Why PWM frequency generated with CCS not stable?

[Somkiat]

I start my project with PICmicro and CCS C Compiler.I wrote my program to generate PWM signal but when I use Oscilloscope measure PWM signal the frequency not stable and period not constant why ?

[SherpaDoug]

Is your processor clock stable? If it is and you don't change the PWM divisor then the outpus has to be stable too.
_________________
The search for better is endless. Instead simply find very good and get the job done.

[Haplo]

erm...are you using the hardware PWM, or is it a software generated PWM?

[Somkiat]

I used hardware to genterate PWM and my clock was stable.
my PWM on OSC :


_____|'''''''''''|___|'''''''''''|___|''''|______|'''''''''''|___|'''''''''''|_______
t1 t2 t1 t2 tx t1 t2 t1 t2

Why tx<>t2 ?


and my code:

//========================
#include <16F877A.h>
#fuses XT,NOWDT,NOPROTECT,NOLVP
#use delay(clock=4000000)

void main()
{

int duty_cycle;

setup_ccp1(CCP_PWM);
setup_timer_2(T2_DIV_BY_1, 66, 1);
duty_cycle=9;
set_pwm1_duty(duty_cycle);
while(1);
}

PWM frequency = 15.15 khz ?

[rwyoung]

This may not explain your "runt" pulse but the argument for "set_pwm1_duty" is a 16 bit (long) integer. If you give it an 8 bit (int) then the compiler is supposed to shift the value right 2 bits.

How are you measuring the pulse width? I assume you are using an oscilloscope. Be sure it is triggering properly (example: rising edge, 2V) and that you have a good ground connection for the probe. Oscilloscopes (and especially digital ones) are very good liars if you don't pay attention...

By the way, what version of compiler are you using? Once in a while, odd bugs creap into and out of the compiler so knowing what version you are using may help us diagnoze your problem.
_________________
Rob Young
The Screw-Up Fairy may just visit you but he has crashed on my couch for the last month!

[Guest]

Try


set_pwm1_duty((long int)(duty_cycle);

It's called "type casting"