I was able to program them and then install them into their imbedded systems, no problem.
But there was a small error in the code and I need to re-program them. Now they will not accept the new programming.
The error was avrdude -1. The IDE suggests using the "-F" option with avrdude. Is there a way to turn on that option?
Further, is there a way to capture the file that is created (.HEX?) to be downloaded? I would prefer to capture the .HEX file once and get a version of avrdude that I can then push the new code to all the the MiniPro's that I need to update.
No, I do not use TX/RX. And I did see something about shorting TX/RX together, but it still does not work.
I am using the USBTiny ISP programmer. I have never gotten the FTDI one to work.
Both of the programmers will cause the reset. I can also still connect the FTDI and read the serial monitor.
I have all A1-A7 used as well as D3 and D4. Nothing else is connected. The FTDI holes are not connected to anything so I can slide in the FTDI and get serial monitor data.
Bruneaux:
Further, is there a way to capture the file that is created (.HEX?) to be downloaded?
Turn on verbose compilation logging (in Preferences). Then do a Verify of your sketch. The last line of the compile output will tell you where to find the .hex file:
I then programmed it AGAIN to see that it is still working. No Problem.
Then I soldered the daughter board to the MiniPro (I left the pins so I could still put wires on the correct pins). Everything's working properly.
Then I disconnected everything from the assembly and connected D10, D11, D12, RST, RAW and GND and attempted to program it again. It failed. I have done this with another daughter board design and it worked fine, as it should. Being able to do updates to imbedded systems is very important.
The only connections are passive components (resistors and capacitors) and they are all connected to A1-A7. D3 has nothing in line. D4 has a 1MOhm resistor to ground.
All I can imagine is that the ADC's are reacting to the capacitors in some way. But programming the Atmel chip shouldn't care, should it?
Any ideas? I have 50 of these assemblies to re-program or I'll have to scrap them. Even more scary is that I will never have a way to update them unless I can find why this is behaving this way.
I have yet to download AVRDUDE and attempt to force it with the -F flag. I hope that will work.
Bruneaux:
Then I disconnected everything from the assembly and connected D10, D11, D12, RST, RAW and GND and attempted to program it again. It failed.
Not surprising since the ICSP programming pins are:
MOSI (D11), MISO (D12), SCK (D13), Reset, VCC, and Ground.
The spec says you can provide 5V to RAW to get 5V out of the 5V regulator. I find that surprising since there is typically SOME dropout through the regulator. If you have regulated 5V I would connect it to VCC.
Sorry I misspoke. Have them connected to d11, d12 and d13. I did try again with the regulated 5v from the PC/USBTiny connected to VCC instead of RAW. I've never seen an issue with using RAW, the device can run reliably to about 4.4v.
But still not happiness.
I use a ZIF for programming and I usually use it unless the ZIF can't get a good connection on those critical pins. So I have female leads from D11-13, +5v and grd from my breadboard as a backup. Have tried both ways (ZIF and direct pins).
I have also completed a test with avrdude standalone. Still will not program.
Will try de-soldering the daughter board and program again with it. Horrible job and it will only be to test. I can't do it with 50 units.