I am working through the tutorial for MPLAB ICD on a 16F877. I have been
able to build the project and download the code, but I have never been able
to successfully run the code. When I try, one of the following happens:
1. I get an error that the operation could not be performed.
2. When I halt the program and get a snapshot of program memory, it's
executing garbage code (the instruction that is all 1s, i.e., 0x3fff).
Scanning through memory shows the same bad instruction at 0x1f00-0x1fff,
where the debug code is supposed to be. The program I downloaded
tut8f77.hex is in low memory, as it should be.
From the junk in high memory, I inferred that the debug code wasn't on the
chip, so I selected "Download ICD Operating System" and downloaded
ICD23000.OBJ. Since then, I have not been able to reprogram the chip. When
I try, I get the message "No Target". I am still able to do "Self Test".
The result is:
DLL Version: Version 1.4100
Firmware Version: Version 2.30.00
Debug Module: Not Exist
Self test: ok
> "No Target" might suggest a lack of power to the target board. Have you
> allowed the ICD to power your circuit or are you doing this separately?
Actually, the ICD is powered by the target circuit. If the ICD isn't
connected to the target at all you will not be able to connect to the ICD
from MPLAB.
By the way, regarding the original post on this thread:
1: You won't normally see the ICD code in MPLAB. MPLAB only knows about what
it downloaded to the chip, via the ICD. The ICD itself inserts the debugging
code to high memory.
2: Downloading the "ICD operating system" installs a new version of code
into the ICD itself (ie: the PIC that is on the ICD module). It has nothing
to do with the code on the target system.
3: Chances are the program was off running in high memory because of an
error in the code itself.
4: Perhaps there is something wrong with version 2.30 of the ICD operating
system? (I am running version 2.21.02 myself).
Perhaps the target chip is toasted?
Bob Ammerman
RAm Systems
(contract development of high performance, high function, low-level
software)
> Actually, the ICD is powered by the target circuit. If the ICD isn't
> connected to the target at all you will not be able to
> connect to the ICD
> from MPLAB.
I'm sure my simple $100 ICD is powered by an external power supply and then
supplies the target board with 5V,0V, Vpp, RB6 and RB7 when programming.
Maybe this is different to the dearer full blown ICD?
> 2: Downloading the "ICD operating system" installs a new
> version of code
> into the ICD itself (ie: the PIC that is on the ICD module).
> It has nothing
> to do with the code on the target system.
There is another flag to set in the ICD to download the monitor code into
the top of the PIC memory space before debugging can begin. (Something I
forgot to turn off and didn't realise why my boards weren't running stand
alone!)
> > Actually, the ICD is powered by the target circuit. If the ICD isn't
> > connected to the target at all you will not be able to
> > connect to the ICD
> > from MPLAB.
>
> I'm sure my simple $100 ICD is powered by an external power supply and
then
> supplies the target board with 5V,0V, Vpp, RB6 and RB7 when programming.
> Maybe this is different to the dearer full blown ICD?
If you have the MChip ICD it _is_ powered by the target board. Go look at it
again. :-)
Now if you are using the prototyping board that came with the ICD, that can
also provide the power, but in this case it _is_ the target.
I am not sure about the original Transdata version.
Bob Ammerman
RAm Systems
(contract development of high performance, high function, low-level
software)
> -----Original Message-----
> From: Peter Betts [SMTP:KILLspamext-peter.bettsKILLspamNOKIA.COM]
> Sent: Tuesday, March 27, 2001 2:29 PM
> To: RemoveMEPICLISTTakeThisOuTMITVMA.MIT.EDU
> Subject: Re: [PIC]: ICD Debug Module / No Target (1 of 1),
>
> > Actually, the ICD is powered by the target circuit. If the ICD isn't
> > connected to the target at all you will not be able to
> > connect to the ICD
> > from MPLAB.
>
> I'm sure my simple $100 ICD is powered by an external power supply and
> then
> supplies the target board with 5V,0V, Vpp, RB6 and RB7 when programming.
> Maybe this is different to the dearer full blown ICD?
