Post by thread:PIC Timing Routines

Hi Jim, You might want to take a look at http://www.people.cornell.edu/pages/shb7/lop.html especially the "results" section. It gives three types of PIC delay loops, along with parameters for two of them which will give exact 10,000 100,000 and 1 million cycle delays. Sean At 11:46 AM 1/25/00 -0800, you wrote: {Quote hidden}> All, > > I am in need of a 10 mS and a 100 mS routine to run on a PIC. > I know I can write them, but if there are some out there, it > would be quicker to use those already written. If you have > such an animal, and would feel like sharing it, or have a > pointer to one, please let me know ASAP. I am trying to do > a quick and dirty test fixture, and time is short. Help me > out if you can. > > Thanks and Regards, > > Jim > >.....jimKILLspam@spam@jpes.com > | | Sean Breheny | Amateur Radio Callsign: KA3YXM | Electrical Engineering Student \--------------=---------------- Save lives, please look at http://www.all.org Personal page: http://www.people.cornell.edu/pages/shb7 shb7KILLspamcornell.edu ICQ #: 3329174

;------------------------------------------ ;call with no. of ms in work delay_xms ;uses temp, temp1 check for reg. conflicts movwf temp1 ;1 # ms in work ms_dely movlw .110 ;1 for fast crystal makes 1 millisecond movwf temp ;1 dlylp clrwdt ;1pretty short duty cycle i use wdt a lot nop ;1 decfsz temp,f ;1 goto dlylp ;2 decfsz temp1 ;1 goto ms_dely ;2 return ;------------------------------------------------------------- alice {Quote hidden}> All, > > I am in need of a 10 mS and a 100 mS routine to run on a PIC. > I know I can write them, but if there are some out there, it > would be quicker to use those already written. If you have > such an animal, and would feel like sharing it, or have a > pointer to one, please let me know ASAP. I am trying to do > a quick and dirty test fixture, and time is short. Help me > out if you can. > > Thanks and Regards, > > Jim > > .....jimKILLspam.....jpes.com

James Paul wrote: > I am in need of a 10 mS and a 100 mS routine to run on a PIC. > I know I can write them, but if there are some out there, it > would be quicker to use those already written. If you have > such an animal, and would feel like sharing it, or have a > pointer to one, please let me know ASAP. I am trying to do > a quick and dirty test fixture, and time is short. Help me > out if you can. ; PAUSE time ; A general-purpose delay routine that puts the PIC into a do-nothing ; loop for a 16-bit number of milliseconds (1 to 65535) at 4 MHz. Requires ; 16 cycles of overhead for call, return and other processing. found at: http://www.dontronics.com/convert2.html#pause Don McKenzie EraseMEdonspam_OUTTakeThisOuTdontronics.com http://www.dontronics.com World's Largest Range of Atmel/AVR and PICmicro Hardware and Software. Free Basic Compiler and Programmer http://www.dontronics.com/runavr.html

Hi Alice, What xtal freq. is this meant for? I count 551 cycles for the inner loop, so it would be around 2.2 Mhz as far as I can see. Is that right? Sean At 12:08 AM 1/25/00 PST, you wrote: {Quote hidden}>;------------------------------------------ >;call with no. of ms in work >delay_xms ;uses temp, temp1 check for reg. conflicts > movwf temp1 ;1 # ms in work >ms_dely > movlw .110 ;1 for fast crystal makes 1 millisecond > movwf temp ;1 >dlylp clrwdt ;1pretty short duty cycle i use wdt a lot > nop ;1 > decfsz temp,f ;1 > goto dlylp ;2 > decfsz temp1 ;1 > goto ms_dely ;2 > return >;------------------------------------------------------------- >alice > > > | | Sean Breheny | Amateur Radio Callsign: KA3YXM | Electrical Engineering Student \--------------=---------------- Save lives, please look at http://www.all.org Personal page: http://www.people.cornell.edu/pages/shb7 shb7spam_OUTcornell.edu ICQ #: 3329174

