Searching \ for 'wdt question-long' in subject line. ()
Make payments with PayPal - it's fast, free and secure! Help us get a faster server
FAQ page:
Search entire site for: 'wdt question-long'.

Truncated match.
PICList Thread
'wdt question-long'
1999\01\07@110733 by Alice Campbell

Content-type: text/plain; charset=US-ASCII
Content-transfer-encoding: 7BIT
Content-description: Mail message body

i have been working on a datalogger project that requires
recalibration of a timing loop based on outside temperature.  Chip is
16c84.  i created a small program to
setup the wdt
clear a counter
run a loop till timer times out
display answer on leds at portb
go to sleep
start over.

im attaching the code.  .

heres the problem:  it doesent work.  when wdt trips, it goes back to
top, instead of being routed to display routine, it reinitializes the
count register, and basically orbits between the loop and
re-initialization.  playing around with it, i cant find evidence that
the wdtimeout really clears bit 4 in status.  as you can see from the
code, i tried testing bit 4 first, then bit 3,, then the other way
around.  i tried catching it after the initialization.  i tried to
catch it by using the presence of data in the count register to goto
the display routine.  i have confirmed that the routine at the top
does implement restart (status,4=1 and status,3=1) ok, sleep looks
ok, and mclr looks ok.  but wdt doesent seem to work...

im using most recent (dos) mpasm on old 486 laptop.

and what will happen if this is in a subroutine?  the thought scares
me.  wakeup on sleep just continues with program counter, wdtimeout
loses the stack???  i stayed home from work yesterday to work on this
and im stumped.

any help appreciated.


Content-type: text/plain; charset=US-ASCII
Content-transfer-encoding: 7BIT
Content-description: Text from file 'pict7.asm'

;======pict7.asm============1 aug 98===============
       list    p=16c84
       radix   hex
;cpu equates (memory map)
portb   equ     0x06
mine    equ     0x0a
mflag   equ     0x0b
check   equ     0x0c
temp    equ     0x0d
status  equ     0x03
tmr0    equ     0x01
intcon  equ     0x0b
optreg  equ     0x81
trisb   equ     0x86
on      equ     1
to      equ     4
pd      equ     3
w       equ     0
c       equ     0
f       equ     1
z       equ     2
rp0     equ     5
       org     0x000
;        btfss   status,4    ;timeout
;        goto    to0ck        ;=0
;        goto    to1ck        ;=1

;to0ck   btfsc   status,3    ;powerdown
;        goto   display      ;to=0&pd=1=wdt
;        goto   go           ;to=0&pd=0=sleep

;to1ck   btfss   status,3
;        goto    display         ;to=1&pd=0  mclr (could use for serial?)works
;        goto    start           ;to=1&pd=1 startup

       btfsc   status,3       ;-pd?
       goto    test1           ;pd=1  means pu or clwdt
       goto    test2           ;pd=0  set by sleep
       org     0x005

test1   btfsc  status,4        ;pd=1,to? clrwdt sets to=1
       goto   start            ;to=1& pd=1->powerup
       goto   go       ;display          ;to=0&pd=1->wdt  ??????????

test2   btfss   status,4       ;pd=0,to=?
       goto   go            ;pd=0&to=0 ->sleep ???????
       goto   display          ;pd=0&to=1 ->mclr

       movlw   b'00000000'     ;load W with 0x00
       tris    portb
     ;  clrwdt                  ;to=1&pd=1  this prevents wdt in go
       movlw   b'11011000'     ;wdt=1&1:1...timeout is 18 ms
       bcf     status,rp0      ;back to bank0
       movlw   0x00               ;?????????????
       movwf   mine            ;put something in mine to start
;        clrf    portb           ;all lines low
       movlw   b'00000111'     ;bsf portb,0
       xorwf   portb           ;is this looping?
;        goto    display

go     ; clrwdt
       movfw   mine            ;if wdt then mine has count in it
       sublw   0x00               ;is it empty?
       btfss   status,z        ;z  well,does it?
       goto    display         ;cool.  show me

calib   clrwdt                        ;at this point prescale is 0
cycle                           ;time 1 wdt cycle
    ;   incf    mine,f          ;1 tick
    ;   movlw   b'00000010'     ;bsf portb,0
       movfw   status
       movwf   portb
       xorwf   portb           ;is this looping? ...yes,but....
       goto    cycle           ;2 ticks       5 total here
       bsf portb,6             ;never gets here if wdt

display; clrwdt                  ;show mine if wdt handled right
       movfw    mine
       movwf   portb           ;did it count??
       clrf    mine            ;setup for next run
       bsf     portb,2
       goto    snooze          ;all done

snooze ; clrwdt                 ;po=1&pd=1
;        movlw   b'11011111'    ;reset prescaler for long sleep
;        option
;        bcf     status,rp0
       sleep                   ;pd=0,to=1.  pgm continues...
       bcf     portb,2
       movlw   b'11011000'
       goto   go               ;loop

;at blast time, select:
;       memory unprotected
;       watchdog timer disabled (default is enabled)
;       standard xtal using 4 mhz
;       powerup timer on

1999\01\07@193838 by Peter Grey

picon face
At 12:08 AM 7/01/99 PST, you wrote:
>Content-type: text/plain; charset=US-ASCII
>Content-transfer-encoding: 7BIT
>Content-description: Mail message body
>i have been working on a datalogger project that requires
>recalibration of a timing loop based on outside temperature.  Chip is
>16c84.  i created a small program to
>setup the wdt
>clear a counter
>run a loop till timer times out
>display answer on leds at portb
>go to sleep
>start over.
>im attaching the code.  .

When SLEEP is executed the program will stop there and on a WDT timeout it
will start at the next instruction and then go to the label "go". It will
not go to the instructions where you evaluate what type of timeout it was.
This should be done after the SLEEP instruction if this is what you want.

good luck,

Peter Grey

{Quote hidden}

1999\01\08@024250 by Dr. Imre Bartfai

it is obvious, that WDT does not work, if it is disabled upon programming.
Enable it before blast.

I hope this helps.

More... (looser matching)
- Last day of these posts
- In 1999 , 2000 only
- Today
- New search...