Freezing Duet with dc42 firmware Mini Kossel

Hi,

I've recently bought the duet board from think3dprint3d and migrated my Kossel Mini from RAMPS to Duet using dc42 excellent instructions; the Kossel itself was from a kit from builda3dprinter.eu but there did not seem to be that much of a difference. While most things worked out perfectly, there is one problem persisting: the Kossel stops mid print -- while it managed to finish one ca. 90 minutes print, everything since then has stopped with a frozen firmware within the first 15 minutes or so: the lights on the duet turn off for a few seconds, as does the extruder fan. Then lights and fan turn on again and the duet does a restart, the print is stopped. Last time it happened right after starting to print.

I do have the 5V_EN jmpered (it came like this out of the box). I do not use an LED PSU, but an ATX PSU (bequiet! SU7-300W, which should give 18A on one 12V-line). Firmware is RepRapFirmware-1.09c-dc42.bin. I sliced with slic3r and cura, same problem. I attached my config.g.

I'm not sure if this is just a broken board or if there might be some other reason for this behaviour. Can anybody help me or has an idea where I can look further?

Thanks a lot
Jost

If the lights on the Duet and the hot end fan turn off, then the Duet has lost 12V power. Here are a few possible reasons:

1. The most common reason for intermittent power is that the 12V input screw terminals are loose. In particular, if you tinned the 12V power wires before putting them in the screw terminals, the heat due to the current flow will soften the solder and they will become loose. Check them for tightness every day (with the power off) until they stay tight.

2. Do you have the required dummy load on the 5V output of the ATX power supply? If not, then you are likely to get an overvoltage condition on the 5V output, and the ATX PSU will shut down.

3. You could also check the soldering of the 12V terminal block pins on the back of the Duet, but I have never known this to be bad.

Edited 1 time(s). Last edit at 07/25/2015 01:04PM by dc42.

---

Large delta printer [miscsolutions.wordpress.com], E3D tool changer, Robotdigg SCARA printer, Crane Quad and Ormerod

Disclosure: I design Duet electronics and work on RepRapFirmware, [duet3d.com].

Thanks a lot for your help, dc42! Going through all of your notes, it now managed to print for 33 minutes, much longer than the last 5-10 attempts, but then it broke off again; another attempt after another retightening of screw terminals broke off after 15 minutes again. This is what I did:

1. My 12V wires aren't tinned, I use ferrules. Nevertheless, I retightened the screw terminals

2. Indeed I did not have a 5V dummy load. My PSU did not seem to have any problems without it up to now, but I nevertheless connected a 12V fan to the PSUs 5V out, just to be sure.

3. I have taken a look at the soldering of the 12V screw terminal, but it looks fine to me.

4. I measured temperatures on the board with an IR thermometer, but nothing is going above 40° C.

I wondered whether with the heated bed the overall load might become too much -- the ramps had one 11A and one 5A input, so I connected one 12V rail of the PSU to each; now I use only one rail, which is supposed to serve 18A. But when the duet powers down now (for about 5-10 seconds before turning on again), there is absoluteley no visible change to the movement of the fan I use for the 5V dummy load, so I suspect it is not the PSU shutting down, and I also don't see why it should shut down after 30 minutes without getting even remotely hot.

For now I keep retightening the screw terminals, maybe the vibrations of printing require this, and I'll try printing without heated bed.

Is there anything else I should look after? Could it somehow be a faulty board?

BTW, dc42: the partial pieces that came out of my printer show a real leap forward in quality. I originally went for the duet to make things faster, but now I'm completely amazed what a difference the duet and your firmware make, even using my mechanical probe with the auto calibration. Do you still have those IR probes on stock? I still hope this will stop crashing some day, and then the probe will definitely be next

Assuming you are using the permanent 12V connection for the hot end fan (see [reprappro.com]), then the 12V input is connected straight to the 12V fan pins. Therefore, as the fan is cutting out, it really looks like the 12V input power is failing. As you have checked the connections, I suspect the 12V rail on your 300W power supply is dropping out. Most people use a 500W or greater ATX PSU to power a Reprap (or a good quality 400W+ one such as the Corsair CX430M), not because they needs that much power but because lower rated PSUs generally have inadequate 12V output current. Depending on how much current your heated bed takes, you could be drawing very close to the 18A rating when the heated bed and hot end are on and the steppers are moving.

But of course you should also check that the ferrules are crimped tightly on the power wires, and that any plug/socket connections between the ATX PSU and the Duet are sound.

---

Large delta printer [miscsolutions.wordpress.com], E3D tool changer, Robotdigg SCARA printer, Crane Quad and Ormerod

Disclosure: I design Duet electronics and work on RepRapFirmware, [duet3d.com].

I guess you were right about my PSU. The dead simple solution (to which I was pointed by Tony from think3dprint3d) for everyone encountering this problem is removing the 5V_EN jumper and giving a separate 5V in, be it over USB or ATX. The short power fails on the 12V rail, if they still exist, don't hurt as they're obviously too short to be noticed by the heaters.

Thanks a lot for your help! I will go for a stronger PSU in the long run, although for now at least the problem seems to be solved