Wiring problem

  • So my printer has been running perfectly happy for almost a year now, about a week ago I smelled a bit of burning plastic, and started to get warnings that Drivers 0,1,2,3,4 were over temperature and the print would come to a complete halt. I figured I would check the wiring and ensure there were no problems. Well, I was shocked when I detached the ATX PSU and found the 12v lines on the breakout board charred. Upon detaching the breakout board from the frame of the printer, the positive 12v lead just fell out of its connector.

    To power my printer, I was running from a 450w EVGA ATX PSU into a breakout board, and from there I connected to a 12v->24v Step up converter before feeding 24v power into my duet board. I had seen the warning on the duet wiring guide that all connections should be in crimped ferrules and that you should not use tinned connections. But... I didn't have the equipment to properly crimp the included ferrules, and I figured that if I clamped the screws down hard enough I would be okay. It looks like the step-up converter I used had tinned ends though. It's my belief that over time the connection was heated by the high currents passing through and the solder or tinning began to run. As the solder began to run, the connection became poorer and the resistance increased. Eventually, the connection failed.

    Burnt Out Board

    To repair the problem I ordered a ferrule set with crimps and reseated all of my screw terminal connections on the printer. I replaced the breakout board, and I cleaned the contacts on the ATX connector.

    This week, however, I have run into a problem with the new breakout board burning out as well.

    Second Burnt out Board

    I don't want to keep throwing parts at the problem and I want to get to the bottom of what is going on so I had a few questions for people who are more knowledgable than myself.

    The first thing I want to do is ensure there are no shorts anywhere in my system. To do this I was planning to use my multi-meter to check the power lines coming off the duet mainboard. Specifically, I was planning to check the hotend heater cartridge, and the hotbed by checking the resistance across them from their wires that currently connect to my duet. Do you think just checking resistance will work? Do you have any idea what resistance I should be expecting? (E3d high precision 24v 40W and Anycubic 43188-247109 12v/24v ultrabase and heated) Obviously I expect to find some resistance and no resistance would imply a short, but what I mean is, do I need a particular resistance to be confident there is no short?

    Next, I plan to replace the breakout board again, and I am concerned the step up converter may have been damaged in the first incident. If so my guess is that there may be a short or problem with the step up converter that caused the second problem. Do you know if there is any way I can safely test the step up converter?

    Lastly I was wondering if there is any way that I can check to see if the ATX PSU needs to be replaced? or if the connections might still be okay. The 5v line is still working and the duet can power up, but I cannot turn "on" the power from the web interface anymore, which is probably cause the 12v power line has a charred connection. (I realized there was a problem with the power again when I saw that the web interface listed the vin as 1.7v and had the red off icon indicating the psu was turned off. I tried to turn on the PSU and when it wouldn't do so inspected the connection and found the second problem.)

    p.s. I want to mention that I do have a smoke detector mounted above the 3d printer, and especially with covid, I am almost always in the same room as the printer.

  • The ATX connector pins are rated to 6A. I think you're just exceeding that limit, or getting close to it. The other connections (screw terminals) aren't the problem.

    If you're using 150W, you're pulling at least 12.5A out of that pin. And since you're using a step-up converter, you would be even higher.

  • @TDK Thanks for replying TDK. The printer did run in this set up for almost a year so it seems odd that it would blow in less than a week with the new board. Though the old I breakout board had the two 12v lines tied together and this one keeps them separate.

    I checked, and it looks like the hotend should only be pulling about 1.67A, but I can't find wattage or amp draw for the heatbed. Do you think setting up the hotbed on a separate mosfet directly from one of the 12v lines would work to reduce the amps across the connection to a safe point?

    Alternatively, I could bypass the breakout board for the 12v line going to the step-up converter and splice/solder the step up converter directly with the 12v line from the psu.

    Do you have an approach you would recommend?

  • @Grudairian I just wanted to add that when I was looking through the questions on the amazon page for my hotbed, someone mentioned that the hotbed should draw about 25 amps(at the 12v config). Looking on the side of the PSU it looks like the 12v lines are rated for 35A. So it seems like splicing the step up converter directly to the 12v line might be a good idea and would get around the 6A limitation of an atx connector.

  • @Grudairian said in Wiring problem:

    Do you have an approach you would recommend?

    Anything that gets rid of the 6A current limitation. If your heater pulls 40W, you're pulling over 3.33A from the 12V line. More because your step-up converter can't be 100% efficient.

    Using directly soldered wires rather than the ATX connector is one solution. Just gotta be mindful of other current limits in the system.

  • @Grudairian I double checked before I started splicing wires, and the step-up converter is intended to handle 24V at 10A, which means that it should need about 20A at 12V input to maintain that. If a single atx pin is only intended to handle 6A at 12V then the wire is probably not rated for too much more than that either. I checked and the wires are 20AWG from the PSU. So now I am thinking that the best course of action would be to take 4 x 12v lines off the PSU and solder all of those together with the positive line in on the step-up converter. That should provide 24A at 12V for input and should be enough power to keep the step-up converter happy. For the 4 12v lines, I figured I would sacrifice the 8 pin cpu connector. I also figured if we are allowing for that much power to go in, we should also allow it to get out and figured I would also splice grounds to be equivalently rated.

    Thanks again for your reply, it has been good food for thought

  • Moderator

    Just throwing this out there, but why not get an actual 24v PSU at the wattage you require? There are Meanwell units with PS_on functionality as well.

  • @Phaedrux said in Wiring problem:

    Just throwing this out there, but why not get an actual 24v PSU at the wattage you require? There are Meanwell units with PS_on functionality as well.

    How dare you, that sounds far too much like common sense to me 😀

Log in to reply