Leaning Print Results

Hi guys,

I have some trouble with my brandnew Moai 130. In and of itself it works well. But I get skewed/distorted results with an additional dimensional deviation in the x-axis (+1mm).

I haven’t printed much beside the 20x20x20 cubes (printed directly on build plate) but I think it would look similar with the calibration cylinders.

I always tightened the button/grip that fixes the build plate in place on the z-arm.

Here are some pictures/visualisations of the problem:

Additional info:

Room temperature: ~ 24°C (I have not yet put the heater into operation because the installation is not that easy. I don’t want to put it behind the vat, but install it in the upper part.)
Build plate type: Easy to Level
Resins: ‘Peopoly Nex Gray’ and ‘Anycubic Eco UV Resin’
Vat: FEP Vat (sits firmly and is locked in place)
Slicer: Asura v2.2.2. (Windows 10)
Firmware version: ? How can I find out? (I think it’s new, the printer has arrived three weeks ago)
Base: The base of the printer is even and it stands straight.

Looks a bit similar to:

A few suggestions/answers:

Update to Asura 2.2.4 and use the “moai100_calib” calibration setting. Ultimately you will want to create your own calibration, but I haven’t done that yet myself.

The firmware version can be seen on the PRINT / SETUP screen when you first turn on your Moai. The current version is V1.18.

Go ahead and set the heater behind the vat for the time being. You can run the power cord out the door and you will still be able to latch the door closed. 24°C isn’t bad, but with the heater you will be closer to 30°C. Eventually I’ll run the power cord out the back on my Moai, but I’d rather be printing than modifying my printer right now. I would be at 19°C without it.

Definitely warm up the whole system along with the resin. I built a heater with a thermostat into mine. I give it 30-60 minutes to stabilize.
If that doesn’t work you may have a controller board issue. I switched mine out and it immediately fixed the leaning. There seemed to be a phenomenon where the trace of the Y axis shifts over the course of the print.

Thanks a lot for your quick answers. I have tried again, taking all your advice into account. My firmware is up-to-date. I have sliced the cube again with Asura 2.2.4. and also used the “moai100_calib” profile. Furthermore I let the heater warm up the build space for 45 minutes and put the cable through the door. Unfortunately this did not help at all. The cubes looks a little bit more sloped now and the dimension of 21mm in direction of the x-axis remains.

What I also noticed is that all cubes of Anycubic’s Eco Resin have a circular dent on the side facing the Vat. Maybe the result of an air bubble or a drop? But I don’t think it has anything to do with the actual problem.

Maybe I really have the controller board issue? I hope not.

Another idea would be to run the printer with the PDMS Vat and relevel it.

Instead of my last thought of trying the PDMS vat and getting it dirty too, I decided to try again to re-level with the installed FEP vat.

The new print looked exactly the same, with a leaning and the dimensional deviation of 1mm in width.

I honestly have no more ideas. Does anyone have any other suggestions as to what I could do?

Is this measurement is taken before or after curing?
After curing the resin will shrink, about 6.5% depending on the shape of the model

I measured it both before and after, it makes no significant difference. It’s just over a millimetre too much. The other axes are correct in their measurements.

Try reduce PM motor speed + vent holes + oriented on supports
that will give the best result

How far down should I go with PM Motor Speed? I’m already at 8.

You say I should print the calibration cube that is normally printed directly on the printing plate with a support structure? Why should I do that?

I could also try to put just one more ventilation hole in it or just turn it 90 degrees to the side so that both holes are free when printing.

I have reduced the PM motor speed even further as suggested. From the above mentioned 8 to 6, which caused the print to fail. I had created a new print file with Asura in which I added supports to the cube in sloping position. I also added the cylinders to the same print file (Asura screenshot attached). The cube was not printed, only its support structure. Cylinder B also failed. At least the other levelling cylinders were made through the over 3 hours of printing. I think the Z height fits so far, but the cylinders are not really round, really elliptical. Furthermore, they have also become slanted, like all other cubes before. As you can see I also noted this measurements. Also the surface of the cylinders does not look good with such a low PM Motor Speed. Now I now how far I shouldn’t go with this number. :slight_smile:

zyl_1_dw_surface zyl_1_dw_top

Does anyone else have any ideas? I’d really appreciate some real help. The thing is that I bought the printer because I want to get real accurate results, more accurate than with an Anycubic Photon or other budget printers. What I have now is unfortunately not enough for my projects! This is very frustrating.

