My machine is one of the first produced…number 133 if the machine name is anything to go by.
I finally got hold of a couple of Bondtech Copperhead heat breaks and followed the hot end upgrade instructions further down the Forum. After re assembly of the hot end with Volcano blocks and CHT type nozzles I was finally looking forward to trouble free printing. Upon carrying out the first bed levelling step there was a spark/ flash when the nozzle touched the rear calibration disc. I switched off straight away as it tried to touch the disc again. I stripped the hot end back down and it looks like the insulation on either the heater element or the thermocouple had split as caused a wire to short out.
I have had the hot ends apart so many times trying different ways to get the thing to just print the insulation must have perished. If it had just worked as it should then I wouldn’t be in this predicament.
I recovered the wires with heat shrink, reinstalled and switched back on. The machine will now not recalibrate as the left nozzle stops 10mm above the calibration disc. rebooting has no effect and is stuck in reset …
Any ideas out there good people?
Circuit board? Firmware? other component?
Is there a UK Snapmaker distributor I can take it to?
I have spent a small fortune above and beyond the initial cost just to try and get the thing to work…
Circuit board is definitely my guess. When I took mine apart and traced all the functions to replace the board with a Duet, I noted as well that Snapmaker did something here that the Duet team explicitly warned about since it may occur without warning by a simple fault inside the heater cartridge as well: the calibration function works by grounding a cable (all contact plates in the bed are connected to GND) that is connected to the hotend metal, but as far as I could see there is no fuse and no means of limiting current in case of a short between 24V and the hotend housing.
Thus, i suspect the flash you saw killed either the cable or the corresponding microcontroller input.
When I replaced the J1 controller, I prevented such a case with an added high power resistor and additional diodes.
@elkarto from my point of view, reliability has increased, and usability has improved a lot for sure!
Speed is something I actually decreased a bit since I have those heavy twin radial fan part coolers and want to keep the linear rails in order. Apart from that, it is known mechanical stability of 3d printed parts decreases if you print them too fast.
The Duet is however well capable of bringing the printer to accelerations and speeds that the mechanics do not survive for long… in my initial configuration, I accidentially had put a “0” too much on these values and started a small test print with that. The printer shook like hell, but it did print correctly until I stopped it. Thankfully the linear rails survived that attempted murder
The stock display does not work indeed, but a chinese PanelDue clone with a smaller footprint than the original just about fits into the cutout of the doors.
Let’s see if I can add at least one picture here…:
I also have a documentation of what I did, but it is in german. Write me a personal message if you want a link.
Edit: one more note: input shaping is currently being reworked by the Duet team. I actually disabled it for now. Basically it works fine and I could nicely optimise it with the small cheap accelerometer board the Duet team suggests. The result (just one part printed however) looked IMHO slightly better as the stock firmware, but not much… the accelerometer told me that resonance frequencies on the J1 tend to shift quite a bit depending on speed which is difficult to compensate. Lowering speeds helps more However, on parts with a lot of small elements RRF3.5.0rc1 I used for testing had some issues which resulted in a layer shift. As soon as that is solved by the Duet team, I will activate it again. The price for using beta firmware
Sorry for the delay in replying. I replaced the temp sensor, reinsulated the heater cables, turned it all back on and it seemed to work! It has been speedily and steadily printing ever since. Why the Duet board? is it better or does it have more functions?
@Printmuppet Wow, you were lucky then, I guess. In theory, the controller inputs survive a 24V surge - practice looks often different however. Congratulations
The Duet board runs RepRapFirmware. Much easier to configure and more capable than Marlin IMHO - and: no compiling needed which literally saved me days when I tried to find out the correct settings for the J1 and created the calibration macros. Any setting you could think of is changed by simply editing a config file (which basically contains just simple gcode) via the web interface and reloading that file or rebooting the printer. Macros are created and edited in the same way. I admit it: I love those boards, and they are reliable as hell.
One of the things I miss in the J1 is how closed it is at the software level… despite having released the marlin source code. On my previous printer I started using klipper just before I got the J1, and I liked the concept.
Maybe it’s the moment to spend some time implementing it in the J1. I need to evaluate the pros and cons.
Please post your progress here. I’m considering swapping out the mainboard with a MakerBase SKIPR which has a CPU and arduino already set up for Klipper. I’ll likely add a hefty side mounted environmental fan as well. I just need to dig through the marlin config to get the steppings, etc. But your solution may be better. Certainly less hardware work.
Yes, the printer not being open and not really supporting open communication standards without was also a reason for me to do that. I am really grateful for @macdylan to have created smuploader for us, but that tool is a can-opener to a nonstandard communication protocol after all. If Snapmaker for some reason decides to go all Apple and blocks that tool, I would have a printer I do not want.
The released Marlin code is nice as well, and it did at least help me by providing the output characteristic of that preamplifier that is built into the hotends (otherwise I would have thrown these Preamplifiers out and used PT1000s instead). But I believe the Snapmaker management only released it more or less grudgingly to fulfill what the Jadelabo team (luckily!) had promised - which I see supported by the fact Snapmaker does not update the code they released despite having released several firmware versions after that.
And knowing the Marlin source does not really disclose the printer… a lot of tasks (including any Wifi communication) are handled / executed by the printer display which as far as I understand is more of an Android-based custom Octoprint than being a simple display. And all that happens in there is closed-source.
That being said: I am looking forward to see a J1 with a Klipper board inside
But enough offtopic talk for now, we strived far from Printmuppet’s thread here
We recommend contacting our technical support team to assist you with this issue. Have you already reached out to them? If not, please do so, and they will help you resolve the issue: Support Ticket Form (formcrafts.com)
If you have any further questions, feel free to let us know.
However, on parts with a lot of small elements RRF3.5.0rc1 I used for testing had some issues which resulted in a layer shift. As soon as that is solved by the Duet team, I will activate it again. The price for using beta firmware 