Leveling / First layer Problems

No issues with adhesion… Also go slow on first layer… See some of my settings…:

Don’t forget you need to level at working temperature, Bed and Module.

Also extruder calibration is important. Those both plus patience (= go slow on the first layer and use a brim) have a greater impact on print quality than the type of filament you are using… Printing with Snapmaker PLA right now - no problem at all…

No, I don’t agree, you need to heat the machine to working temperature first. When you start the calibration the heat will be turned off as it can interfere with the induction sensor. The team is looking at solutions to heat in between locations, it has been verified the bed changes especially the first bed (around 7600 units), the new stronger frame should warp less but the temperature does matter. Calibrations of any item are always at a certain temperature.

In case you decide to move the sensor, here is the guide to do so:

Through experimentation many people who have been unable to get anything close to a level bed have, indeed, gotten something useable by heating up the bed first. How do you ignore this experimental data in favor of your theory? I pose to you that perhaps your single bed and platform structure is not indicative of many others, and that there may be more than one “best” way.

Not interested in proving you false, I"m just saying that you’re making something binary that’s not necessarily binary.
Heres how we used to do this:

“having heat on can only make it worse,”
Some people get better mesh with heating up bed before calibration"
“They got lucky!”
I am not trying to PROVE you are wrong OR right. ALL I am saying is, some people do get better results, and here you are saying “No, don’t do that”, for some reason.
I tell people, Why not give it a try, see if it works better for your situation.
There is not always a best or only answer.

All that said we’re all hoping that this firmware patch will be the magical one that solves all the problems. lol

I guess you weren’t around a year ago.

Heating the bed is a known workaround for the thermal expansion of the platform, and is an essential part of the calibration workflow for some people here on the forum.

Here’s my results on the matter. For me, thermal expansion was much smaller of an issue than the deformation in the webbed platform.

The correct way to heat and level the bed was to connect a serial device, send the M140 SXX, wait for preheat and then calibrate the bed (the heat is always on like normal printing).
I know the discussion about electromagnetic fields but this worked.

You could believe me, the heat stays on if you are on with serial connection. Try it or believe it.
You could call it bug or feature.

Here is what I believe and observe, I know what you are talking about but you simply did it wrong.

You have to set the bed temp first and then perform the calibration.

Video about
Dropbox - VID_20210710_214048.mp4 - Simplify your life

The video in my previous post works now, have a look.

1 Like

Shame the camera work doesn’t work :wink:

Support (tracy) walked me through a hot calibration in the last few days at their request to troubleshoot the 1mm slope on my plate.

  1. @WilliamBosacker is right (like he is 99.99% of the time) if one issues a G1029 it most definitely turns off the heating as soon as the process starts, if one is connected with a serial console one sees this in the serial output and can be monitored on the touchscreen.

  2. @xchrisd you started a 5x5 calibration from the touchscreen - in this scenario the temperature is indeed not turned off and will remain at whatever temperature was set with M140 S70 (this is what tracy asked me to do via serial, I assume octo does something similar?)

This means it is currently impossible to do a hot calibration above 5x5.

I don’t know why there is a difference - for me it confirms that what the touchscreen does during calibration is not identical to what happens with G1029 commands. This also explains why the head and base movement to get from home to first calibration point are sequenced differently when using G1029 than when using touchscreen. This raises even more questions in my mind as to what else is custom by the touchscreen and not accessible from G commands.

1 Like

I´m sorry if i am wrong but i have never read about a higher grid calibration than 5x5 in this topic.
I wanted nothing more than saying the same like in this post which is posted before from Tracy.

Octoprint has the same terminal alias serial-interface or serial-connection like Luban.
The touchscreen works this way too.

Unfortunately i have no idea to get it working at a bigger grid.

Hi guys, i got back from vacation and decided to invest some time today, to run a 11x11 calibration. after a few problems with connecting luban to the printer (all solved), i did the 11x11 leveling. After that i thaught i should try to print a test print. Now the results of the test print are as you can see in the picture. I do have to be honest and say that i left the printer heating up and came back to late to stop it befor doing so much damage. So i know this is also a bit my fault but still i don’t quit understand why the printer did what it did.

Further information:

while the printer was leveling (11x11 grid) it sometimes poked the printingbed.

i was using the instructions @WilliamBosacker posted on another thread (Required 3D Printing Calibration). (NOT saying it’s anyones fault)

i carefully adjusted the Z-offset at the end of the leveling process.

If it poked the printingbed enough to move the bed, it’ll give you a really poor mesh which could cause this. when you say “poked”, did it visibly push the bed down?

Yea, agreed (on the poor resulting levelling data). Sounds like the sensor is too high and needs to be lowered.

That’s the cool thing, it’s not your fault. It was assembled wrong and this is a common-enough problem support had to write a self-help article about it this last January.

Thanks for the advice. I will have a look at the sensor and the manual on how to adjust it.

I am not quite sure, but i definitly thought about this scenario too. And i think especially when handling the module alot because you change between 3D, laser and or the milling moduel, the chance is quite high you accidentally push the sensor out of position. So it’s probably not a bad idea to add this scenario to the calibration page.

1 Like