MPLAB and debug mode error PK2Error0028

[edeca]

I've just spent 30 minutes wondering why debug mode didn't work on a 16F88. Everything seemed to compile and program fine but I constantly got the error "PK2Error0028: Unable to enter debug mode".

It was all because I had:

#include "p16f88.inc"

Instead of:

#include <p16f88.inc>

So this is mainly here for my memory, but perhaps it will help somebody else in the long run.

 

[edeca]

Actually that hasn't helped at all. I'm now getting the same error again. The 16F88 is wired the same way as the 18F1320 I just took off the board. Pull-up resistor and diode to MCLR.

My program is as simple as they get:

		errorlevel -302		; Disable bank warnings
		#include <p16f88.inc>
		list p=16f88

		__CONFIG	_CONFIG1, _CPD_OFF & _CP_OFF & _DEBUG_ON & _LVP_OFF & _MCLR_ON & _PWRTE_ON & _WDT_OFF & _WRT_PROTECT_OFF & _INTRC_IO & _BODEN_OFF

		;cblock 0x20		; Variables
		;	test
		;endc

		org 0
		nop
		goto Start

Start	clrf	STATUS
		clrf	PORTA
		clrf	PORTB

		bsf		STATUS, RP0 	; Bank 1
		clrf	TRISA			; All PORTA output
		clrf  	TRISB			; All PORTB output
	
		; Disable A/D, all ports digital
		clrf 	ANSEL
	
		; Disable comparators
		movlw	0x07
		movwf	CMCON
	
		; Set 4 MHz clock
		;banksel OSCCON 
		bsf		OSCCON, IRCF2
		bsf		OSCCON, IRCF1
		bcf		OSCCON, IRCF0
	
		bcf		STATUS, RP0		; Bank 0
		movlw	0xFF
		movwf 	PORTA

Loop	goto 	Loop

		end

Do I need to do anything special with the Junebug to make ICD work? Programming/verifying is fine even when I'm getting the error.

 

[blueroomelectronics]

Remove
_MCLR_ON & _PWRTE_ON & _DEBUG_ON

 

[edeca]

Remove
_MCLR_ON & _PWRTE_ON & _DEBUG_ON

Thanks Bill, another stupid error! Do I need to put them back if compiling for release? Or will changing profiles in MPLAB do that for me.

 

[edeca]

Hm. I still get the error sporadically. Here's the output, which I think is fine except the last error.

Initializing PICkit 2 version 0.0.3.61
Found PICkit 2 - Operating System Version 2.32.0
Target power detected ( 4.99V)
PIC16F88 found (Rev 0x8)
PICkit 2 Ready

PKWarn0006:  The local copy of program memory has been changed since the last program operation.  Should PICkit 2 program the target (fix) before proceeding?
Programming Target (08/03/2009  21:53:46)
PIC16F88 found (Rev 0x8)
Erasing Target
Programming Program Memory (0x0 - 0x13)
Programming Program Memory (0x800 - 0x803)
Verifying Program Memory (0x0 - 0x13)
Verifying Program Memory (0x800 - 0x803)
Programming Debug Executive (0x-F00 - 0xFFF)
Verifying Debug Executive (0xF00 - 0xFFF)
Programming Debug Vector
Verifying Debug Vector
Programming Configuration Memory
Verifying Configuration Memory
Running Target
PK2Error0028:  Unable to enter debug mode
Debug mode entered, DE Version = 0.0.0
PICkit 2 Ready

My program hasn't changed, except your recommended removal which makes config:

__CONFIG	_CONFIG1, _CPD_OFF & _CP_OFF & _LVP_OFF & _WDT_OFF & _WRT_PROTECT_OFF & _INTRC_IO & _BODEN_OFF

This is driving me nuts now. I thought I'd be able to sit down and write out "my first ASM program" but I haven't got past flashing the status LED because I can't debug. How infuriating

 

[blueroomelectronics]

It looks ok, what version of MPLAB are you running?

 

[edeca]

MPLAB 8.2. I've just switched from the ICD header to the pins along the top of the Junebug and get the same results. I have also tried powering from the Junebug rather than external power.

MPLAB also constantly tells me that the project is out of date, even if I don't change anything. I'm unsure what causes that either.

In any case, I'm not sure my Junebug has ever been right since I had to separate the tracks with a knife! I sit down for a day to play and never seem to get past ICD errors when I start trying hardware instead of simulation.

I'm tempted to pick up a PICkit2 for £30.. if only there weren't so many cheap Chinese knock offs!

 

[blueroomelectronics]

I've tried your program with both a genuine PK2 and my Junebug. Both give the error with MPLAB 8.20a but you can still single step.
I've modified the program for a 18F1320 and it works fine. Seems like an MPLAB 8.20a bug...

 

[edeca]

Weird. Sorry to sound negative about the Junebug, my frustration got the better of me!

I'll bug it at Microchip if I can find the right place. I need the A/D on the 16F88 or I'd swap it out for another part.

 

[UTMonkey]

I have had this with the PICKIT2 - unplugging the USB and closing down MPBLAB seems to get me back on track. Doesn't really happen that often though.

 

[edeca]

I've tried that a few times. The only thing I haven't tried is another machine, I'll see if I have time tonight to install the software and try it on my laptop.

 

[blueroomelectronics]

I wonder if it's the new absolute vs relocate mode in 8.20?

PS the 18F1320 has an improved AD vs the 16F88. It doesn't have the comparitors though.

 

[edeca]

An interesting thought. I'll try an older MPLAB tonight and also try both of the modes (I don't remember which I chose). I haven't raised a support ticket yet so I will wait until I've tested.

I'm flying to Canada for a week (hi Bill and anybody else there ) so wont have much time to play.

 

[blueroomelectronics]

Enjoy your stay here eh.

 

[edeca]

There's also this: Microchip Technology User Forums

But I normally do this anyway, as I get annoyed by the IDE reminding me to reprogram.

 

[blueroomelectronics]

Make sure you check program after succesful build.

 

[edeca]

Bill, a quick look at the Junebug circuit suggests the two 4.7K pull-down resistors are included, so there's no chance I need those?

 

[blueroomelectronics]

It's the same as the red button PICkit2. Try your program with the 18F1320 and see if the problem goes away. Any reason the A/D in the 18F1320 is not good for your application?

 

[edeca]

Actually I was just thinking I'd try a 16F part as I've never used them. Perhaps I'll stick to the 18F for now, that's not a bad idea.

Thanks for the clarification on the Junebug. I'll target my application at the 18F1320 and perhaps try another project on the 16F88 later.

 

[blueroomelectronics]

The 18F are so much nicer to use than the 16F and cost about the same. The A/D also includes a programmable acquisition timer so it's really easy to use.

PS set the OSCCON as early as possible to because the debugger will run so much faster...

movlw 0x62 ; 4MHz
movwf OSCCON