luke Ionno <address@hidden> wrote:
> Question for everybody...
>
> I’ve got a TWOG-based airframe with a servo-operated release mechanism; it’s
> a little tricky to reset, and unfortunately, at boot, all the servos cycle,
> causing the release to trigger.  Is there any way to prevent the initial
> servo cycling?
>
> Thanks!
>
> ~Luke
> _______________________________________________
> Paparazzi-devel mailing list
> address@hidden
> https://lists.nongnu.org/mailman/listinfo/paparazzi-devel
>
>