Arduino 2009 cpu died for no appearent reason

After a few days of successful tinkering, my Arduino 2009 is dead.

The pin 13 led does not blink, uploading gives error, the tx/rx leds blink occasionally before the error message occurs. Xtal runs, usb and power section check out OK.

I guess I somehow ESD'd the micro, either that or the bootloader vaporized. I did not expose the unit to any dangerous voltages or situations, but it's gone nevertheless. I need to take better care next time..

I ordered replacements, no big deal - only annoying issue is that I cannot play with it this weekend.

What is the accumulated experience here? Does the bootloader flush spontanuously, or is the cpu known to be fragile?

Using a switchmode 12V adapter and longish wires on the data pins is appearently pushing it, I guess I'll need to add better anti-ESD ckt's.

I have numerous 2009 and homebrew boards, the processor isn’t particulary sensitive in my experience. I cooked the bootloader once but thats it.

Yeah I have also never fried anything on my arduino boards but I thought I had when I used my arduino while it was sitting on some stripboard (doh! lots of random shorts) but it is still working now...

That is precicely the reason why I made my acrylic covers for the arduino boards though: http://www.flickr.com/photos/mowcius/sets/72157622349824171/

Mowcius

yep and the same reason mine is mounted on a purdy piece of plywood, altho it does nothing for the top

but I have cought myself a time or 2 recently where I take off my jacket (I am a smoker , and I smoke outside), ground myself to something and still have enough static left over to zap my hand on something else a second time

so good ESD practices in wintertime might be a wise idea

Agreed. When I take off my jacket, I always touch the metal pull-string for my overhead light (which usually delivers a nice static shock), but that stops me from shorting other things out.

One time I had a lose +5V jumper hanging around, and it touched the trace to the pin13 LED. That thing lit up red/orange very brightly, but I was able to get the jumper off before any noticeable damage occurred.

I suppose the moral is just to always be careful.

or my personal guideline

don't fiddle with wires while its on!

don't fiddle with wires while its on!

Or use an arduino cover ;)

Mowcius

have one for a modified bare bones board?

If you can give me the dimensions then I could get one sorted for you. Unfortunately it will be a week until I can do any more but normally I would be able to do stuff within a week. PM me if you are interested. I would also be interested in doing one for the BBB anyway. Annoyingly the design files do not seem to be on their website.

Mowcius

don't fiddle with wires while its on!

Indeed. This incident was on the second day I had the thing ;). I as still an uber-noober ;D!

I have been building AVR based applications for some time, some in RC planes - and found them rather bullet-proof, hence my amazement over the sudden demise of my Arduino.

My theory so far is that the 12V switchmode adapter is (as is obligatory to meet CE requirements) fitted with HF bypass caps. This generates ~110V ac commmon mode into a high impedance towards ground. I was measuring the motion of the lsb digit of my gas meter using an IR gate on long leads - and the gas meter is well grounded. The board appearently died when I plugged in the adaptor.

I should have known better.

Just out of interest - is there a scenario possible where the cpu still lives, but only the bootloader is flushed? (in which case I might be tempted to dig up my old AVRISP)

Thanks for all the comments so far!

EDIT The plot thickens...

I used the http://www.geocities.jp/arduino_diecimila/bootloader/index_en.html trick to reprogram the chip. Weird thing is that reprogramming works perfectly. No errors. verifies OK, can read fuses etc. ..but it does not work..!

Am I right in assuming ...arduino-0018\hardware\arduino\bootloaders\atmega\ATmegaBOOT_168_atmega328.hex is the right bootloader? (tried them all - no luck)

It could just be you have zapped some I/O pins which makes it look like nothing is working but the CPU is OK.

Just like someone in a coma, brain OK but no I/O.

I got the replacement 328 chip, added some protection resistors on the ADC inputs en PWM outputs.

I also made a 4xNiMH pack (originally a RC plane RX pack) compatible with the USB connector.

I'm back in business! It works again ;D

I know the server is running slow but if you keep refreshing then you send loads of replies! Any chance you could delete a couple?

Glad it's working again for you now though :)

Mowcius

Sorry for inundating the forum with identical messages - the connection kept timing out and I presumed the message was not sent. I deleted the excess stuff. 8-)