After using @JoshK’s calculator and 4-cube form file, and from reading various ideas on the topic here, plus noodling a bit myself, I created a new excel calculator to insure dimensional accuracy in all three axes, as well as a step-gauge and test cube.
Dialing in your printer is really important if you’re prototyping injection molded parts that need to snap together, so that’s why I built this. I could get X & Y pretty good with Josh’s calculator, but my Z dimension was way off, and was causing problems.
Give it a whirl and let me know how it goes for you. After calibrating my printer with this, the 2cm cube model is literally within 10um in all 3 dimensions.
Calculator: (rename extension .obj to .xlsx. I had to change to upload it….)
Form1_Calibration_Calculator.obj (9.1 KB)
Step Model:
form1-step-gauge-100mm3.form (445.6 KB)
Test Cube:
hollow-2cm-cube.form (77.5 KB)
If you find bugs, or see ways to improve it, please let me know.
-C
7 Likes
This is really cool, Chris. Thanks for sharing.
1 Like
Looks legit. I will set this to run when I leave the office tonight.
If you do tune your printer this way, you have to remember to scale each object individually in Preform to the calculated value every time you use Preform. You’ll probably only forget once…
My settings are now:
Scale:
0.984%
Tuning:
X: 1.01%
Y: 2.05%
I’d be interested in what others’ calculated settings are.
What resolution should this be printed at?
Sorry for delay - I printed @100um. As I’ve mentioned in way earlier posts ( @Formlabs1 ), I cannot get a print to work @ 50 or 25. My symptom is always a part will print beautifully at 100um, but whenever I try at anything higher, the supports break part way up and the print fails. I tried to do the cube at 50um, but I will admit, this is not a new tray, and the white resin I used is pretty old, so that may be adding to the issue.
Here’s an example of my typical result: 100um on the left, 50um on the right.
These supports took forever! Are the points 1.2mm!?
hehe, yeah, they’re huge. I had to do that to make sure the model stayed dimensionally stable. I have to say it’s kinda cool seeing the model printed on your desk (cool desk by the way). How did it print for you? Looks pretty good in the picture.
I know the feeling… Seeing something you designed someplace you didn’t put it is pretty neat.
It printed well, although i did manage to crack one of the sides pulling the supports off. Still haven’t had a chance to measure it though – the calipers has been eluding me 
I guess I must be pretty lame at converting the OBJ to .xlsx Could someone enlighten me. This calibration sounds like EXACTLY what I need. I’m a mechanical design engineer and I have been struggling with this since I got my unit. I’ve used a lot of resin to get the part to meet my Solid Works models.
@Steve_Wicksman Hi, all you need to do is rename the file extension to .xlxs - there’s no conversion happening. Windows, and maybe Mac too (can’t remember, I use Linux) do not show you extensions by default. You’ll need to turn that on somewhere in the explorer’s options. Once you do that, then you will see how to rename just the extension.
@Formlabs If you allowed .xlxs extensions in your upload tool, I would not have had to hack that. In fact, a blacklist rather than a whitelist would really make more sense, possibly coupled with a max size limit.
1 Like
Open File Browser (Drive and Directory listing) and just rename the file by changing the extension from OBJ to XLSX.
1 Like
OK I was able to do it, but I had to rt click on the file link above, “save link as” and then add the extension at that point. It saved just fine then. Not sure why my browser does not show the file with the extension, but maybe there is an “option” that I have to change. I can’t wait to get home to build the part and measure it!! Thanks everyone…
Ahh right. He meant download the file, then change the extension. Changing it via save as works too.
As I said…I went home to print it and got up at 4:30 AM to see how it looked (sign of a true geek). It came out perfect. I printed at 50um. I poped it off the platform easily. Put it in the IPA and went back upstairs to shower for work. I’m thinking in the shower…which way was the part positioned on the platform? I’m thinkin’ I was really still asleep when I took it off. No way to get that information now. So then I’m driving to work and I start thinkin’ again…did Christopher put some kind of detail on the part so that I can tell which way is “X” and which way is “Y”. I did not even think to look. OK so I have to ask…is there something on the part for X and Y? Without that I have a 50-50 chance and then I run the part again to see if I got it right.
I laughed a little to myself… and then froze because I realized I no longer remembered the orientation of my part either 
Then I realized that there was no way the base was symetrical all the way around. If you left the base on, you will notice 2 of the arms have 4 triangles for prying, and 2 arms only have 3.
This difference should be enough to identify which way was which. BUT. It would absolutely be a good idea to add some sort of identifier to the base crossbars or something.
Of course I took the base off when I was asleep… Do we know if the X axis is parallel to the hinge or is it the Y axis?
Preform’s X is opposite to what you’d expect - it is Front-to-Back on the platform.
Yeah - it really should have markings, oops. I use the supports to guide me (the ‘Y’ axis has step supports up the peel side only) but having clear X and Y marks would probably help a lot.
Printing now with madesolid vortex resin. Does the type of resin play into the densities of the calibration artifact?
Yeah, good question - and also resolution as well. I can’t say. I printed the tool in FL white @ 100um. Will FL Black @ 25um be accurate now? No idea, but I’m guessing it’ll be way better than it would be had I not done the calibration at all.
I have some Vorex in the mail to me now - should be here tomorrow. Very interested in seeing how it performs.