Go Down

Topic: Bad Atmega328 chips? (Read 3483 times) previous topic - next topic

bHogan

Hi you all,
Twice now I've had the problem where I can burn the bootloader on a chip but it won't accept an upload. My situation is this:

I made a little board with a ZIF socket with ISP and FTDI connections. When I need to program a batch of chips, I burn the bootloader from the IDE with a USBtinyISP, then switch to a FTDI cable and upload the program with the IDE. This works great, but in two batches of chips now, I found one that would take the bootloader but not upload the program (resp = 0 IIRC). Putting the chip in the Arduino = same.

Does anyone know of something to try to bring these two chips to life? (They were brand new - one from DigiKey and one from Sparkfun.)  I guess it's possible that the Rx & Tx pins are not working, but I want to make sure before I toss them. (They are, after all, the soul of the Arduino.)

Thanks
John
"Data is not information, information is not knowledge, knowledge is not understanding, understanding is not wisdom."
~ Clifford Stoll

Coding Badly

The most likely culprit is that the fuse settings are not correct.  Before chucking the processor, at least make one more attempt to burn the bootloader paying close attention to the fuse settings.

bHogan

Thanks CB,
I loaded the bootloader several times on these chips and even checked them with Avrdude. The fuses look fine.
They just won't accept an upload. Must be a problem with the UART or something.
"Data is not information, information is not knowledge, knowledge is not understanding, understanding is not wisdom."
~ Clifford Stoll

spycatcher2k

Dont chuck them (donate them to ME!!!!!!!) - only joking, why not just program them with your programmer without the bootloader, an extra 2k to use, and because you were going to bin them, they are essential FREE.

Drew.

Drew.
http://www.uk-pcb.co.uk - My UK Based PCB Fab & Assembly Company
Design work undertaken
SMD & Thru-Hole assembly

Jack Christensen

This sounds very similar to something I had happen with a 328p chip.  Which bootloader are you burning?
MCP79411/12 RTC ... "One Million Ohms" ATtiny kit ... available at http://www.tindie.com/stores/JChristensen/

Jack Christensen

Check the thread below for my tale of woe and intrigue. Never did solve it to my satisfaction, but I did find sort of a workaround. Would be very interested in comparing notes as there's still a mystery there...

http://arduino.cc/forum/index.php/topic,59394.0.html
MCP79411/12 RTC ... "One Million Ohms" ATtiny kit ... available at http://www.tindie.com/stores/JChristensen/

bHogan

Jack,
Looked over your post and yes it sounds very similar. I didn't give the crystal part a real close read - my proto board uses 22pF caps and I have the same problem if I put the chip in the Arduino. I don't have the inclination to spend the time trying to get 2 out of 30 chips working by futtizing around and loading sketches with the ISP. If they not going to work like all the rest, I don't want to deal with them. Good luck on getting your going, however.  :)
"Data is not information, information is not knowledge, knowledge is not understanding, understanding is not wisdom."
~ Clifford Stoll

Jack Christensen

Yeah I'm not sure how much the part about the crystal mattered in this case. Bottom line was the errant chip worked with the ATmegaBOOT_168_atmega328.hex bootloader, but would not work with Optiboot.  So I was just curious which bootloader yours didn't work with.
MCP79411/12 RTC ... "One Million Ohms" ATtiny kit ... available at http://www.tindie.com/stores/JChristensen/

bHogan

Oh, - it was just just the standard bootloader loaded from Tools in the IDE. I think the bootloader loaded fine. Got the flashing LED. Just couldn't uplaod a sketch after that. Same as your problem. If it was only one chip I'd just say it was a fluke. But 2 & yours - it's a pattern!
John
"Data is not information, information is not knowledge, knowledge is not understanding, understanding is not wisdom."
~ Clifford Stoll

Jack Christensen

Agree, but which bootloader? What board did you have selected when you burned the bootloader?
MCP79411/12 RTC ... "One Million Ohms" ATtiny kit ... available at http://www.tindie.com/stores/JChristensen/

bHogan

Oh (x2) "Arduino Duexxx or Nano w/ATmega328"
"Data is not information, information is not knowledge, knowledge is not understanding, understanding is not wisdom."
~ Clifford Stoll

Jack Christensen


Oh (x2) "Arduino Duexxx or Nano w/ATmega328"


Hmmm, that's the one that worked with my chip, it wouldn't work with the Uno (Optiboot) bootloader, where other chips did. So backwards I guess. Weird. Was hoping for a pattern. Still seems like similar symptoms. Date code on my chip is 1051.
MCP79411/12 RTC ... "One Million Ohms" ATtiny kit ... available at http://www.tindie.com/stores/JChristensen/

bHogan

Just to follow up . . .
I was up to 4 "bad" chips - out of ~50.
Using the same setup described in the 1st post, I did the same operation on another machine - same REL 22 installed.
I was able to bootload and program all 4 of the "bad" chips.
A mystery I'm not inclined to pursue, but just in case it happens to you or a loved one.
"Data is not information, information is not knowledge, knowledge is not understanding, understanding is not wisdom."
~ Clifford Stoll

Coding Badly


Jack Christensen


Just to follow up . . .
I was up to 4 "bad" chips - out of ~50.
Using the same setup described in the 1st post, I did the same operation on another machine - same REL 22 installed.
I was able to bootload and program all 4 of the "bad" chips.
A mystery I'm not inclined to pursue, but just in case it happens to you or a loved one.


Update appreciated.  I still have the one that gave me trouble, so I'll try that approach.  Maybe that'll chase the, ummm, gremlins out.
MCP79411/12 RTC ... "One Million Ohms" ATtiny kit ... available at http://www.tindie.com/stores/JChristensen/

Go Up