I would say people have different play in their actuators, they have different humidity. Humid climates are going to have more adhesion problems. Some people may have inherently flat and level beds to begin with. The heating could cause the bed to warp differently depending on the air currents in the room! I think that bed frame is a real culprit. I can’t wait to get my composite frame replacement done to see how it works.
@Tone can you confirm the Z height is measured as @Isaac11 referenced?
I think for my A350 it most likely comes down to not having an inherently flat and level bed.
DroneOn, I did something like yours but instead of the 1x/Jx locations I just made plain 20x20 squares (posted on another thread as well), that way you can see the first layer coverage in each calibration location and hopefully be able to adjust the grid/mesh pattern better.
CalibrationTest.stl (71.4 KB) CalibrationTest_40300071.gcode (129.6 KB)
2 Likes
looks good @coppertop_01 nice work.
We are aware of this problem. The strong magnet, by design, should eliminate this issue. However we still received compaints from people regarding this issue. I will have @Edwin take a look at your case.
Can @Edwin look also at my case please.
1 Like
Oh I’ll also try that, thanks 
Presume the squares align to the Ix/Jx locations, looks like they do.
Let me know how they line up, it was kinda a estimated guess…I still need to run mine, weather has been nice and I’ve got tons of outdoor projects.
Hi Issac, we have a good conversation via PM, and your linear modules are defective in some degrees. Please help me to locate the problem more clearly. Sorry for the inconvenience.
1 Like
Can you just share this as a jpg or something please?
And shouldn’t the Snapmaker team have provided us with this and some instructions on how to play about until we get the bed level? I never even knew these coordinates existed until I read this post.
Part of the problem is that you cannot use the touchscreen to just go to a particular spot on the 5 x 5 matrix to raise or lower until you get it perfect, but need to go through the whole cycle again. Life would have been much easier if they had made this possible, maybe even with a larger matrix, without having to fiddle with Gcode strings.
Ummmmm, sorry but I do it all the time, just press the spot on the screen you want to goto and it’ll go back, I screw up the process almost everytime and have to go back.
No you can’t anymore do that on version 1.8 
Try it with 1.8.0. Not happening. One step forward, two steps back it seems…
seriously I pretty sure I had to do it with the 5x5 pattern on mine…doesn’t matter for me, I’m going to be print my test pads and adjust the mesh through Mcode shortly
I apologise - its true it was not working before. but today I re uploaded the 1.8 and tried it again and this time it let me go to the point I clicked on… The touchscreen is a little inconsistent sometimes because I can assure you it did not work when I previously tried to go to a point to recalibrate only that point. Anyway I will now be doing the sensor adjustment procedure and re-running it on auto and we shall see what happens…
My glass bed (first time using glass) arrived earlier than I was expecting. Picture of first print after a manual 3 x 3 calibraiton.
For a first print it was a resonablestart, compared to the first print after a 5 x 5 Auto with the magnetic bed in my previous post. There are a couple of the areas at the top of the print are still too low.
No adjustments to layer height or width either for first layer. I print first layer at 40% speed.
Print comes off easily once temperature has lowered.
1 Like
I still have a problem that the front part of the bed is always printing a little too low.
Here are some manual calibration for 3 x 3 matrix using:
Digital probe and bed at 60C
others < 0 1 2
others < 0 +9.350 +9.250 +9.200
others < 1 +9.600 +9.600 +9.550
others < 2 +9.600 +9.750 +9.850
Reading of probe at these points
0.01 0.02 0.04
0.04 0.01 0.04
0.03 0.01 0.04
Can see there is a delta of 0.65mm between 0,2 and 2,2
3D Module and bed at 60C
others < 0 1 2
others < 0 +8.550 +8.400 +8.400
others < 1 +8.750 +8.750 +8.750
others < 2 +8.750 +8.850 +9.050
Anything else I can do? I find that the front area the Z is too high (presume this is due to the mesh averaging?) so if I lower the front manually via M 421 then what will happen is the back of the bed will get closer as well and be too close to the bed. Any advice @Tone ? 
Where can I get a bed like that one please? Is it a Creality glass bed? Any information would help as I would try to source one even locally.
Also I want to suggest something that might be useful. You might want to try using neodymium (rare earth) magnets to attach the bed. They are extremely strong and will even work through a finger for instance so I imagine a thin piece of glass is as nothing. You will need a handle though as they would be difficult to prise off.