Resin Calibration

[czyzczyz]

I'd be happy to do what I can to try and make some files that will work on the Elegoo Mars Pro. 

I looked at the .stl file, "UvTest2.stl" that was used as an initial source for the photon resin test in meshmixer. 

• It seems to be an object that's 12mm tall. 
  
• The first 1mm is a plate or bed
  
• the 2nd mm is blank
  
• and then the rest is a 10mm-tall stairstep arrangement of test strips (i.e. the first strip is 1mm tall, the last is 10mm). 
  

If I'm understanding the test correctly, all the test strips need to be exposed on the same layer -- so the plate is kept from moving upward between layers somehow. The test-mode.gcode file supplied in the github only cranks down the acceleration and gets rid of the peeling step, so maybe the files have been hacked in some other way to get rid of z increments?

Somehow this object must be sliced in such a way that the first layer is exposed for a long time to make the bed, then the z-axis moves up 1 layer, and then the remainining layers are each exposed in sequence to the same layer without any further movement of the plate. 

My guess would be that to do this right one would need to:

1. scale the z-axis of the object such that the 12mm in height drops to 12*layer_height. So for .05mm that'd be 5% scale.
   
2. slice the object into a .ctb file, with the first layer set to expose for (some long amount of time)
   
3. hack the .ctb file to remove any further z-axis advance between layers 2-12.
   

I could be wrong about step #3, but I think the whole point is for all the exposure to happen on the same layer.

And then before printing that ctb file one would need to run the gcode to set the printer to not peel between layers - but do we know if that gcode works on the Mars Pro and more importantly if the test-mode.gcode works, does the print-mode.gcode file also work and are those the correct settings for the Mars Pro? I'd hate to successfully run the test and then leave the printer set to a non-standard peel or acceleration setting. 'Course if it doesn't save the settings to eeprom then a reboot would fix it.

---

And of course while searching to see if anyone had reverse-engineered the .ctb file format I came across this description of the format as a template for a hex editor. And stepping back a couple of directories from there, it appears that this file is part of a branch of the resin test created specifically for the Elegoo Mars Pro. (and the repository was created 5 months ago -- was the Elegoo Mars Pro in circulation back then?)

The readme simply states "Based on https://github.com/altLab/photon-resin-calibration but adopted for ELEGOO Mars Pro 3d printers."

It appears this might be the answer. Run the test-mode.gcode, print the .ctb, run the print-mode.gcode - and there's the test. Yay!

There's only one .ctb file in the repository, "resin-test-50u.B100.2-20.ctb", which would be the 50-micron 100-second, 2-20 exposure-range test file. I'd love to generate a 25u file as well but don't know how. I wonder if this tool is at all involved in creating the test file -- the same user who made that Elegoo Mars Pro version of the .ctb file also contributed a patch to that project. It looks as one of the functions of that tool is to convert between various sliced file formats for SLA printers, and it recently added support for the .ctb format. Maybe that tool can be used to convert the Elegoo Mars or Photon versions of the test files into ones for the Elegoo Mars Pro?

---

I'm going to give it a try. Because I don't know if that print-mode.gcode sets the printer's parameters exactly back to normal, I'm going to use this test mode gcode file from a different repository which doesn't save settings to eeprom. That way the printer should go back to normal after power-cycling. 

For what it's worth, it does appear that there's a file on the supplied USB drive from Elegoo named "Machine parameters.gcode", so assuming that's all well and correct I should be able to get back to the standard parameters regardless what I do with these setup gcode files. Good to know.

[czyzczyz]

Success!

[czyzczyz]

Success!

---

   

The gcode that didn't save settings to eeprom, followed by the .ctb test file produced this two layer print out of the stl. i had to clean it using paper towels soaked in IPA directly on the print plate, as my IPA-filled pickle jar isn't large enough to fit the plate itself, but that seemed to work fine. I then cured the print for 15 minutes in front of a UV light and tried to remove it from the plate. It was well-adhered, fragile, and brittle, and broke into pieces. So I printed another test, cleaned it, and removed it from the plate before curing, which worked fine. 

Now I just need to figure out how to read the results, to my eye #6-#10 look very similar, with maybe #10 looking the best. This makes me think I'll need to try another test with longer exposures.

Here's a series of images of the test print –these results are for Anycubic "Skin" resin, printed on the Elegoo Mars Pro. Scanning the test print produced a useless peach-colored mostly-solid rounded-rectangle image. Photographing it is a little difficult, so I tried photos aiming to look at the result using transmissive, diffuse, and reflected light.


   
   
   
   

---

Following the directions for the card gives me an answer:

QUICK READ OF A CARD

• Start from top and locate first well formed circle column
  
  
• Move to bottom last visible Circle
  
  
• Move right untill the circle starts to clog
  
  
• Backtrack one colum
  
  
• That's your sweetspot
  

The first well-formed circle column is #6, the bottom-most circle on that column that shows any negative space clogs up in column #7, so #6 is the winner.

The test file I used was "resin-test-50u.B100.2-20.ctb", which means it was a 50 micron test, base layer exposure 100 seconds, and the exposure of the 10 columns ran from 2-20 seconds from left to right. so column #6 is a 12-second exposure.

So I'll be adjusting my next prints using this resin at that z-layer height to do 12-second exposures per layer. When I last printed the Amerilabs Town calibration object, I'd used a layer height of 0.04mm and an exposure time of 7s. So it will be interesting to compare a print with the new settings to the old print.

