SyntaxTerror

It's simply a small PCB that adds additional fan drivers so the fans on the second head can be controlled individually. As the main board lacks any (unused) buffered outputs from the microcontroller, I put together a little board with the missing driver transistors.

Would it help to add ferrules to each connection? (Specially the heater?) Hmm.. it might? Is there a lot of stress on the wires in the UMO?

If you have some fine wire cutters or a scalpel, you should cut out U2 and clean the circuit board with a natural hair brush and denatured alcohol. That'll make it possible to inspect the PCB for delamination and/or burn damage - then we can see if your board is repairable.

It certainly looks like there's been some heating from the blackened look of the screw head. What's going on there? Is the cable even fully seated in the terminal? There's a lot of copper showing. If I were you I'd snip off a centimeter of all the cables and reinstall them. I'd also do my best to clean the terminal block to ensure a proper connection. It's true, the cables aren't supposed to be tinned. That's not to say it won't work with tinned ones.

Damn, I guess the crossed fingers didn't do the trick. Did you remember to scrub the gold fingers on the display flex cable fingers also? Two sides to a connector and all. You don't happen to have access to an oscilloscope, do you? I wouldn't mind having a closer look at what's going on with the display signaling. I find it somewhat puzzling that it would just stop working while the rest of the printer keeps on trucking. Did you ever manage to get something printed directly from SD? Also, have you tried updating the firmware (in case something's corrupted)?

Shouldn't be a problem

As long as the bed is kept close to the glass temperature of the filament in question, there's always going to be an increased risk of warping (for obvious reasons). Getting the lower layers to stiffen up always helps prevent the next layers from pulling them off the bed.

Oh god, you're right. I must have some brain issues here. And I do think the heater resistance is rapidly moving upwards due to excessive heating, I figure that's the only reason anything can be working.

Ahh excellent! Happy to hear everything worked out in the end. It is a terribly odd coincidence that it didn't work to begin with on two separate systems. Linux in particular usually does not suffer from driver issues like this. But hey, who cares. The firmware's in now

Do you know Arduinos? Basically they're microcontroller breakout boards where someone was brilliant enough to take away the hassle of dealing with a dedicated programmer. They did so by giving the microcontroller a bootloader that allows the user to reprogram it via the serial port (which is not normal behaviour). So basically there's a USB->TTL serial converter that lets your PC interface with the microcontroller and upload new code. That's exactly what Cura wants to do with the board. If the bootloader is borked, then we won't be able to ever do this. That's where the ICSP port comes in. That'd be the normal way to program the microcontroller if it didn't have a bootloader. I don't think it's drastic, only inconvenient. Hell, that's how things are programmed under normal circumstances. It just requires you to have a programmer of some sort. I must admit I'm simply not certain at this point. Just.. too damn rusty with drivers. More thinking required on my part Nah, don't worry about that.

Definitely worth a try. You never know! Regardless of whether or not it works, that's a super useful post for the tinkerers. Off topic - why does the forum insist on indenting the first line of code?

I can't say I know the supplier, but I figure they look trustworthy. What's printed on top of your individual circuits is unimportant, and not necessarily the same. There's date codes, production facility codes, lot numbers and such so they may differ. It's the right part you linked to, so go ahead and grab it.

I would be tempted to say to re-flash the boot loader through the ICSP header, but if it is broken, how was the initial firmware loaded? Well, it certainly wouldn't be the first time an arduino bootloader got screwed up for some unclear reason. In my time playing with various Atmel parts and Arduino, it's happened to me at least a dozen times. In any case, I think flashing from the onboard headers would be a very good idea. Was actually going to make a how-to concerning that at some point since there seems to be little step-by-step instruction available for the UM2 board. Hmm. I must admit I'm very much out of practise with driver binding these days, but I figured since we get lrwxrwxrwx 1 root root 0 Apr 20 20:26 driver -> ../../../../../../bus/usb/drivers/cdc_acm something must be going right? The UM2 board is dead as a hammer when only connected to USB.

A qualified guess would be that the heater is damaged. Dumping too much power in any resistive load usually ends up with the resistance starting to increase, which then lowers the overall power dissipated due to reduced current.

Well there's your issue, I guess. When @gr5 posted the PID for 40W, it was 40W on 24V. What you have is in reality 80W since you've doubled the rated voltage. I'm actually halfway surprised the heater is alive still?

Basically, I just figured it would be a way to see if the USB interface is in fact functional. If you can print directly from Cura via USB, then it must be. In which case you might have a bootloader problem on the ATMEGA2560 on the main board.

Hmm. I think I'm barking up the wrong tree. It's odd though. Has it ever worked for you, I mean is it the first time you try uploading new firmware on it? Also, have you ever tried printing from USB instead of the SD card?

I wouldn't order any sort of IC from some Chinese vendor. It's incredible how even individual components get faked. They're so cheap I don't see what they point is in the first place unless someone is buying bulk (in which case they should know better). I guess somebody must be able to make money that way somehow.

Okay, can you go into that dir and do ls -la ?

Yeah just rmmod cdc_acm modprobe cdc_acm ... just to be on the safe side.

Alright, can you try plugging it in again and checking what's in /sys/bus/usb/drivers/cdc_acm/ Assuming you're getting the same "cdc_acm 7-1:1.0: ttyACM0: USB ACM device" there should be a directory called 7-1:1.0 there. Otherwise I'm guessing definitely a driver problem.

Have you tried unloading and reloading the driver?

Odd. Why doesn't it register?

Well to me it looks like your machine is not loading the driver. Try: sudo modprobe cdc_acm lsmod | grep cdc Then plug the ultimaker USB back in and see if dmesg gives you a little more this time.

Hold on, you didn't get [10870836.371712] cdc_acm 2-1.2:1.0: ttyACM0: USB ACM device [10870836.372227] usbcore: registered new interface driver cdc_acm [10870836.372229] cdc_acm: USB Abstract Control Model driver for USB modems and ISDN adapters ?