| View previous topic :: View next topic |
| Author |
Message |
Ttelmah
Joined: 11 Mar 2010
Posts: 19215
|
|
 Posted: Fri Jan 20, 2023 11:37 am |
 |
|
He started asking for a frequency generator, then switched to talking
about measuring frequencies, but when asked how he was generating
the frequency (which is vital, since if he is using a timer, this is not
available for measurement), did not answer. At this point I gave up.
He seems to want us to answer things telepathically, rather than giving
us real data. |
|
 |
temtronic
Joined: 01 Jul 2010
Posts: 9093
Location: Greensville,Ontario
|
|
| Posted: Fri Jan 20, 2023 12:57 pm |
|
|
Well his request did get me to dust off my PICkit3-r1, FIND some 683s,cut code and program a 19KHz signal !
I also found some NE544s and NE5044 while looking for the PICs....turned on the EICO 950B R-C-Comparator and the 'magic eye' still works !
One of these days I suppose I should clean up the 'lab'. |
|
|
sifirzero1
Joined: 14 Dec 2022
Posts: 37
|
|
| Posted: Fri Jan 20, 2023 2:35 pm |
|
|
| It is important to measure the 18khz and 25khz range and to have a timer0 input. |
|
|
sifirzero1
Joined: 14 Dec 2022
Posts: 37
|
|
|
temtronic
Joined: 01 Jul 2010
Posts: 9093
Location: Greensville,Ontario
|
|
| Posted: Fri Jan 20, 2023 3:21 pm |
|
|
Well it is possible, but a LOT easier if Timer1 could be used.
You should google 'measuring pulse width' for more options on how to code.
BTW I don't know the language that posted link is, so I can't download whatever code there is. It'd be nice to see it in English. |
|
|
temtronic
Joined: 01 Jul 2010
Posts: 9093
Location: Greensville,Ontario
|
|
| Posted: Sat Jan 21, 2023 6:08 pm |
|
|
| Code: |
//683FC1
//test code for 25KHz frequency counter
#include <12f683.h>
#fuses INTRC_IO,NOWDT,PUT,NOPROTECT,MCLR //basic working fuses
#use delay(internal=4M) //4MHz internal
#use RS232(baud=9600,xmit=pin_a4) //data via TTL<>USb module
int8 lc=0; //loop counter from Timer0 interrupt
int16 f=0; //frequency
int8 tc=0; //timer0 count
#int_timer0 //timer0 ISR
void isr_timer0(void) { //number of loops of 256 counts
lc=lc+1; //of input to Timer0 (1 count=1 Hz )
}
//================================
void main()
{
setup_comparator(NC_NC_NC_NC); //disable comaparators
setup_adc_ports(NO_ANALOGS); //disable ADC pins
setup_timer_0 (RTCC_DIV_1 | RTCC_EXT_L_TO_H); //no prescaler, external source,rising edge
enable_interrupts(GLOBAL); //enable interrupts
delay_ms(1500); //premain loop delay
printf("\n\r Timer0 frequency program \n\r"); //sign on messgae
while(TRUE) //do forever
{
lc=0; //loop counter
tc=0; //timer0 count
f=0; //frequency
set_timer0(0); //preset timer0
enable_interrupts(int_timer0); //software 'start' for timer0
delay_ms(1000); //do nothing for 1 second
disable_interrupts(Int_timer0); //software 'stop' for timer0
tc=get_timer0(); //get tiemr0 value
f=(lc*256) + tc; //calculate frequency
printf("lc %u tc %u f %lu \n\r",lc,tc,f );//send to PC terminal program
} //end of forever while
} //end of main
|
OK, decided it was easier to cut this 'Timer0 only' code than try to 'find' the 1,000 emails that Gmail 'lost' from my account....
Considering it's running on the internal RC oscillator, the 'math' says it's within 2% accuracy, probably not the most efficient or neat code. Uses about 1/4 of ROM and 20% of RAM.
Jay |
|
|
sifirzero1
Joined: 14 Dec 2022
Posts: 37
|
|
| Posted: Sun Jan 22, 2023 2:26 am |
|
|
| Thank you so much. I will try the code as soon as possible. good luck with your work. |
|
|
temtronic
Joined: 01 Jul 2010
Posts: 9093
Location: Greensville,Ontario
|
|
| Posted: Sun Jan 22, 2023 7:38 am |
|
|
I changed the sampling time to 1/2, 1/4 and 1/8th of a second, corrected the 'math' and at 1/2sec, +1% error, 1/4sec, +3% error, 1/8th, +7% error.
So, if you can live with the error (or edit the 'math'), you can sample faster then the PIC could do more than just be a frequency counter. |
|
|
sifirzero1
Joined: 14 Dec 2022
Posts: 37
|
|
| Posted: Sun Jan 22, 2023 8:48 am |
|
|
| The first value of the codes you have given is correct, there is a deviation in the next values. Calculating wrong. The codes are working very nice. |
|
|
temtronic
Joined: 01 Jul 2010
Posts: 9093
Location: Greensville,Ontario
|
|
| Posted: Sun Jan 22, 2023 10:29 am |
|
|
To get real accurate results, you need to KNOW what frequency the PIC clock is operating and then adjust delay_ms(xxx) so that the 1 second 'delay' (where counting is done ), is exactly 1 second.
The internal RC is a 'nominal' 4MHz not 4.00000000 MHz, the delay_ms() function may be off, say 1/2ms, that 500usecs, so 'errors' are to be expected.
You should be able to use delay_us(xxxxxx) to get a more accurate 'delay' during the counting stage.
Also if you use a crystal cut to a 'binary' frequency, like 2.457600 MHz, you'll get better results.
What I posted was a 'quick and basic' frequency counter using Timer0, easy to copy/edit and make better.
Jay |
|
|
sifirzero1
Joined: 14 Dec 2022
Posts: 37
|
|
| Posted: Sun Jan 22, 2023 1:20 pm |
|
|
| 20mhz oscillator delay_ms(986); as accurate measurement |
|
|
sifirzero1
Joined: 14 Dec 2022
Posts: 37
|
|
| Posted: Sun Jan 22, 2023 1:21 pm |
|
|
| thank you so much. I understood the logic |
|
|
|
FAQ 
Search
Register
Profile 
Log in to check your private messages 
Log in
