Major damage to Sanguinololu on my Mendel90

The weekend started out well with some successful printing at low layer heights with my new ABS. Then after completing 3 out of the 4 prints I was working on, the PCB Mk2 Heatbed suddenly wouldn't come on any more.
A quick inspection of the electronics revealed the problem.



My fault completely. Nophead warns in the instructions that the heated bed connector isn't rated high enough for the current but I'd already built mine by then and it seemed to be working perfectly, so I left it. And now I've paid a high price.

After that, things went from bad to worse.

Rewiring the bed to connect directly to the body of the MOSFET and ground, as nophead suggests, didn't make it work again so I suspected that there was something else damaged - possibly the MOSFET itself.

I started taking voltage readings from various places on the board whilst sending commands from the laptop to turn the heaters on and off to test which lines seemed to be working and then disaster struck. Whilst touching the probe to the 12V pin on the IO header, it inadvertently made contact with the adjacent 5V pin as well. I knew straight away that something bad had happened because the laptop instantly shut down.
Fortunately, I was able to restart the laptop by removing the battery for a few seconds, but it appears that both of the USB ports on the right side are now dead.

Back at the printer, things weren't looking too good either.

• The board connects normally and displays the Marlin start-up message.
• The bed thermistor reads correctly, but the bed heater still doesn't come on.
• The extruder thermistor reads 109C and doesn't change, but the heater does work.
• None of the motors move
• The voltage regulator and several of the pololus get very hot.

Removing the 1284P and replacing the old Atmega 644P fixed the nozzle thermistor problem, but none of the other problems, so I suspect that there is more damage somewhere on the board.

Any suggestions as to what I should do next? Try and rescue the board or abandon and start again?
Apart from the Atmega chip, which I'm pretty sure is damaged, what else is most likely to have suffered from a 12V shock down the 5V line?
I'm concerned that the pololus may have been damaged as well.

I would appreciate your advice.
richgain

12V on the 5V line will kill the pololus and damage the Atmega. I managed to do that on my previous RAMPS board. Same result - pololus get very hot and cannot move the motors. You will have to replace them.

The wires for the heated bed may have shorted and killed the MOSFET. Luckily MOSFETs are cheap.

---

Help improve the RepRap wiki! Just click "Edit" in the top-right corner of the page and start typing. Anyone can edit the wiki!

Nightmare! Yes putting 12V on the 5V rail can damage any of the chips connected to the rail, which is all of them. I am surprised the FTDI still works and the 1284P mostly worked.

Try removing the Pololus that get hot. Maybe the others have survived, but its a long shot. They probably failed in a different way.

---

[www.hydraraptor.blogspot.com]

Up and running again now.
I bought a new board, chip and steppers from Think3DPrint3D and received them 36 hours later. Excellent service at very reasonable prices.
I plan to replace individual components on the fried board and see if I can get that one working again to keep as a spare (or make someone else a printer!)
I did have one moment of panic when the heated bed still didn't work even with the new board and nophead's suggested wiring. Then, late last night, it suddenly dawned on me that when he says connect the heated bed directly to the body of the MOSFET and the PSU, that didn't necessarily mean Ground on the PSU (I've edited the wiki now). I tried changing it to the MOSFET and 12V on the PSU and 'Bingo', I was back in business. So the one on the old board might not be burned out after all.
richgain