I think I destroyed my SMD Uno

I have a SMD Edition Uno. It used to work. It no longer accepts uploads.

  • I tried the loop-back test. Passed.
  • The +5V and +3.3V pins read correctly.
  • The "on" LED lights up.
  • Attempts to program it via ISP fail with "unable to enter programming mode"

The SMD processor chip will be difficult or impossible for me to remove and replace. Any suggestions? Or just throw it into the bin?

I suppose it could be retained as a USB to serial converter.

Were you previously able to program it via ISP?

Yes, previously it worked normally.

Any idea what might have happened to it? I accidentally programmed the RSTDISBL fuse bit recently. That slowed the whole process right down.

I suspect static electricity. I'm usually careful but maybe I picked it up when I had a charge.

Do you have a high-voltage programmer? (Maybe the static discharge flipped the RSTDISBL fuse bit?)

What did you last upload to it? Does that program still run?

I do, but I don't think I can program it without removing the chip. Or can I? Atmel warn against attempting that with chips on boards, but maybe it would be OK. I suppose it couldn't get any worse.

I'm not sure. I do so much ad-hoc testing of stuff for the forum it could be practically anything.

From compiling stuff for the forum, I have already killed one of my dip based ATmegas. Maybe you burned up a flash cell (or two)...

[quote author=Nick Gammon link=topic=131648.msg990198#msg990198 date=1352511492]

I do, but I don't think I can program it without removing the chip. Or can I?[/quote]

I forgot about the stuff connected to RESET. The Atmega16U2 may not survive the 12 volts or the auto-reset circuit may interfere.

Good news! (and very bad) You have a new oversized USB-to-TTL-serial converter.

Ah well. At about the same time my USBtinyISP stopped working. Not a good day. :slight_smile:

Nick,
I can replace that chip for you. I have a hot air station, and even a couple of chips I think.
The TQFP chips, not the leadless chips - board looks like it has pads for both?

what happens when using usb/arduino IDE?

Need to fill out that question some more cjdelphi, you are mixing things together a bit too much.

I'm curious how on earth he killed one, i've tried (by accident) several times lol

If you've got nothing to lose, I would try replacing the chip.

Put it in a toaster oven on high. It takes 5 minutes in the oven I use.

Once the solder is loose, carefully take the board out. It helps if it's on a tray in the oven. You don't want to bump it against anything, everything will be loose. Using tweezers, pick off the '328. Allow the board to cool.

Using a fine point soldering iron, solder on a new chip. Lots of tutorials on youtube on doing smd soldering by hand.

I havn't done this technique on a Uno yet, but I have done it on a breakout board when I got the chip on backwards. Standard disclaimers apply.

CrossRoads:
Nick,
I can replace that chip for you. I have a hot air station, and even a couple of chips I think.
The TQFP chips, not the leadless chips - board looks like it has pads for both?

Thanks! I think it is leadless. As in this photo:

However that is not the photo of it. (That has the extra pads). The broken one has a chip the size of the lower one (SMD edition) but that particular rev board did not have the larger pads. I don't have my camera handy right now.

cjdelphi:
what happens when using usb/arduino IDE?

The IDE serial monitor will talk to the USB chip if you set up the loop-back test.

However an attempt to use the bootloader times out. As does an attempt to access it via the ICSP header.

cjdelphi:
I'm curious how on earth he killed one, i've tried (by accident) several times lol

I think I must have zapped it with static. I usually touch my aluminum keyboard before reaching for parts like that, but might have forgotten once. Sometimes I get quite a zap from the keyboard when I touch it.

I've done in a couple - loose 12v wire hits a pin, and its dead.