4 MHz clock. Attached is the asm code. Andrew On Tue, 25 Jan 2000 11:46:16 -0800 James Paul <@spam@jimKILLspamJPES.COM> writes: {Quote hidden}> All, > > I am in need of a 10 mS and a 100 mS routine to run on a PIC. > I know I can write them, but if there are some out there, it > would be quicker to use those already written. If you have > such an animal, and would feel like sharing it, or have a > pointer to one, please let me know ASAP. I am trying to do > a quick and dirty test fixture, and time is short. Help me > out if you can. > > Thanks and Regards, > > Jim > > KILLspamjimKILLspamjpes.com Content-Type: application/octet-stream; name="delays.asm" Attachment converted: wonderland:delays.asm (????/----) (00015DEF)

Ouch! shoulda checked that better before i sent it. It is for a 4MHz and the constant should have been 200 for 1001 cycles. 0xff lashes with wet noodle to me. alice. {Quote hidden}> Hi Alice, > > What xtal freq. is this meant for? I count 551 cycles for the inner loop, > so it would be around 2.2 Mhz as far as I can see. Is that right? > > Sean > > At 12:08 AM 1/25/00 PST, you wrote: > >;------------------------------------------ > >;call with no. of ms in work > >delay_xms ;uses temp, temp1 check for reg. conflicts > > movwf temp1 ;1 # ms in work > >ms_dely > > movlw .110 ;1 for fast crystal makes 1 millisecond > > movwf temp ;1 > >dlylp clrwdt ;1pretty short duty cycle i use wdt a lot > > nop ;1 > > decfsz temp,f ;1 > > goto dlylp ;2 > > decfsz temp1 ;1 > > goto ms_dely ;2 > > return > >;------------------------------------------------------------- > >alice > > > > > > > > | > | Sean Breheny > | Amateur Radio Callsign: KA3YXM > | Electrical Engineering Student > \--------------=---------------- > Save lives, please look at http://www.all.org > Personal page: http://www.people.cornell.edu/pages/shb7 > RemoveMEshb7TakeThisOuTcornell.edu ICQ #: 3329174

Subject: Time: 1:22 PM RE: Pic Timing Routines 1/26/00 Date: Alice Campbell wrote > and the constant should have been 200 for 1001 cycles. But 1001 isn't 1000, and what about all the overhead in the outer loop and the call and return? Pickey, pickey. I find that four instructions/loop works better. See if the following makes any sense. All of the following assumes a 4 mHz crystal. loops of the form: dlylp nop decfsz temp, f goto dlylp take 4 usec for each time through except the last which takes 3. If initializing the counter took 1 usec and we used W to save a register we could come up with an exact 1 msec delay: msecdly: movlw .249 ; one less time thru accounts for call and return mseclp: addlw H'FF' ; subtract one from W btfss STATUS, Z ; reached zero yet? goto mseclp ; no, subtract again return ; 2 + 1 + 248 *4 + 3 + 2 = 1000 usec If we envelope this with an outer loop having a certain number of usec overhead we have to somehow subtract out the overhead if the time is to come out right. We can't subtract 1 from the number and call the routine that number of times because the 1 represents over 1000usec. Each outer loop is also of this form and represents 4 usec overhead. The number we want to use to initialize the inner loop is not 249 but: ( nbr = number of millisecond we want to delay) 249 - (nbr/4) - (#instructions used to calculate nbr/4) how about this: delay_xms: movwf temp ; save # msecs delay bcf STATUS, C ; clear carry rrf temp, W ; divide by 2 and save .. movwf nbr ; in nbr, (temp not changed) rrf nbr, f ; divide by 2 again comf nbr, f ; compliment nbr incf nbr, f ; twos compliment movlw D'247' ; 249 -2,(for about 8 instruction overhead) addwf nbr, f ; subtract out outer loops overhead, save in nbr innerinit: movf nbr, W ; initialize inner loop dlylp: addlw H'FF' ; subtract one from W btfss STATUS, Z ; reached zero yet? goto dlylp ; no, subtract again decfsz temp, f ; now decrement outer loop counter goto innerinit ; more outer loops return This won't work out exactly because of the integer division but it should be closer than the code you suggested. I haven't tried this, ( or even tried to assemble it yet ), but I would be interested in any comments anyone has. Stan