I tried to print an object that used almost all the length of the A350 bed. The side of the very front end displaced itself. At that side there should be 2 identical cut poligons. For one of them the printed line got displaced and the skirt in that region was also displaced. I then realized that in that corner the printed filament didnt stick to the bed as well as in the other parts. Bedleveling was done the day before.
I was wondering what could cause this problem? Z-Offset should be fine. There wasn’t any problem for the last print. Could it be I printed the filament to cold? I have the A350 in the enclosure. Sometimes there is strain on the filament. Could this also be a cause?
You are experiencing a very common issue with the snapmaker 2 which is often discussed on the forum. There are a lot of good threads about this, and I am not the best person to ask about it, but suffice it to say:
The machine is not very flat, and the calibration grid by default is not very thorough.
to start you can try to adjust the calibration grid in the settings to 5 x 5.
you can try to pre-heat the bed to a little above the temp you want the bed at, then do the calibration (which shuts the heater off) and when the heat warps the sheet as it does, it will compensate. id let it sit above temp for 10 minutes or so first.
beyond that, theres other tricks to do, but if you are a starter thats probably the most reasonable path forward.
if you are kinda used to the machine i can show you how to do an 11 x 11 heated calibration to see how it goes.
also do yourself a favor and look around on some of the threads in the forum about flatness, leveling, adhesion, there are a billion little things you can try to make it better.
As a general rule of thumb beyond that… some good tricks for adhesion is start the print bed hot, start the first layer hot, start the first layer slow and a wide line width.
you can also look at a glue stick, magigoo, blue tape, hair spray… so many different methods out there.
I tried 5x5 about a week ago. The first layer was there after however printed in a strange way. Do I need to do any adjustments if I use the 5x5 pattern?
I would be interestet in the 11x11 heated calibration.
I will read on further about the subject of bedleveling. Seems like the last reads weren’t enough.
Well, as you see, the sheet is not consistant, so some areas will print better than others.
The more calibration points you have, the better it is going to know how to accomodate for it.
The heated bed causes the sheet to shift further, hence why I recommend a pre-heat to the bed prior to running the 5 point calibration.
Why not give this a shot first? You don’t need to do anything specifically different for the 5 point calibration than the 3 point, other than assume that the machine has a more detailed concept of how to work around the lack of it being flat.
Honestly i was pretty satisfied with the 5 point pre-heated calibration for a very long time. I am only now looking to expand it because I want to challenge myself.
Over time you will learn to compensate for the topography of your specific bed by deciding to place certain objects in certain areas, such as on mine i have to put large objects in the center because the the top and bottom are different and if i try to print something large where it doesnt have enough consistant space it has a higher chance of failing due to partial adhesion failure.
so i will often take smaller parts and move them to where they would seem to be easiest to print based on my knowledge of my bed.
if i dont have any large parts i always start from the middle.
Your bed, however, may have a completely different plane to it, so only experience will tell you.
That’s what I thought as well. That’s why I dont understand why the first level was printed differently. First I thought it was the filament acting weird. After the second filament gave me the same result I calibrated again with the 3x3 matrix and the print behaviour was normal again. Any idea what could have happened? May be I messed something else up in the 5x5 calibration? Z offset?
If this is with 3x3 grid, your machine is a good one .
I clean my bed before every print with alcohol, because of dust (have an enclosure too).
The filament spool should be protected against dust, dust could cause a clogged nozzle, never had a problem because of this…
If you have adhesion problems, Printing slower and maybe printing hotter is the way to print.
Gluestick also helps but the solution above is better.
What speed do you print the first layer?
I would suggest 10-15mm/s.
Happy to hear that. I must have underestimated the effects of all the different variables like heat, print bed, etc.
What kind of alcohol? Good to know about the dust.
If I print in the middle the adhesion is good.
Temperature initial layer: bed 70C and nozzle 200C (depending on filament according to temperature tower) but may be I could raise both? Speed 18mm/s. So I could lower that to 10mm/s. I guess travel speed for initial layer shouldn’t have any effect on the quality of the layer?
Here is the situation after I autocalibrated with a preheated bed:
first layer print temp: 70C
preheating for 15-20Min at 75C (distance with 70C was to far, with 80C to close
print first layer at 10mm/s
I think you might be reaching for the stars a little hard right now with what you will be able to get away with, at least until you have a lot more time under your belt.
However…
You can try to do a custom heated 11 x 11 calibration. It is kind of wonky especially to set your z-offset as the touch screen will not assist you for this process. The touchscreen jog menu does not allow for very fine adjustments like luban does, but you could always adjust the z-offset as a whole afterwards as well from an active job.
This routine will sample your bed in 121 places. It appears to be the most we can get away with in terms of processing power.
At this time, I hesitate to recommend anything beyond a pre-heated calibration as opposed to an actively heated one, I am still testing how the heater affects the sensing range. You are of course welcome to try.
This is a complex problem with many factors which the community is actively sort of working together on to resolve in various different ways. There are some custom firmware adjustments being made, some accessories to help level on a glass bed, replacement build plates being designed and tried, and many other things being tried right now to improve this issue.
You may just be looking for perfection that might not be acheivable with your bed. @moserroger you’ve never posted any pictures of your setup.
Share some photos of your machine. From front at level of bed and at height of z-towers from front titled down are great. Just make sure there’s nothing we’re missing that’s funky about your machine or assembly.
Thank you MooseJuice for clarification. I wasn’t aware this was so complex and that there was so much being done about this issue. And yes, as the Snapmaker seems more complete or better built than many other machines I somehow assumed that all is already very well adjusted and will just work like a charm.
I will also have a look at the other thread. All looks very interesting. I might have to get back to you about the z-offset tough.
I wanted to try your 11X11 calibration. How would I save the adjusted z-offset? Would it be possible to adjust z-offset through luban and save it by g code in the second makro?
The second makro doesn’t seem to work for me. The toolhead isn’t homed.
When you are done with the first marco, you have to use the touchscreen from that exact position to manually set your z offset, do not move x or y at all
you can also use luban to get a finer adjustment.
afterwards, you execute the second macro.
i dont know why it wouldnt home from there, but those other lines before that are what saves your offset based on the nozzles position and finalizes the calibration.
what did luban say when you did it? can you paste the text from the console for what you did afte rthe first marco finished?
keep in mind this does not work like the touchscreen calibration - and the z-offset value you have set on your screen before doing so will remain at that number (if you had it manually adjusted to anything beyond zero) but you can still use that to move it up or down during a print from there.
You mean the message after the macro 1 finished or after I started macro 2?
It seemed to me that nothing happened when I ran macro 2.
I will redo both and let you know about what is written on the console. I tried to do a test print and had the impression that z-offset was set to the position at which the nozzle was at after macro 1 and not the position I moved it to. I only moved Z not X nor Y.
.
