 |
 |
| View previous topic :: View next topic |
| Author |
Message |
tnay
Joined: 11 Feb 2007
Posts: 4
|
| Extremely weird CCP behavior |
 Posted: Wed May 20, 2009 10:40 am |
 |
|
Hello everyone,
I have been troubleshooting this problem for hours. I have encounter a very unique problem here. I try to shrink down the code and problem as much as possible. My CCS version is 3.219.
Here's what I need.
I want to call CCP compare on certain given timing, and change pin status. Whenever a match occured, it would first change the pin status, then a value is assigned to CCP_1 for next match. As of this discussion, which pin status to change is not important, but I like to bring your attention to the timing. Say I have 3 value, so generally it goes like this:
First match = value A
Second match = value B
Third match = value C
Repeat.
Here is what happen.
This particular code work for most A, B and C combination, limited to 16-bit. However, there are several occasion where it does not work. For example:
When A = 200, B = 3130;
When A = 250, B = 2363;
Instead of giving me a perfect match (+/- 1), it consistently give out B+60, become like this:
A = 200, B = 3190;
A = 250, B = 2423;
It will only happen when the above value is in this particular sequence and order. I can't figure out any relation between them. The problem seems to only affect PIC18, and not PIC16.
Here is the code that didn't work
| Code: |
#int_CCP1
CCP1_isr()
{
if(set)
{
output_C(temp);
CCP_1 = PWM[count];
temp = 0x08;
count++;
if (count == 2)
set = 0;
}
else
{
output_C(temp);
CCP_1 = 10000 - PWM[1] - PWM[2];
temp = 0x00;
set = 1;
count = 0;
}
}
void main()
{
set_tris_c(0x00);
output_C(0x00);
setup_timer_1(T1_INTERNAL|T1_DIV_BY_1); // timer1 enabled, instruction clock /8
setup_ccp1(CCP_COMPARE_RESET_TIMER); // set CCP1 to cause an interrupt on match
enable_interrupts(INT_CCP1); // unmask compare1 match interrupt
enable_interrupts(global); // enable all unmasked interrupts
while(1)
;
}
|
But funny enough, the following code works, every time. the different is that I bring up the temp variable before calling CCP.
| Code: | #int_CCP1
CCP1_isr()
{
if(set)
{
temp = 0x00;
output_C(temp);
CCP_1 = PWM[count];
count++;
if (count == 2)
set = 0;
}
else
{
temp = 0x08;
output_C(temp);
CCP_1 = 10000 - PWM[1] - PWM[2];
set = 1;
count = 0;
}
}
void main()
{
set_tris_c(0x00);
output_C(0x00);
setup_timer_1(T1_INTERNAL|T1_DIV_BY_1); // timer1 enabled, instruction clock /8
setup_ccp1(CCP_COMPARE_RESET_TIMER); // set CCP1 to cause an interrupt on match
enable_interrupts(INT_CCP1); // unmask compare1 match interrupt
enable_interrupts(global); // enable all unmasked interrupts
while(1)
;
} |
|
|
 |
PCM programmer
Joined: 06 Sep 2003
Posts: 21708
|
|
| Posted: Wed May 20, 2009 11:08 am |
|
|
Post a complete test program. Add the lines that are missing, such
as the #include for the PIC, #fuses, #use delay(), and all variable
declarations. |
|
|
Guest
|
|
| Posted: Thu May 21, 2009 4:08 am |
|
|
Sorry for that, I exclude those with the intention of reducing the code length. Totally forget about how others are to run it without the heading.
Anyway, here goes:
| Code: | #include <18FXXXX.h>
#fuses HS,NOWDT,NOLVP
#use delay(clock=10000000)
long PWM[2] = {250, 2363};
int temp;
int set = 0;
int count = 0; |
Please replace the 18FXXXX with any PIC18F. I tried a few, neither of them don't work. You can also try the 16FXXXX, it works on them though.
The delay is not the focus of this discussion, but the instruction cycle. So any clock speed with do.
Thanks.[/code] |
|
|
tnay
Joined: 11 Feb 2007
Posts: 4
|
|
| Posted: Thu May 21, 2009 4:12 am |
|
|
Opps. That was me. Forget to sign-in   |
|
|
PCM programmer
Joined: 06 Sep 2003
Posts: 21708
|
|
| Posted: Thu May 21, 2009 10:28 am |
|
|
I can't test it without a specific PIC. Post a PIC that it fails with,
and post one that it works with.
Is this a Proteus project ? |
|
|
tnay
Joined: 11 Feb 2007
Posts: 4
|
|
| Posted: Sat May 23, 2009 3:41 am |
|
|
Try any PIC16F and PIC18F that support CCP. I'm using PIC16F877A and PIC18F458 here. I'm not familiar with Proteus.
I tested the timing using MPLAB SIM. Tested on my prototype circuit, and uses another PIC to capture the timing. It consistently happen around that particular value (+/- 3). Any value of outside that range, be it smaller or larger value, it just disappear. |
|
|
|
|
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