>
The standard Microchip ICD definately takes it's power from the target
rather than the other way around. I've modified a cable so that the ICD is
continuously powered, and supplies power to the target when plugged in.
Stops MPLAB having a fit when I compile some code and the ICD has been
powered down.
> There is another flag to set in the ICD to download the monitor code
> into the top of the PIC memory space before debugging can begin.
> (Something I forgot to turn off and didn't realise why my boards weren't
> running stand alone!)
Are you referring to the "Enable Debug Mode" checkbox on the ICD Options
dialog? I have checked that. Or is there another flag I should set?
> "No Target" might suggest a lack of power to the target board.
> Have you allowed the ICD to power your circuit or are you doing
> this separately?
I am using the Microchip ICD evaluation kit. There is a power plug on the
Demo Board (in which I've connected a 9V regulated power supply). The Demo
Board is connected to the ICD Module through a telephone wire. Shouldn't
that take care of power to the ICD Module? The ICD Module is also
connected through a serial cable to the PC, of course, although I assume
that's just for data.
> I am using the Microchip ICD evaluation kit. There is a
> power plug on the
> Demo Board (in which I've connected a 9V regulated power
> supply). The Demo
> Board is connected to the ICD Module through a telephone
> wire. Shouldn't
> that take care of power to the ICD Module? The ICD Module is also
> connected through a serial cable to the PC, of course,
> although I assume
> that's just for data.
Yes you are 100% correct. Ignore my previuous posting, I took a guess and I
was wrong! :-(
> Are you referring to the "Enable Debug Mode" checkbox on the
> ICD Options
> dialog? I have checked that. Or is there another flag I should set?
Yes. The "Enable Debug Mode"
I wouldn't just load the code and hit RUN. I would load the code and then
reset the PIC and STEP through the code until it falls over or behaves oddly
It sounds like it's your code that's trashing and taking you to undefined
memory areas and then the debugger crashes or has problems because of that.
If STEPing is going to take you a long time try iteratively jumping over
small chunks of code setting a breakpoint each time. If it falls over you at
least have narrowed down where to look for the problem, somewhere between
the previous breakpoint and the new one.
Is there a pre-assembled version of the code or a finished assembler file
within the tutorial? Have you tried using this instead in case your typing
in of the example tutorial is flawed somewhere? This would at least test you
hardware was O.K.
Also are you happy with using the Simulator functions? Make sure you
understand how the code is meant to work under simulation before you delve
into the hardware.
No need to apologize for your wrong guesses. You've made a number of
helpful suggestions and comments that have increased my
understanding. While I have a lot of experience with computers, I'm brand
new to PICs so am pretty ignorant about them.
As for the code, it came with MPLAB, so it's probably not buggy. I didn't
type it in myself.
I'll try swapping boards, chips, and cables and see if things work better
today, as well as trying your other suggestions.
>As for the code, it came with MPLAB, so it's probably not buggy. I didn't
>type it in myself.
Do they not supply an example with deliberate bugs in it so that they can teach
using the ICD to find them? I also seem to recall that when the ICD was first
released there was an error in either documentation or code that came with it,
and there was an update on the M site. I may be wrong here, but there is
something lurking in the back of my mind that this happened ...
-- http://www.piclist.com hint: PICList Posts must start with ONE topic:
[PIC]:,[SX]:,[AVR]: ->uP ONLY! [EE]:,[OT]: ->Other [BUY]:,[AD]: ->Ads
I finally got it to work. I had connected the telephone wire between the
ICD module and the demo module. It worked after I connected the ICD module
to the ICD header. (If you want to see what I'm talking about, look at p.
17 of http://www.microchip.com/Download/tools/picmicro/icds/mplabicd/51184d.pdf.)
I had thought the two were equivalent, but I must have been wrong.
I greatly appreciate the help and support I got on the list. I had been at
wit's end (without a magazine).
Ellen
-- http://www.piclist.com hint: PICList Posts must start with ONE topic:
[PIC]:,[SX]:,[AVR]: ->uP ONLY! [EE]:,[OT]: ->Other [BUY]:,[AD]: ->Ads
mills.edu
