Initial Calibration Feedback

The Second gen snapmaker is quite the unit, however in inherits a few annoying bugs from the first snapmaker.

The first is that calibrations pre-print should be done with an at-temp buildplate. I used to trick my old snapmaker into doing this over the USB connection then quickly calibrating it while the plate was still hot. The new one auto-calibrates before prints, making this tougher to pull off.

The snapmaker should bring the plate up to full temp THEN do the calibration, as the plate will have reached it’s final shape. Then it can go check stops and such while heating up the hot-end.

The second is to have some sort of filament guide, retraction spring (you can find these on thingiverse), as long moves on the Z-axis can cause the filament to fall off the spool over one of the z-axis linear modules.

I’m sure I’ll have more, but this is my immediate feedback.

3 Likes

I will also add the fact that we need lower step for Z axis. 0.01 should be an option.

And I still don’t understand why the printer need to run 2 times the 9 points auto calibration. 1 for the calibration and the other time just before printing

I got my 2.0 A350 today and have been doing some practice prints. I have found that 9 points on a big build plate just doesn’t go well enough, one half of a print will be too low and the other too high and we are talking about a 80mm round print. I also agree with Rhynri that the calibration should be done when the bed is heated. I also like the idea by Isaac11 to have the option of 0.01mm for the Z axis. I like that it can be live tuned.

I would say that the calibration at the start of the print is necessary in case the print surface has been moved about. I think though for the A350 there should be a 7x7 calibration. 3x3 is not good enough.

Another issue I have found is with the magnetic build plate. It can pop up in spots which suggests it is not strong enough sticking to the magnets on the bed. I would suggest a PEI sheet like the ones by Prusa. https://subtle.design/products/everflex are planning to let people order a custom size version so you can order the size you need.

Little bit of teething issues but im sure it will get better.

2 Likes

I second having a much tighter grid for the calibration. Doubling it at least seems warranted with the size of the build plate.

I did figure out that if I set the build plate to 20C above the temp I want it at it’ll cool to roughly the right temperature during the autocalibration step of starting a print and I’ll get fairly even results.

1 Like

Also having the possibilty to add trusted computer to avoid the need of pushing accept button for wifi connexion

4 Likes

Yes, 100x that. I understand why you want to accept the first time, but every time thereafter… no.

1 Like

I also see some warping between these 9 points, so a denser net of probed positions should be an improvement. And it should be done once before the print starts, not before heating up the buildplate.

At the same time I would prefer to have it configurable. It looks like the probing is completely done in firmware, no G-Codes involved. Is it possible to move the calibration out of the firmware and inside the G-Code files?

Hey Everyone,
I notice that my calibration test on my A150 only test about 75% of the build plate
Is anyone else having this issue?

@thomasl, you could add some Gcode before every print with Simplify3d, other Slicer software or with WordPad yourselv.

I have no snapmaker 2.0 yet, -

G29 or M420, M421 look not bad
You have to try and figure it out, if you could check more than 9 points

The point is the firmware makes the first calibration, then comes your data for printing

From the looks of things here it seems you can make an arbitrary grid to probe.

I have the A350. I have had a successful small print n the middle of the plate, but anything large I try does not stick outside of the middle. Seems to be coming in to high in those areas. When my print head is calibrating it does not touch the build plate when going to the 9 points. (I do bring it to the bed at the last point as in the video).
All this is frustrating as I sold my original SM to buy the largest one just for the real estate. It sound like many people are having this problem. Is there a time line for a fix?
Thank you in advance for your help.

Hi, when auto calibrating, the print head doesn’t touch the build plate as designed for the first 8 calibration spots as designed, the last spot you have to manual adjust the z-offset so that the head is lightly touching the heat bed.
Depending on your prints, sometimes the prints don’t have a strong foundation so you need to add supports. You can customize and add supports under the ‘Profile’ tab.

Hello,
I wish it was so simple. I have tried the auto calibration, I have tried the manual calibration does not matter. It has not even started on the print, it is just doing that ring around the base and in the corner it is almost 2mm above the plate. Something is wrong.

Hi, the probe does go to the corner just slightly above the heat bed before printing. This step is actually normal. It seems like you are having a problem with the printer not being able to start a print? If that is the case please send a request to our customer support at support@snapmaker.com. Sorry for the inconvinience.

Has anyone figured this out? I want to add autoleveling with simplify 3d but don’t know the command, I was reading over the gcode output for the file from luban and couldn’t figure it out

For what it’s worth, on my S3D-sliced prints, the printer firmware has added it for me… not that I wanted it to.

G1029 Pn where n is the size of the grid (3 is default and this command is not needed to run a 3x3 level but you can do a P5 or P7 for a 5x5 or 7x7 respectively)

G1029 A ; perform the auto level
G1029 S ; saves the mesh
G1029 D0 ; ends the process
M500 ; saves mesh to EEPROM

2 Likes

You need to set the Z offset in-between the G1029A & G1029S.
Otherwise your mesh will be off by the value of what the Z offset should have been.

1 Like

Yep apologies I was going from memory which ain’t what it used to be.