Hey everybody!
Long post incoming, so TL;DR : What affects print precision/detail level and how to optimize for as close to pixel-perfect as possible?
I've recently gotten an Anycubic Photon S for printing minis and general learning.
Testprinting the resin test (only one part, sliced multiple times with different exposure times since I have the S, 0.02 layer height, from 2s to 8s expsoure, anycubic green) has resulted in underwhelming precision for me. neighboring features merge together heavily, with the least merging at 4s exposure, at merging below 0.3mm at the cost of features < 3px not printing/failing to print over multiple layers. If I want those to show up I have to go up to 8s at which point details merge at distances < 0.7mm. (non-free standing details blend together totally, so rivets and stuff need to be bigger than 1mm to even look like rivets and not just be bumps) small details are also (of course) not at all their width in px anymore but heavily smoothed and about 0.3mm wider in all directions.
I've also tried the same procedure with elegoo black (thinking it might be diffusion and black absorbs light right??) which performs even worse, only starting to show 1px detail at 12s.
I've inspected all the different testprints under a microscope, so It's not just that the material is hiding the detail. I also understand that I'm doing the equivalent of pixel-peeping by now, but it's more about principle and learning at this point^^
now to optimization:
as far as I can tell, the points I can optimize are: the resin, the timings, temperature, the layer height, the FEP thickness, -clarity and tension and leveling of the build platform.
I've read through as many resin suggestions as I could find and Phrozen ABS Like Grey seems to be a very good performer, as well as MiniQ LCD-5400s (post by
Boon Kheng Tang
showing it of a few weeks ago) the latter being much more expensive for me. Apart from that I've been hearing that a lot of people are happy with anycubic green, so I'm not quite sure how much of a factor this is.
my room temp is relatively low, about 18c. which is out of the sweet spot if I'm not mistaken. Is that much of an Issue?
timing test are very slow on the S right now, since we can't use the resin test that tests multiple timings at once yet.
What would need to be done to make that compatible with the photon S?
I tried a few prints with a layer height of 0.01mm, but got the same blended results. My thinking was that with reducing the layer curing time, I could theoretically reduce the amount of resin that cures through scattered light, but that seems to be the wrong Idea of how the process works. Do you guys generally get more detail with more cubic voxels? (eg 0.05mm layer height)
I've replaced the FEP that was shipped with the photon since it looked relatively scratched and cloudy to me already. Since I had no reference to what an actually bad film looks like, I replaced a perfectly good one and nothing changed... Does anybody have experience with thinner films? Maybe they diffuse the light less between LCD and resin? (also is that even an Issue?). The film is tensioned to 350khz.
build platform is as level as I can get it. also that should be a non-issue after the base sticks, right?
Sorry for this extremely long post. I'm hoping that you can help me understand how the process works in as much detail as possible, hopefully to write a comprehensive optimization and recommendation guide and get the best possible prints for detail out of the anycubic. The possibility is there, especially with antialiasing, for reealy nice prints with this machine, I just can't get it there without help...
翻譯年糕
Ray Baumanns
2019-05-08 07:53:04
Brad Senn
2019-05-08 11:13:12
Chris Graf
2019-05-09 22:03:53
