Re: [avr-chat] avrdude : chip not responding

[Vincent Trouilliez]

On Wed, 2005-08-31 at 22:20 +0200, Peter wrote:
> Have you allready tried exchanging the resistor on MISO against 330 Ohms yet?
> I had the EXACT same problem with uisp and an Atmega8 yesterday, and changing 
> the resistor from 1,5k to 330r fixed it. You might even try this line without 
> a resistor, as long as the PC is allways on while the Atmel is connected to 
> it.

Hi Peter,

Yes, I undid my cable and replaced the 1K resistors with 330ohms ones,
still no joy. I think I will remove the resistors from the cable DB-25
connector, and simply implement them directly on the avr board so that I
can easily/quickly fiddle with them. 
I then shortened the cable as much as I could, from 2.65 metre down to
one metre, but still no joy.

I am currently transferring the avr from the breadboard onto a
prototyping board, so all connections are tight and soldered, can only
help.
After that, I will try and add a line buffer on the avr board. This way
I could get a perfect signal from/to the avr while getting rid of the
resistors in all safety on the MOSI and SCK lines.
I have already tried swapping avr's (I bought two of them, thinking I
could get them to chat with each other), but again no joy.
Using the "-F" option with avrdude, I could "happily" ;-) do a flash
memory dump with the MISO line DISCONNECTED, and the avr was still send
ing stuff on this line ! So it's now 200% sure, the avr does "answer",
but apparently not appropriately ! Maybe the stuff he sends is just
garbage because it's read protected (somehow ?) and rejects all read
instructions and just "echos" on MISO whatever comes from MOSI/avrdude ?

If only I had someone around me with a "proper"/working avr programmer,
I could put my chip in it and see what it does ! Is there anybody in
France ? Or in a nearby country ? The AVR is very lightweight and small
so I could probably sent it by post for very little money. Since I have
two of them, I could still experiment on one while you see what's going
on with the second ?

If only I had a DSO with tons of memory to record exactly what's going
on on the SPI lines... :o(


Regards,

--
Vince, 6h40 AM, resuming experiments...