argh... meant to copy the list in case anyone had something to add.
-dave
--- Begin Message ---

Bob Paddock wrote:

 There is no
program to run yet, you can't program
the I2C disable bit. :-<

OK, that *is* bone-headed.

After reading the whole thread, the best idea that comes to mind is to build a small fixture that is a repeater board between the production programmer and the product. The germ of the idea is that the repeater board should disable the I2C part, and then become transparent to the ISP programmer.

Possible implementation:

1) ISP headers in and out for connection to programmer and product.
   These are wired through, or may need buffers.
2) A small AVR, wired to bit-bang the SPI lines.
3) Software that:
   a) bit-bangs the magic "I2C-B-Gone" sequence onto the
      outing SPI header. (Hard coded, or stored in eeprom,
      or whatever)
   b) tri-states the AVR so that the ISP can drive

I suppose it may also require some buffers or so forth so that the bit-bang sequence doesn't back-drive the AVR-ISP or something, but perhaps some or all of the lines can simply be wired through.

And, it will become a production line problem unless a very simple software interface can be designed to drive the whole programming rig in the right sequence from a script. So, maybe the fixture board needs to have an RS-232 serial interface on it so that it can triggered from a script.

-dave


--- End Message ---
_______________________________________________
avrdude-dev mailing list
[email protected]
http://lists.nongnu.org/mailman/listinfo/avrdude-dev

Reply via email to