>Shawn wrote:
>
>
>>they mention a CCP1RIL bit.  But I'm looking at the datasheet for the
>>16f627 that they are basing the design off of and I can't find the bit
>>anywhere...
>>
>>
>
>I think they meant the CCPR1L byte - changing that between zero and the
>calculated 50% duty cycle value would give you glitchless on/off modulation
>of the output.
>        Jason Harper
>
>--
>http://www.piclist.com hint: PICList Posts must start with ONE topic:
>[PIC]:,[SX]:,[AVR]: ->uP ONLY! [EE]:,[OT]: ->Other [BUY]:,[AD]: ->Ads
>
>

>
> Note that each "bit" wil be a specific number of periods of the
> PWM signal. So you can just as well count the number of PWM
> periods output and, when the correct number is reached, just
> shut of the PWM output.
>
> This can be done easily by using the "overflow interrupt" of the
> timer controling the PWM module. In the ISR you then just
> decrement a counter and, when it reach zero, shut the PWM output
> off.

> Using this method it's also easy to get clean on/off so you
> always gets whole and clean PWM pulses. You could either also
> control the "on" of the PWM in the ISR, or just monitor the
> "timer interrupt flag" somewhare else in your code.
>
> I used this concept in an IR sender using the PWM module to
> produce the 38Khz carrier, and using sequences of PWM
> pulses of different length to idicate "start", "1" and "0" bits.
> Looks very much the same as this application...
>
> Jan-Erik.
>
> --
> http://www.piclist.com hint: PICList Posts must start with ONE topic:
> [PIC]:,[SX]:,[AVR]: ->uP ONLY! [EE]:,[OT]: ->Other [BUY]:,[AD]: ->Ads

>
> >
> >
> >>If that's the case,  then how would you go about doing that?
> >>Would this require a second timer that runs at some "baud rate"
> >>that pulls data from a  register and sends it out the pwm?
> >
> >
> > You don't have to tie up a hole counter for this, just
> > another software counter.
>
> A counter is a counter though.

> Shawn Wilton wrote :
>
>
>>>First you must have a method to put the PWM output
>>>on/off. The AP note uses the TRIS bit of RB4 for this.
>>>Probably OK.
>>
>>You think the tristate should be flipped eh?  Why not just
>>use the duty cycle as a switch? You have a preference I take it?
>
>
> No particular preference, I'd just use what's easiest.
> If you'd use the duty cycle, you have to "remember" what it was set
> to until it's time to start the PWM again. The TRIS bit are just a
> single bcf/bsf instruction. Faster and less code.
>
>
>>>
>>>>If that's the case,  then how would you go about doing that?
>>>>Would this require a second timer that runs at some "baud rate"
>>>>that pulls data from a  register and sends it out the pwm?
>>>
>>>
>>>You don't have to tie up a hole counter for this, just
>>>another software counter.
>>
>>A counter is a counter though.
>
>
> Sorry, I should have written "tie up a hole timer"...
> I read "a second timer" in your text as a "hardware timer",
> which isn't needed, as I wrote.
>
> Best Regards,
> Jan-Erik.
>
> --
> http://www.piclist.com hint: PICList Posts must start with ONE topic:
> [PIC]:,[SX]:,[AVR]: ->uP ONLY! [EE]:,[OT]: ->Other [BUY]:,[AD]: ->Ads