[SirMoo]

Awesome work there.
Did you get any further with this ?

[rAi]

I could not get any of the previous links to work on a relatively new Mars. Starting around eary 2020 they shipped with Firmware v4.3.x and that version does not run the .photon files. 

Therefore, I created a new set of test prints that span exposures from 1 to 20 s in one shot. 

You can find how to use them and extend them at https://github.com/r-a-i/mars-resin-xp-finder

[hardcak]

I could not get any of the previous links to work on a relatively new Mars. Starting around eary 2020 they shipped with Firmware v4.3.x and that version does not run the .photon files. 

Therefore, I created a new set of test prints that span exposures from 1 to 20 s in one shot. 

You can find how to use them and extend them at https://github.com/r-a-i/mars-resin-xp-finder

I have purchased the Elegoo Mars 2P this month- great machine! running V4.3.9... Im interested in calibrating various resins prior to printing, so need a good workflow.

The gcode files to put into print mode seem to work ok, but the .cbt file are not valid on this firmware. I have managed to re slice the original file in CHITUBOX following the settings in the readme (attached). I dont know how to create the multiple files using 010 editor - but Im not that bothered as I only want a couple for 50 micron no AI, so I created a single 50 micron file. It only prints the rectangular base without any features. I set the layer height 0.05mm - should this be 1mm as per below.? Im not sure the editor changes this.. 

other than this test it prints fine using the ABS like Elegoo grey Im using. My exposure is ok, and repeatable prints. It mainly for other resins Im about to try..

[RedDevilLax31]

I don't have it yet but planning on getting a Mars 2 Pro around Christmas time so I have been lurking and planning around it a little. I wanted to be sure to have an exposure test so I went through and edited the model, then processed the ctb file with 010 editor to adjust all needed settings and remove the lifting for the layers after the base layer and should have a 50 micron file for AA 4. Not 100% sure if this works or not but thought I would share it in case its needed. its also set up for the Mars 2 Pro to have .25 second intervals instead of the 1 second in the standard file. I assumed with a exposure time of 2-3 seconds the .25 multiples would be much better than 1 second multiples. If someone wants to test the file feel free and let me know how it works.

---

sorry wouldn't let me attach the .ctb file, ill port somewhere and provide a link

---

Forked the GitHub to upload the file there, can be found at https://github.com/RedDevilLax31/resin_x...r_mars_pro

---

I could not get any of the previous links to work on a relatively new Mars. Starting around eary 2020 they shipped with Firmware v4.3.x and that version does not run the .photon files. 

Therefore, I created a new set of test prints that span exposures from 1 to 20 s in one shot. 

You can find how to use them and extend them at https://github.com/r-a-i/mars-resin-xp-finder

I have purchased the Elegoo Mars 2P this month- great machine! running V4.3.9... Im interested in calibrating various resins prior to printing, so need a good workflow.

The gcode files to put into print mode seem to work ok, but the .cbt file are not valid on this firmware. I have managed to re slice the original file in CHITUBOX following the settings in the readme (attached). I dont know how to create the multiple files using 010 editor - but Im not that bothered as I only want a couple for 50 micron no AI, so I created a single 50 micron file. It only prints the rectangular base without any features. I set the layer height 0.05mm - should this be 1mm as per below.? Im not sure the editor changes this.. 

other than this test it prints fine using the ABS like Elegoo grey Im using. My exposure is ok, and repeatable prints. It mainly for other resins Im about to try..

hardcak, from what I found out you need to go into the file in the 010 editor and run the template, then go in manually and update the settings in each layer created, https://youtu.be/s_NIeiNoKi0?t=204 was a link in one of the github pages for it that sort of shows the process that needs to be used but is a little confusing too because some of the settings in 010 editor are not in the same spots as the video

[hardcak]

thats good.. Ive downloaded the files and will have a look when I get a chance and feedback anything useful. If you are buying one I would recommend. I cant really fault for the price. Im after dimensional accuracy - so between this and something like a formlabs I dont know how they compare, so the next challenge is to optimise for this..!

[Sepitus]

Hey,
I'm new to resin printer and wondered if this can be used for the Elegoo Saturn as well?

[demon_lee]

I don't have it yet but planning on getting a Mars 2 Pro around Christmas time so I have been lurking and planning around it a little. I wanted to be sure to have an exposure test so I went through and edited the model, then processed the ctb file with 010 editor to adjust all needed settings and remove the lifting for the layers after the base layer and should have a 50 micron file for AA 4. Not 100% sure if this works or not but thought I would share it in case its needed. its also set up for the Mars 2 Pro to have .25 second intervals instead of the 1 second in the standard file. I assumed with a exposure time of 2-3 seconds the .25 multiples would be much better than 1 second multiples. If someone wants to test the file feel free and let me know how it works.

---

sorry wouldn't let me attach the .ctb file, ill port somewhere and provide a link

---

Forked the GitHub to upload the file there, can be found at https://github.com/RedDevilLax31/resin_x...r_mars_pro

Hi!

What is the exposure time of the first model and what is the last in test?
0.25 sec - 0.5 - 0.75 - 1.0 - etc.?

and 50um_a4_UvTest3V5.ctb is wrong. lifting of the table between layers is about 1mm, not 50 microns.