Good evening everyone!

As you may know, I already uploaded a bootloader into an Atmega325P.

I'm trying to do it again for another MCU of the same model. This time however, I can't communicate with the chip.

Armed with my multimeter, I searched everywhere for the problem's source and I think I found an interesting hint: I don't have any signal on the MISO pin when my usbasp is trying to connect.

Does this ring any bells? Reminder of a past problem?

Thanks!

I realized there isn't supposed to be a signal on MISO.

Back to square one.

I tested the MCU on another PCB and it works fine. When I put it back where it belongs, nothing. I did communicate with it one time, but I never succeeded in reproducing the results.

The two PCBs are almost the same, except some minor changes and smaller traces in the new model.

All the tests were done on almost empty PCBs: only the MCU and a 14.7456 MHz crystal.

I wonder if the smaller traces could interfere with the SCK signal...

I finally discovered what was wrong...

My first MCU was apparently broken. The second one had a small short between two pins. When I finally succeeded in connecting to it, I uploaded my program and at the same time I unwillingly uploaded new fuses' settings.

These problems might seem quite ordinary, but when together, they leave a solid WTF impression.