Pass through mod idea

Ok, so now that we will all potentially have access to tiled carving (I was beta testing easel tiling). I was thinking, what we want as a mod (community or OEM) is a sheet feeder tied to tiling. So if anyone has used an automated bandsaw to cut down stock, you are familiar with the concept. You load a long bar of stock, enter the interval length you want, a quantity and on command the machine feeds the stock forward by the specified distance (with kerf compensation if it’s good) and cuts pieces off. Now for tiling you’d want to have some feed wheels that push a piece of stock (plywood, or whatever) using some rubber wheels or whatever out the back of the machine by a specified and precise amount (would probably want servos with rotary encoders on the idler wheels). You’d want to make it g-code driven I guess. So at the end of a tile instead of the pause you get now, you get a pause and feed in y-/+ by the tile length (you’d probably want to specify an overlap distance if you wanted one). Like the stepper drivers in the prusa you’d probably want collision detection. The wheels would serve as hold-downs as well as the tiling mechanism. I guess this could be the 4th axis if that makes programming easier. Not sure the force needed to slide a sheet of plywood or god help you thick MDF on the waste board, and if normal steppers can do that (I guess if you did the whole thing via ball-screw instead of wheels you’d get better gearing, but then you would be length limited - with wheels it could keep feeding all day). Just had this idea while watching one of the big industrial CNC routers move it’s own materials on and off. They don’t need to tile as it takes insanely long stock (like an entire RV side panel in one op)

Ideas? Comments? (Haven’t had coffee yet, so it may be silly)

1 Like

Maybe drink some coffee and think about it. :sunglasses: :sunglasses:

I messed with an encoder wheel to measure my material sliding through on tiling jobs. While a fun pandemic lockdown exercise, it was not as accurate as eyeballing a pencil mark or using registration holes. An automatic feeder that was reliable would most likely be more expensive and bigger than your X-Carve.
I still wanna see you do it, though!

Did it fail because the encoder wasn’t high-enough resolution? Or the contact wasn’t reliably in contact?

It was actually pretty accurate, but I never gave it the finish (enclosure, wire management, etc) it would need to make it regularly usable. The biggest problem I could see was because the way I designed it to roll on the surface for measurement, it would measure any dips and raises in the surface. Really accurate on dead flat material. Just easier to make a pencil line if I can’t use index holes. If I did a bunch of tiling, I might revisit it.