| View previous topic :: View next topic |
| Author |
Message |
pokiko
Joined: 27 Jul 2007
Posts: 33
|
| tocki and timer1 |
 Posted: Sat May 17, 2008 6:07 am |
 |
|
hi
i use timer0 and timer1 of pic16f877a with an osc of 16Mhz. I have a 50hz input from the RA4/tocki pin as signal input. All i need is to specify a 1oms time resolution from Tİmer1. When i setup_timer_1( T1_INTERNAL | T1_DIV_BY_8 ) then 10us corresponds to 10000 counts where i expect 5000.
So where does this doubly occur. Is there something which scales timer1 count by 2. I looked at datasheet but i couldn't find out anything.
Last edited by pokiko on Sat May 17, 2008 3:57 pm; edited 3 times in total |
|
 |
Indy
Joined: 16 May 2008
Posts: 24
|
|
| Posted: Sat May 17, 2008 8:05 am |
|
|
| What is your compiler version number? |
|
|
pokiko
Joined: 27 Jul 2007
Posts: 33
|
|
| Posted: Sat May 17, 2008 3:54 pm |
|
|
| it is 4.020 |
|
|
ckielstra
Joined: 18 Mar 2004
Posts: 3680
Location: The Netherlands
|
|
| Posted: Sat May 17, 2008 4:06 pm |
|
|
| The first releases of the v4 compiler were of very bad quality and only became more or less usable at version 4.030. I recommend you to upgrade to a newer version or use the older v3.249. |
|
|
pokiko
Joined: 27 Jul 2007
Posts: 33
|
|
| Posted: Sat May 17, 2008 4:26 pm |
|
|
| so what about 4.049? |
|
|
ckielstra
Joined: 18 Mar 2004
Posts: 3680
Location: The Netherlands
|
|
| Posted: Sat May 17, 2008 5:07 pm |
|
|
4.049 is higher than 4.030, a reasonable stable version. Still getting the same bad results?
Can you post a short example program, including all #fuses, etc. So we can copy/paste it into our compiler? |
|
|
pokiko
Joined: 27 Jul 2007
Posts: 33
|
| ok |
| Posted: Sun May 18, 2008 1:45 pm |
|
|
| Code: | #include <16F877A.H>
#device ADC=10
#fuses HS, NOWDT, NOPUT, PROTECT, NODEBUG, NOBROWNOUT, NOLVP, NOCPD, NOWRT
#use delay(clock=16000000)
#define maxPERIOD 9200 //nearly 10000
#define minPERIOD 4000
#int_TIMER0
void timer0_isr()
{
set_timer0(255);
set_timer1(0);
}
#int_CCP1
void pos_trig()
{
output_high(pin_e0);
delay_us(1);
output_low(pin_e0);
}
#int_CCP2
void neg_trig()
{
output_high(pin_e0);
delay_us(1);
output_low(pin_e0);
}
void main()
{
.......
set_adc_channel(0);
delay_us(10);
ampereREAD = read_adc();
if(amperSET>ampereREAD)
{
CCP_1-=10;
if (CCP_1<minPERIOD)
{ CCP_1=minPERIOD;
CCP_2=CCP_1+10000;
}
delay_ms(10);
}
else if(amperSET<ampereREAD)
{
CCP_1+=10;
if(CCP_1>maxPERIOD)
{ CCP_1=maxPERIOD;
CCP_2=CCP_1+10000;
}
delay_ms(10);
}
}
void init_MCU(){
.............
SETUP_TIMER_0(RTCC_EXT_L_TO_H | RTCC_DIV_1);
setup_timer_1( T1_INTERNAL | T1_DIV_BY_8 );
setup_timer_2(T2_DISABLED,0,1);
setup_CCP1( CCP_COMPARE_INT );
setup_CCP2( CCP_COMPARE_INT );
setup_CCP1( CCP_COMPARE_INT );
setup_CCP2( CCP_COMPARE_INT );
} |
Last edited by pokiko on Sun May 18, 2008 4:18 pm; edited 1 time in total |
|
|
ckielstra
Joined: 18 Mar 2004
Posts: 3680
Location: The Netherlands
|
|
| Posted: Sun May 18, 2008 3:01 pm |
|
|
The program is not complete and contains syntax errors (minPERYOD instead of minPERIOD).
How can we see from this program that your Timer1 is running twice as fast? I mean, which variables or what output should we check?
