PWM problem CCP2 port

[Darrel Taylor]

NL2TTL,

Seems everyone is forgetting the analog ports these days.

ADCON1 = $0F

Added: It will probably be better to have subroutines that set the dutycycle of the CCP modules. Then call them only when needed.
Continually setting the registers in the main loop may cause some problems, and isn't needed since the hardware will continue putting out the PWM while the main program is doing other things.

Also, by using these statements Duty3 = Gval *2, it effectively reduces the PWM resolution to 9-bit since bit0 will always be 0.

Added2: The RB3/CCP2 pin is the alternate CCP2 output. You'll need to change the CCPMX configuration fuse for it to work there.
<br>

[NL2TTL]

@Darrel Taylor

Thank you for you replay, about adcon and the Rb3 i understand, but what exactly do you mean about:

Also, by using these statements Duty3 = Gval *2, it effectively reduces the PWM resolution to 9-bit since bit0 will always be 0

How can i fix that?

[Darrel Taylor]

In the binary world, multiplying by 2 is the same as shift left 1 (<< 1).

This will always leave a 0 it the bit0 location. So if you had a dutycycle of say 257 (100000001), * 2 would be 514 (1000000010).

The easiest way to fix it is to just use bigger numbers. 10-bit numbers range from 0 to 1023. <br>

[NL2TTL]

I've added:

ADCON1 = $0F

But still ccp2 doesn;t work (also on portc.2 it gives problems)

The pin is always high.

[NL2TTL]

And correct me if i'm wrong but when i don't make a loop in the program i get no response on any pwm port

So i need the Goto mainloop

[Darrel Taylor]

Did you also change the CCPMX configuration fuse for RB3/CCP2 alternate?

portc.2 is CCP1, are you saying that that one stopped working ?

And yes, you need to have a mainloop. Otherwise it goes to the END statement and enters Sleep mode. But changing the CCP registers doesn't need to be in the mainloop.
<br>

[NL2TTL]

it's working thank you very much!