@DirkW i missed your PM motor speed on your settings, sorry about that, 8 is already really low
orient the cube to be about 30-45 degrees,you don’t need to have the 4 leveling pegs but you need more vent holes on the cube or just make a big vent hole at the bottom

also make sure run the Moai Checkup guide

to ensure key parts are secured. unsecured or misaligned parts could also lead to shift parts

Thanks for trying to help. I have considered everything, all screws of the z-axis have checked the peeling motor and its mounting, there is no friction or jamming. I have used the extended calibration instructions completely and readjusted the mirrors. The pre-adjustment was quite off-center. Then I calibrated to the circles, it fit so far, but the point on the arcs between the axes was always a bit too far out, the movement didn’t seem to be exactly round.

Then I made a print with supports and a print directly on the printing plate. The edge dimensions were now a bit better, but the dimension of the Y-axis (from left to right) was still almost one millimeter too big. The slant was still there, but a bit weaker.

To get the desired dimensions, I calibrated further with the settings and adjusted the x and y size accordingly. Now the dimensions of the cube are exactly 20x20x20 mm. Nevertheless the slant of the cube is still there. Since I have done everything possible and there are no improvements, I assume that this is a hardware problem where something seems to be wrong with the control of the galvos.

It seems to be exactly the same problem as in this thread:

I’m done with it and don’t know what else I could do to fix this problem. I do not believe that it is no longer in my hands to solve this problem. In my opinion, it would now only be a matter of the manufacturer’s warranty.

I know that you have big problems with the coronavirus at the moment, so I certainly cannot expect a timely solution like a return and refund. I wish you all the best for this and hopefully a good end to the current situation, but I hope to hear from you soon.

Here are some pictures of my last print with the best dimensions - unfortunately there is no right angle to see.

cube_after_advancedcalib_01 cube_after_advancedcalib_02 cube_after_advancedcalib_03 cube_after_advancedcalib_04

How is your checkup?
is there any play in the parts?

Okay I try to make it overall more visual.

Checkup & Adv.Calib Ready!

Adv. Calibration Circle

After that I printed this new Cubes

After I corrected x and y size for Cube 2. in the settings I got this, egg shaped calibration circle, but the measurements of the cube were good (apart from the slanting).

I have therefore reset the settings to the default settings of 900 units for x and y size.’

Then I had the idea to create and print a test circle with a diameter of 120mm

In Asura

After Printing

Very detailed information, it was very helpful
I suggest you to go through the Asura custom calibration profile, and set the x and y size back to 900 while you do this
You are having distortion between Moai dimension and Asura’s, this would be much clearer once you printed the Asura cross and take measurements , it might have 1 side out of the build plate in your case

My first print of the circle was also made with the default settings of 900 for the x/y size. The difference with this first test was that I did not use the calibration profile “moai100_calib”.

Okay, so I made both: One version with my settings from the cube and one version with the standard settings of 900 for x/y size and both were sliced with the Asura calibration (moai100_calib) profile.

This is what came out:

What i meant was to use this file : https://drive.google.com/file/d/1oibVc5TPFBH5NJ9tkiiyTT682b6L_TUj/view
slice it in Asura using none profile and from that take your measurement and make your own calibration profile

Yeah, I came across that myself. I did that and it worked. The measurements are now correct and should be sufficient for my purposes. Here’s an overview. What I still do not understand is that my prints are still printed at an angle to the door. Also, during my tests I found that the y-axis is also slightly tilted. Furthermore, the center of the print is also shifted. You can see all this from my pictures. Can you tell me why this happened?

I printed a new 20x20x20 cube with my new Asura calibration profile and it’s also slanted.

The shift is usually due to the tilting mechanism to separate layers
You can try to print without this feature by turning z reset position and z follows to 0
It was made to reduce the peel force on SLA printers but it seems to cause you trouble