I'm getting to the end of a full teardown and rebuild of a 2-year old v1.2 B9, hitting several maintenance items and alignment checks along the way.
One of the big things I'm trying to correct in this rebuild is the machine's geometry. I haven't studied the flatness of each panel before, and was surprised to find out just how bowed each panel and major component of the machine was! In fact, the only things that weren't significantly out of flatness, were the linear motion guide rail (the back plate most DEFINITELY was out... I had to shim by over 0.018"... the THK specs call for a surface with less than 50um deviation, lol... soft aluminum or not, it was pulling the rail out of shape). Most components were checked with a known good granite surface plate, and several ground machinist's straight edges and squares. I'm amazed how well the printer was doing previously, even with these loose tolerances. Solid AK-like design philosophy! Put the precision where it counts.
Long story short, I've found that the polymer vats themselves are still rocking from front to back (a rotation about the x-axis) on the red x-slide plate. This happens regardless of thumbscrew tension, or with the thumbscrews removed or installed. The plate itself is very flat, and isn't rocking in the rails (rails have been sufficiently, but not perfectly shimmed). Also, it's pretty apparent that everything but the thumbscrew ears are raised up by a few thou' above the level of the plate. So, the vat is effectively floating, only held down by the two ears.
I did a quick search and couldn't really find much info relating back to the poly vats, and how they should be behaving.
So, I checked various vats for flatness. The bottom surfaces were indeed extremely flat... no noticeable rocking on the granite surface plate. Also, I was measuring from both the frame's ridges and the raised locating features on the corners of the vats WITHOUT the ears on the plate. Both were very flat! However, as soon as I found a way to measure the ears and the rest of the plate (without the raised locating features being measured), boom, rocking, just as when mounted on the red sliding plate.
Measuring across two vats (I have more that I can check as well), I found that one was raised a lot higher than the other, but both had the same rocking behavior. So, I'm assuming it's a design feature.
Is this a normal thing? I could see the extra clearance and flex acting to protect both the vat and x-axis components from damage due to compression during curing and movement.
Does sanding the vat ears flat to the level of the rest of the vat (or shimming it so it doesn't rock) improve printing performance at all?
I've read through a few older posts on vat position repeatability, and it sounds like there's a definite improvement in print quality if the pdms comes back to approximately the same place for each layer. Is this still the consensus?
Of course, it's not as big of a factor as projector to z-axis alignment changes during printing. But, I'd like to eek out every ounce of performance from our printer if possible.
Thanks in advance,