You can get better prints by knowing about the errors in your Form1+. You can even print this at the same time as other models so you don’t have to do a special run. And if you post your results, we can discover what is typical is too.
And from the various measurements I was able to discover the following errors:
X-scale = 0.50% too big
Y-scale = 0.10% too big
X-offset = 1.35mm too far toward hinge
Y-offset = 0.85mm too far toward front
Z-loss = 0.643mm
Platform height = Full adhesion, do not adjust.
Z-scale = (can only be properly tested with a tall model and supports under it, not included here)
Z-loss is the most interesting to me, it is the extra thickness I must put under any model I print without supports to maintain dimension. Do NOT adjust your platform height based on this calculation. I know my platform is as high as I can safely go because a little higher and things don’t adhere good.
I also learned I need to reduce my X-scale by 0.50% and my Y-scale by 0.10%. Also good to know!
As a note, z-loss can be minimized by raising the build platform until one of the blocks does not adhere when you print. If your build platform and PDMS in your tank are parallel with micron accuracy, z-loss can be reduced to zero. I don’t believe any printer is that perfect though.
I just got my replacement printer back, and printed a 2cm test cube @25um in grey resin in the right front corner of the build platform. It was about 125 - 200um undersized depending on axis (which I forgot to note before I popped it off the platform…). I did it on default supports @ default orientation. One thing I noted about the surfaces was the 3 faces facing up (support-side, while printing) were rough, and the 3 faces facing down were smooth. Do you see this kind of surface characteristic as well in your prints? The roughness was not caused by the supports themselves.
I downloaded your .form and will use this to calibrate. Couple of questions:
Q: Did you print these little blocks with no supports, then measure their postions afterward while still adhered to the build platform?
Q: How accurate were the blocks themselves? Were there differences between blocks?
Yep, no supports and measure while on platform. I did measure my blocks and they had a little variance, but I chose not to make any judgements on them because the blocks are so small. The large measurements are the same thing with less percentage of error. Error being shrinkage as the resin layer solidifies and layers not stacking perfect enough to escape the precision of a tool like a digital caliper. I haven’t noticed any surface differences, but maybe with an old tank the peels would be harder and damage your surface?
Good questions.
Cool - thanks for getting back to me. I’ll print your file and post the results - maybe you can suggest options on how I should calibrate after the results are in. Thanks again for all your posts - they are all genuinely helpful.
Thanks, It’s good to hear my posts are helpful.
From your X-scale measurement:
10.0203 - 10.0000 = 0.0203mm difference
0.00203 / 10.0000 = 0.00203
0.00203 * 100 = 0.203% error (positive number mean too big)
So your X-scale is 0.20% too big
and your Y-scale is 0.58% too big
Using the Preform fine-tuning tools you can subtract the extra percentage to get back to 10cm on your next print.
It also appears the print is a bit twisted on the platform, so each time you lock the platform on you can tweak it by very lightly rotating it in the direction that compensates as you flip the lever down. This might save your large prints from hanging off the edge some day. Mine was like that too and it’s no problem to compensate.
I ran your calibration test last night and was presently surprised how accurate the printer was -
In the Y-Axis I measured 100.10mm (0.10% out) and in the X-Axis measured 100.25mm (0.25% out).
Following your lead, I’ve created an Excel calculator to calculate the required % increase or decrease from the calibration print measurements. It can be downloaded here – Form1+ Calibration Calculator
I also uploaded an STL version of your test to Sketchfab…. more of an excuse to try Sketchfab, then anything else.
Thanks. It seems the default calibrations are pretty good for everyone. I think the left-to-right direction is the X axis though. And both textboxes in the photo say “Decrease Y-Axis by:” There is an edit button on each post, you can use it to change the picture or fix links. Thanks for posting your results!
On the Formlabs website it gives the axis directions as Y = Hinge to peel & X = front to back - pic.
There is some logic in the Excel sheet that automatically tells you weather you need to increase or decrease the scale based on the calculations. e.g. if i entered a value of 99.25mm for Y, the calculator would tell you to increase Y-Axis by 0.75%.
I’m having problems reprinting the test as some of the cubes aren’t sticking to the build plate. I think I will add rafts to the cube bases and try that.
Really? That’s wack. I guess I should have run the scale test twice.
If you had a cube or two not adhere that probably means those corners of the platform are too far from the tank. Try lowering your platform height 0.1 and try again.
This is a great print to calibrate. I use it all the time. I have noticed that the calibration can change (slightly) if the temperature changes dramatically in the room. Probably not a major issue for most but I do a lot of mechanical parts