Please post a small (max. 1 page) and complete program. The program should contain all #fuses commands, etc. so that without modification we can copy/paste it into our compiler and run your example. |
|
|
pokiko
Joined: 27 Jul 2007
Posts: 33
|
|
| Posted: Mon May 19, 2008 2:38 am |
|
|
thanks
Last edited by pokiko on Wed May 21, 2008 12:48 am; edited 1 time in total |
|
|
ckielstra
Joined: 18 Mar 2004
Posts: 3680
Location: The Netherlands
|
|
| Posted: Mon May 19, 2008 5:30 pm |
|
|
| Code: | | setup_spi(SPI_SS_DISABLED); |
SPI_SS_DISABLED may only be used on a slave! For a master this define equals the define SPI_CLK_DIV_16.
To disable the SPI hardware you should use use:
| Code: | | SETUP_ADC(ADC_CLOCK_DIV_2); |
At 16MHz this is creating too short conversion times, check Table 11-1 in the datasheet. For 16MHz use DIV_32 or slower:
| Code: | | SETUP_ADC(ADC_CLOCK_DIV_32); |
CCP_2 is not initialised at startup.
| ckielstra wrote: | | Please post a small (max. 1 page) and complete program. |
The program is not small and does not compile because of a missing '}'. This means you never tested this version. Also there is an endless loop, again making this a useless test program.
The posted program contains a lot of irrelevant code. Please post a smaller and tested version.
You may think I'm lazy, and that's true, but there are also other reasons why I'm continuing to ask for a small and tested program:
- In making the program smaller there is a good chance you will find the problem yourself.
- I want to make sure I'm testing the exact same program you are using. Too many times in this forum the bug was found to be in the part that 'was left out for simplicity'.
When posting the new program also tell us what is wrong; what output do you expect and what are the results you see? How do you know these results are wrong? How did you measure this? |
|
|
ckielstra
Joined: 18 Mar 2004
Posts: 3680
Location: The Netherlands
|
| Re: tocki and timer1 |
| Posted: Mon May 19, 2008 6:13 pm |
|
|
| pokiko wrote: | When i setup_timer_1( T1_INTERNAL | T1_DIV_BY_8 ) then 10us corresponds to 10000 counts where i expect 5000.
So where does this doubly occur. Is there something which scales timer1 count by 2. |
I was looking at your original question again and am not sure if I understand what you mean.
You have configured Timer1 with a DIV_BY_8. So Timer1 ticks with:
16MHz / 4 / 8 = 1 tick every 2us.
Here you want to create a 10ms delay.
But 10.000 * 2us = 20ms.
Now what is the time you are measuring?
And what is your problem? |
|
|
pokiko
Joined: 27 Jul 2007
Posts: 33
|
| sorry, |
| Posted: Tue May 20, 2008 1:57 pm |
|
|
I am trying to write a rectifier program, controlling an 50hz ac signal. RA4/tocki input is connected to a
TLP-521 detects nearly zero cross point which is set to detect the signal's low to high point. As you mentioned,
2us*5000 counts have to correspond 10ms and 2us*10000 have to correspond 20us. But here i think 2us*10000 counts
coresponds to 10ms and 2us*20000 corresponds to 20us. Am i wrong?
Last edited by pokiko on Thu May 22, 2008 6:25 am; edited 3 times in total |
|
|
ckielstra
Joined: 18 Mar 2004
Posts: 3680
Location: The Netherlands
|
|
| Posted: Tue May 20, 2008 2:35 pm |
|
|
| ckielstra wrote: | | Please post a small (max. 1 page) and complete program. |
I'm wondering about the size of your pages... My pages contain 40, max 50 lines. Not over 340 lines.
I don't want to look into a large program like this, maybe I can post a small test program for you later this evening. If you are in a hurry you'll have to create a small test program yourself. |
|
|
pokiko
Joined: 27 Jul 2007
Posts: 33
|
|
| Posted: Tue May 20, 2008 2:54 pm |
|
|
| i am not in hurry. Thank you very much ckielstra. |
|
|
PCM programmer
Joined: 06 Sep 2003
Posts: 21708
|
|
| Posted: Tue May 20, 2008 2:57 pm |
|
|
You appear to have identical code in many of the cases. You can greatly
reduce the size of your code by letting each identical case "fall through".
Example:
| Code: | switch(chKeyPad)
{
case '0':
case '1':
case '2':
case '3':
case '4':
case '5':
case '6':
case '7':
case '8':
case '9':
cursorValue[ind]=(int8)chKeyPad-'0';
lcd_gotoxy(cursorPos[ind],1);
printf(lcd_putc, "%c",chKeyPad );
printf(lcd_putc, "\b" );
break;
} |
|
|
|
|
FAQ 
Search
Register
Profile 
Log in to check your private messages 
Log in
