Updated leveling process


Can we get the DIY-Test pillar file as an stl instead of just a gcode? I’d like to swap vats and test the level on my other vat, but I’d rather not swap out the resin that’s in it just to run the leveling test. In order to run that resin correctly, I need to adjust the laser speed, and can’t do that with the gcode file…

I could just remake the pillars myself, but it would be easier on me if you could provide for download.


here are the STL files:




OK My suggestion is to adapt a camera ball and socket clamp, as below.
One half screws to the Z axis arm ( may even be the same thread size)
The other half is screwed into the inside of the build plate ( Id like to take a build plate apart to see if this is easy enough?)

The software would need prior adaption to allow for any height difference, but once done the plate could be lowered down loose, the ball tightened and the build plate is level, Would take all of two minutes…
Ive a similar ball and socket clamp in front of me attached to a tripod scooter, and can def say a decent quality one will be more than strong enough…


cheaper option costing a few £$


That idea was suggested some time ago. The major issue, IMHO, is that the mechanism tends to drift slightly due to the mechanical offset of the locking mechanism. This does not affect a camera alignment much or even visibly, but could cause issues with the tolerances of a laser SLA printer. I believe Wanhao uses a similar setup on the Duplicator 7, but that is a DLP printer and perhaps more forgiving.


#off topic, removed!# :wink:


just got an idea to think about alternative way of leveling, think of those cheap small digital weight scales, for jewelry (or kitchen scale but smaller) - imagine placing 4 of them on each corner - and get the same weight
on all of them in order to level…

  • also thinking about how nintendo wii balance board has 4 corners used to control the games - also it obviously measures balance - and I’ve got one of those that I don’t use… would be an interesting hack to tear it apart, place them in corners under moai build plate and use the bluetooth to show results on your mobile … sounds like a hackaton project

… this scale idea might be worth it… just something to get a real time feedback on how to get most accurate leveling in fastest way possible…


Ordered my kit from Matterhackers on Tuesday Night had everything assembled by last night. This thread has been a huge help. Im on my 7th set of test risers, I think Im where I need to be, Just wanted to make sure before I went ahead with burning through more resin,

A 9.20
B 9.71
C 9.21
D 9.68

A 10.56
B 11.04
c 10.74
D 10.99

A 10.57
B 11.00
C 10.73
D 10.94

A 10.74
B 10.81
C 10.73
D 10.66

A 10.74
B 10.81
c 10.77
d 10.68

A 10.85
B 10.93
c 10.95
D 10.92

A 10.79
B 10.87
C 10.89
D 10.85

Will that .1 difference between A and C make a difference?

Thank You.


from #4 and onward, you would be fine with printing. #7 is very good. no need to get perfect as long as you are in the 10.7 - 11 range


So after you Get the corners at around 10.80
You keep your Z at 1877 ? (Since i have Them at 10.80 at 1877)

Thank you


Yes that is correct. If you are having adhesion problems you can increase that and that will lower the build plate .100mm for each increment.


10.7mm will stick better so if you like to print large, maybe go for that extra 0.1mm


This is what I was looking for. I have been struggling to get a print started and I was pretty sure my calibration was off. The ring test worked great and my four cylinders all printed fine, but they keep coming out at about 10.1, 10.15, 10.2,10.1. I kept tightening the nuts underneath a half turn each time I ran the calibration test. Every time it was just about the same until it wouldn’t print them anymore. I suspected that I needed to adjust the Z reset position, but this has confirmed that suspicion. I am hoping this will help, because I’ve been working on it for days and have not been able to successfully print anything other than four cylinders and the same ring.


the whole idea of leveling for Moai can be narrowed down to this:

There has to be enough pressure from build plate to the vats such that it will turn an 11mm tall cylinder in STL file into 10.7mm in print. The loss of the 0.3mm is the measurement of that plate pressure.


Is there some reason not to just use a bubble level to make sure the vat is level? Is the issue that typical bubble levels aren’t precise enough, or that it’s easier to translate from a test print height at each corner to screw turns at each corner?


it has nothing to do with the absolute levelness to the gravity. it is about applying equal force at a specific amount to the PDMS layers which can be uneven. therefore, the bubble leveler will not be useful and have misled some users in the pasy.


I see. Well, in that case, I would recommend a terminology change. We’re not leveling the vat at all, then; we’re making the vat parallel to the build plate. It’s very confusing to call it leveling. If it were leveling, then we’d have to re-do it when moving the printer, but not when changing vats.

I’d also like to echo the sentiments of others that this should be built into the printer in some form. Even if the mechanics of making the vat parallel to the build plate are infeasible to automate, there should be a way to incorporate pressure sensors to provide realtime feedback about the parallelization effort.


Technically, the term is tramming. But it’s been called leveling ever since the first consumer FDM printers came onto the market, so that’s the common term used by everyone.


Tightening the nut half turn will increase cylinder by about 0.3mm while loosening the nut by half turn adds about 0.3mm.

Could you please fix this? It states that both loosening and tightening the nut add height to the test cylinders when printed.


Use this instead.