Shrinkage/dimensional accuracy between white/biomed amber

Hi all

I’ve been working on a theory of repairing mandibular fractures in cats with 3D printed splints between their teeth. I’m modeling the splint from volume rendering of the CT scans and prototyping with standard white resin first, printed at 50 microns with a form 3B+, and washed and cured as stated. This was all going well and I had tested to my satisfaction and I had tested them on a skull and shown a good fit.

I switched to a prototype with biomed amber and suddenly the splint is very loose; Around 0.2-0.5mm of space around the teeth when the splint is placed over them in Biomed Amber, where the exact same splint was fine in white.

All cured according to specifications, using full rafts, in the same orientation.

The most important bit of my troubleshooting is to know which one is true to the dimensions of my digital model; the white or the biomed amber (or neither)

My best guess at the moment is: I think the CT scan may add a small volume to the teeth when scanning (In humans, I believe this may be why they combine this with laser scanning as well, but I don’t have access). As i’m modelling a splint based on a slightly larger than real life model, my splint comes out with some wiggle room when printing in Biomed Amber, and this is the one that is dimensionally accurate…Whereas white has a bit of inaccuraccy which by chance happened to fit perfectly with the inaccuracies from the CT?

This might help to tune the accuracy of different resins: fit-tuning-test