New print test
This one is 25 microns Z 50 microns XY Orientation is horizontal
Intially I had issues with it not sticking to the build table, but it turns out I had the attachment layer exposure too high. So build settings for this was 20 seconds attachment layer exposure and 10 seconds regular layer exposure. There were 4 fill layers as attachment layers and then a bunch of support structure. I think the exposure was still too high, the supports were pretty well cured so I think I should have set it lower, like 7 or 6 seconds. Build time was about 1 hour. The quality is excellent, way better than the other professional printers I've used. The X and Y resolution is completely smooth, the Z resolution is only noticeable at the most subtle of angles for a surface. It looks much better than the photo.
1 issue though was the sweeper, since the table barely moves, after a few layers the suction keeps material from getting under the table properly, so it ended up with holes in the layers where material didn't get to, that caused like the tip of the gun not to come out right and caused those damage dots that you can see. Once I noticed I removed the sweeper and it worked just fine. I'll do another test later to confirm that's what caused those issues, but I'm pretty sure. The sweeper isn't strictly necessary so in these cases you need to remove it.
The other side is messed up though, I didn't put enough distance between the model and the attachment layers and so it ruined that side, the exposure time I'm sure also contributed to that.
One thing I learned was that if you are having issues where the layers aren't attaching to the build table then it might be that your exposure is too high, of course it's also possible that your exposure is too low, but this is just to confirm that you can indeed set it too high.
In my next test I'm also going to try at an even higher Z resolution, I'm pretty sure it can do something like 15, which would be fantastic.
I'm also pretty sure the squashing issue I had isn't here anymore, though I can see 1 faint line across it, that might be the same thing as the bumps I saw on the earlier print. I still don't know what that is, but if it's unavoidable I might be able to reduce it by pre-distorting the model so that it cancels it out. It wouldn't be noticeable on surfaces that aren't uniform.