 |
 |
| View previous topic :: View next topic |
| Author |
Message |
reddyysnr
Joined: 15 Apr 2013
Posts: 6
Location: India
|
| PWM spikes |
 Posted: Sun Sep 15, 2013 11:50 pm |
 |
|
Hi,
I am trying to generate PWM on 16F690. This is the code below.
| Code: |
#include <16F690.h>
#FUSES NOWDT, HS, NOPROTECT, BROWNOUT, MCLR, NOCPD, NOPUT, NOIESO, NOFCMEN
#use delay(clock=40000000)
void main()
{
setup_timer_2(T2_DIV_BY_1,250,1);
setup_ccp1(CCP_PWM);
set_pwm1_duty(0); /// 0 - 255
}
|
I have set the duty cycle to zero. I tried simulating this using Proteus. It shows spikes now and then. Is there an error in my code or is it a bug of Proteus?
Here is a Screenshot of Proteus |
|
 |
Ttelmah
Joined: 11 Mar 2010
Posts: 19496
|
|
| Posted: Mon Sep 16, 2013 12:39 am |
|
|
With the code as posted, the chip will actually be asleep, and the oscillator stopped. No anything....
I'd say Proteus has problems with this, and keeps trying to reawaken the chip.
Proteus also does not check other things for being 'possible'. What is the maximum clock rate that HS supports?. What is the maximum clock rate the chip supports?. OS01 in table 17-1. Is 40MHz ever going to work?.
Then in a real chip, you need:
1) MCLR pulled up (since you have it enabled).
2) Decoupling close to the chip.
3) You really should not leave the other pins floating. Either drive them (set as output), or have a pullup resistor on them. Floating pins _will_ lead to erratic power consumption and behaviour - again Proteus misses this.
4) Clock. For HS, you need a crystal, and loading capacitors. Proteus will 'believe' you if you say you have a 40MHz crystal, and it is working, even if the real chip won't....
This is why you will learn a lot more especially about problems, using a real chip.
| Code: |
#include <16F690.h>
#FUSES NOWDT, HS, NOPROTECT, BROWNOUT, MCLR, NOCPD, NOPUT, NOIESO, NOFCMEN
#use delay(clock=40000000)
void main()
{
setup_timer_2(T2_DIV_BY_1,250,1);
setup_ccp1(CCP_PWM_L_L); //Note the change
set_pwm1_duty(0); /// 0 - 255
while (TRUE); //stop the chip 'dropping off the end'....
}
|
The other thing is that using an eight bit value to set the duty, leaves two bits of the PWM 'unprogrammed'. These can be set in the setup_CCP1, using the syntax I show. Without this, the actual result can be 'indeterminate'. However on the real chip, given it'd be asleep, you would see nothing. However if you keep the chip awake (while etc..), you might see continuous steady pulses, depending what is in these bits. They are meant to be zero on boot, but I have in the past seen these containing values.
Best Wishes |
|
|
artohautala
Joined: 17 Nov 2011
Posts: 187
|
| PWM works fine now |
| Posted: Mon Nov 25, 2013 7:51 am |
|
|
Lot of thanks Ttelmah
I'm using this setup now and everything is OK
My clock frequency is
| Code: |
#use delay(clock= 11059200)
//init PWM1
setup_timer_2(T2_DIV_BY_16,255,16);
setup_ccp1(CCP_PWM);
//set_pwm1_duty(100); /// 0 - 255
|
I measured PWM pulse in practice with oscilloscope frequency is 675 Hz
period is 1.48 ms.
Ontime is 0.57 ms and offtime is 0.91 ms.
all the best
  |
|
|
|
|
You cannot post new topics in this forum
You cannot reply to topics in this forum
You cannot edit your posts in this forum
You cannot delete your posts in this forum
You cannot vote in polls in this forum
|
Powered by phpBB © 2001, 2005 phpBB Group
|
FAQ 
Search
Register
Profile 
Log in to check your private messages 
Log in
