I "cheated" since I don't need the resolution of the two LSBs. I just 
set them both to 0 and don't write them. That greatly simplifies my 
code, as I only need to set CCPR1L.

I have a routine for the user to set the value of CCPR1L, and all is 
good. I go from full off (actually, I set the lower limit to '5' to 
prevent a full-off backlight) to full on, and 250 steps between.

I did have a minor 'trip-up' since PWM1 is port C,2 and PWM2 is C,1. 
But, I quickly found that.

I also had to 'invert' my settings since I wanted 0 to be full off and 
255 full on. My interface inverts the on/off state. Again, minor issue.

I was thinking/making PWM much more difficult than it is. But, when you 
don't know squat about it, even the simple things help.

Now, on to the fan part of the code! I already have a step up on that, 
but my PWM output may make the counting a challenge if the timers are 
interdependent.

Joe M.

ivp wrote:
> Just one little point
> 
>          movlw   d_lo            ;put duty cycle bits 1-0
>          swapf   wreg
>          iorwf   ccp1con         ;into bits 5-4 of CCP1CON
> 
> will not put the bits into 'fresh' locations
> 
>          movlw b'11001111'
>          andwf ccp1con
> 
> clears ccp1con <5:4> before including the duty cycle bits as above
> 
> wbr
> 
-- 
http://www.piclist.com PIC/SX FAQ & list archive
View/change your membership options at
http://mailman.mit.edu/mailman/listinfo/piclist