> -----Original Message-----
> From: michael brown [SMTP:spam-me@HOUSTON.RR.COM]
> Sent: Wednesday, February 05, 2003 4:18 PM
> To:   PICLIST@MITVMA.MIT.EDU
> Subject:      Questions about PWM using TMR2? and CCP
>
> Hello all,
>
> I'm posting using OE today as I can't run MPLAB in Linux yet, please
> don't send me any quoted-printable stuff.  ;-)
>
> I'm trying to figure out the hardware PWM feature of the 16F628.  My
> John Peatman book leaves a little to be desired here.  :-/  Fortunately
> the datasheet actually seems more clear about things in this instance.
> I would like to use this feature to control an ordinary R/C servo motor
> (1 - 2ms width) approx 50 times/sec (20ms period?).  It appears that my
> using a 4Mhz crystal may preclude the ability to do this.  It looks like
> the longest period I can come up with is 244hz (about 4ms?) allowing for
> a full but unnecessary (for me) 10 bits worth of PW precision. ;-)
>
> Questions:
> 1) Am I getting this stuff right?
>
Looks good so far!

> 2) Would it be a bad thing to try and drive the servo using a 4ms period
> since that is still longer than my longest pulse width requirement?
>
Possibly.  Servos are generaly very tolerant of the PWM period.  However,
this is quite a long way from 20ms and not all servos are the same, so it
would be worth trying this for the particular servo you wish to use.

> 3) Since (AIUI) the PWM function uses TMR2 (though all other
> capture/compare modes use TMR1), is there no way to use the postscaler
> to increase the period?
>
Unfortunately the post-scaler only affects the rate at which the TMR2IF flag
is set.  The only way to go slower after you have set 1:16 pre-scale and
PR2=255 is to use a slower clock.

> 4) If not, is there any way to use TMR1 instead
>
Not through the hardware, it could be used as a timebase for a software
solution though.

> I would like to do this with the hardware automatically toggling the
> line at the correct rate, without having to handle anything in
> interrupts.
>
I know what you mean, I prefer to have proper hardware peripherals I can set
and forget rather than software buring up my CPU cycles.  However, in the
case of a servo control, the PWM period is so slow that an interrupt driven
solution would have negligible overhead.  There are also lots of code
samples on the net.

Regards

Mike

--
http://www.piclist.com hint: To leave the PICList
mailto:piclist-unsubscribe-request@mitvma.mit.edu