As I read this forum, I occasionally encounter people mentioning to beware of turning Atmel MCU into a brick with wrong fuse value. Since the consequences are so terrible, may I ask what are the things to look out for to manage this risk?
One thing I can think is never use RESET pin as PIO because ISP programming will be disabled. Can the more experienced members add more?
If you set the RSTDISBL fuse, you can recover if you have a high-voltage parallel programmer. These are pretty rare there days, though. So this is probably the more serious problem.
If you set the clock source wrong, you may need to inject a clock to get to a point where you can program the chip "normally." This is relatively easy, but certainly non-trivial. (for instance, if you have an SMT arduino, the pin that you need to inject the clock into a pin that is only connected to some rather small components. Not impossible, not as difficult as doing HV parallel programming, but probably more than enough to leave the average arduino user with a dead board...)
During the last week I have managed to brick a ATtiny85 at least 5 times.
What I was trying to do was setting the fuses to use an external 32.768 x-tal.
The result was that the tiny was not responding to the programmer, it could not read the device signatur, so for me it was bricked.
Erni:
What I was trying to do was setting the fuses to use an external 32.768 x-tal.
The result was that the tiny was not responding to the programmer, it could not read the device signatur, so for me it was bricked.
When you run the clocking down that slow, you will need an ISP programmer that can also slow down the sck clock enough. For my AVRISPmkII, I need to set it down to a 4KHz sck clock to program an ATtiny13A @ 16KHz. I use the avrdude -B250 option for that.
and the -B250 option too, so I guess I must use another programmer.
I am using TinyISP which have an option for slowing down the clock, it workes OK for Attiny13 at low clockspeeds.
It also could be the fact that my clock x-tal is not oscillating, so I will keep on experimenting.