Oh dang thats a cool setup.
I was going to say you DO have a cnc, but that looks a lot funner to be honest.
Pretty much a time saver anyhow
Oh dang thats a cool setup.
I was going to say you DO have a cnc, but that looks a lot funner to be honest.
Pretty much a time saver anyhow
Spotting will tell you where the high places are. If theyāre really high, it doesnāt matter how you take them down, as long as you can do it in an incremental and controlled fashion. Iād first trying using a dremel tool with a rubber-bonded abrasive. Itās likely to be both fast and controllable enough. Itās way less work than trying to jig up for a belt grinder.
Great feedback. I think Iāll try but with something like a silicone baking mat in between (complete silicone one, not one of those with fiberglass in them). Those are typically ok with much higher temperatures and the direct contact with the wood (and thus heating of the wood) would be less of a problem.
This is of course assuming that the tolerances are less than those are not worse than from the carriage But if the wooden bed underneath is flat I think the heated bed and print plate together should be stiff enough for small variances in the silicone layer. And based on the marketing material Iāve been finding the thermal properties (expanding/shrinking) should be ok too.
So something like this: https://www.amazon.com/Silicone-Nonstick-Resistent-Countertop-Protector/dp/B078Y1H4YQ
edit: Or in stead of silicone, something like this: High-Temp Felt Plumbers Pad (https://www.amazon.com/Inch-High-Felt-Plumber-Store/dp/B01NAJZT8Z/), although I think the silicone would be more convenient and āflatterā
Iām not too worried about that in my personal case as my printer is located in the basement which has rather stable environmental conditions throughout the year and the mechanical ventilation also manages the humidity in the air (up to some point).
I was chatting with devs on the marlin discord today to understand more about the reference implementation of ABL - they seemed to think some level of fade is essential. So I started looking at why SM2 fade is 0 and ended here.
so question - is snapmaker implementing some level of fade or applying it over the whole model height
(they, sigh, also asked me why i wasnāt using UBL as, and i quote āthatās so superior its laughable we still keep ABL in the code, we only do it for old machines with little memoryā)
I think fade is the best thing to use as well however snapmaker has bastardized leveling by co-opting for integrated z home offset, which is supposed to be done using M206.
In order to use fade the home offset MUST be moved into M206 otherwise it will result in bizarre behavior.
By default there is no fade, the bed mesh is carried through the entire model. It would take manual work to split the values into the correct locations before fading could work.
I recently ran tests using UBL and BIL on my Ender 5 Plus, and to my surprise, UBL was terrible and unusable. When I switched to BIL, the leveling was spot on. UBL starts out with a course bed level, and then narrows in in areas where it detects greater differences, which is how it misses some areas. Both levels were based on a 5x5 grid, and I ran at least 10 UBL tests prior to switching to BIL. If an 11x11 grid is used UBL may perform better, but I donāt trust the results after seeing it work.