PIC16F64 - flickering rest

[E-ten]

Please appreciate that I have not done any 'serious' electronics in over twentyfive years! Also note that my experience of PIC was only a couple of projects for some simple switching.

Anyway having dusted some old chips, updated my MPASMWIN (from XP!), I am now about to tear my (little left) hair out trying to understand the problem.

The problem is that when I run the program indicated below I get a flicker every 2.5 seconds or so, which looks as if the system is resetting.

I have removed a 555 timer I had incorporated in the board. No change.
I have disconnected the regulator, again fitted in the board. No change.
I have changed the oscillator from RC to XT and tried two different crystals. No change.
I have rewired MCLR and RTCC directly. Had these through 10ks because I intend to use them. No change.
I have tried three different chips with identical software. No change.

So I am beginning to thnk it is a software problem or a programming one. The programmer is one of those cheap XGecuTL866 II Plus.

Please examine the program, the hex and the read-back from the chip and let me know if you see any issues.

[Mod edit: put code in code box]

       LIST      P=16F54, R=DEC
        __FUSES   _XT_OSC & _WDT_OFF & _CP_OFF
        include   "P16F5X.inc"

;--------------------------------------------------------------------------
; Variables
;--------------------------------------------------------------------------

Ram             EQU     h'0C'
Count           EQU     Ram+1


;--------------------------------------------------------------------------
; Program Code
;--------------------------------------------------------------------------

        ORG     0
                   
   MOVLW   B'00000000'   ;Set PORTB for output
        TRIS   PORTB
 
   MOVLW      93               ;Set something to see
   MOVWF   Count

Loop   MOVF   Count, 0   ;Dump value to port to see something
   MOVWF   PORTB           ;Dump W to PORTB

   END

The Hex file is:-
:020000040000FA
:0C000000000C06005D0C2D000D02260017
:021FFE00F90FD9
:00000001FF

And the chip read-back is:-
0C00 0006 0C5D 002D 020D 0026 0FFF

Any ideas would be much appreciated.

Regards,
ep

 

[E-ten]

As an aside, this was my first post with code and it is seriously ugly.
Perhaps someone can tell me which icon boxes the code so that it stands out.

ep

 

[Harald Kapp]

It's right here.

 

[PETERDECO]

Welcome to the forum. It sounds like it could be the Watchdog Timer timing out in a loop. Try disabling it.

 

[E-ten]

It's right here.
View attachment 52678

Thanks for that! It is under a PLUS sign till activated.

 

[E-ten]

Watchdog Timer? Surely one only needs to deactivate it in the Fuses line?

Regards,
ep

 

[PETERDECO]

You can deactivate it in the configuration bits in the programming software. Usually near the Code Protect on and off.

 

[E-ten]

I think I already have done that - in the Fuses line as shown in the listing.
Unless there is some other setting I have not noticed?

 

[E-ten]

Still don't understand what is going on and what is the solution.

Is the programmer not setting the fuse _WDT_OFF ?
Or does one also have to do something else within the software?

Am afraid I do not have the experience to decode your:-
"You can deactivate it in the configuration bits in the programming software."
I thought that was what I was doing in the Fuses line.

Regards,
e10

 

[E-ten]

Somebody asked whether I had considered the fuses being not being configured.
Of course I hadn't!
When I tested for crystal operation I was using an old chip that I knew was crystal based since I have never used RC till now.

Further tests suggest that, if the fuses are not configured RC takes precedence although badly - I notice that pin 15 has got a nicely squared waveform at the same frequency of the oscillator rather than the expected Fosc/4.

So I now have a different question, different enough to warrant its own thread.

Regards,
ep