Hi all, we recently got our system set up and completed calibration (actually 2 now). I cut 2 - 30" diameter circles and both are a little bit oval, its almost not noticable unless you really look for it. I have the machine set up in my sign shop so I can conviently cut exact shapes out of vinyl with the plotters. When the guys when to mount the vinyl to circle I cut with the maslow the border was noticably uneven. So I cut 2 more 10" circles and cut a 10" circle with the vinyl plotter as a test. It’s actually larger than 10" top and bottom and thinner in the sides, its also not a smooth cut its kindy of jagged - not a great pic but you can see the black circle sticking out the edges
Another issue i had - which i may have solved with feed rate is uneven depth cutting, it is almost like as the sled moved up and around it was tilting backwards or lifting up. You can see the router path and the part that doesn’t get cut away gets thicker at the top
I thought I saw somewhere in the forums or build guide the optimal degrees to place your frame when in the vertical format and now I can’t seem to find it anywhere. Ours is at 66 degrees, I think it should be plenty rigid, its full 2x4 frame and fully sheeted with 1/2" plywood. We 3D printed the anchor points and push 3/8" bolts through them to hold the belts.
These test cuts were made over 2 different sets of firmware and 2 different calibrations with the same results. So there has to be something I’m missing. Any help would be much appreciated.
Hi all, we recently got our system set up and completed calibration (actually 2 now). I cut 2 - 30" diameter circles and both are a little bit oval, its almost not noticable unless you really look for it. I have the machine set up in my sign shop so I can conviently cut exact shapes out of vinyl with the plotters. When the guys when to mount the vinyl to circle I cut with the maslow the border was noticably uneven. So I cut 2 more 10" circles and cut a 10" circle with the vinyl plotter as a test. It’s actually larger than 10" top and bottom and thinner in the sides, its also not a smooth cut its kindy of jagged - not a great pic but you can see the black circle sticking out the edges
what was your calibration fitness score? this sort of thing improves with better
fitness. Accuracy is something that Bar is going to get back to working on as
people start getting through calibration successfully.
Another issue i had - which i may have solved with feed rate is uneven depth cutting, it is almost like as the sled moved up and around it was tilting backwards or lifting up. You can see the router path and the part that doesn’t get cut away gets thicker at the top
making the sled slicker so that it slides more evenly will probably help this.
I thought I saw somewhere in the forums or build guide the optimal degrees to place your frame when in the vertical format and now I can’t seem to find it anywhere. Ours is at 66 degrees, I think it should be plenty rigid, its full 2x4 frame and fully sheeted with 1/2" plywood. We 3D printed the anchor points and push 3/8" bolts through them to hold the belts.
it’s not at all clear that there is an ideal angle.
These test cuts were made over 2 different sets of firmware and 2 different calibrations with the same results. So there has to be something I’m missing. Any help would be much appreciated.
I don’t think you are missing anything, just hitting the current limits.
Thanks @dlang , I can’t remember and don’t know how to check it now. It occured to me that I used feed rates and spindle speeds I found online for the materials I was trying to cut and they are just potentially way too fast for this machine. I’m going to slow it way down try another test cut or 2
Thanks @dlang , I can’t remember and don’t know how to check it now. It
occured to me that I used feed rates and spindle speeds I found online for the
materials I was trying to cut and they are just potentially way too fast for
this machine. I’m going to slow it way down try another test cut or 2
Yes, the maslow is not a very rigid machine, so we have to go a lot slower and
be a lot less aggressive.
with the earlier maslow, I had really good results going with single-flute
cutters.
Looks like that to me too.
When I measure there is a difference between x and y accuracy but not as much of a difference as you are showing.
My set up is horizontal, maybe that makes for less difference between x and y which in theory at least would be logical.
The sled lifting due too friction also can alter the shape of the cut in an uneven way (directionally).
This is top of my todo list. I’ve gotten pretty bogged down in bug fixing, but I’m really hoping to get back to accuracy testing soon. We pretty much re-wrote the entire calibration process since the last time I had a chance to dig into any real accuracy testing. Once we’ve got everyone up and cutting, that’s going to be my next main focus.
Thanks everyone, so I waxed the bottom of the sled and re- ran a calibration and ended up at 0.74 is that good? I’m going to do some test cuts tomorrow at really slow speeds to see how things line up.
Update, with the newest calibration, polishing the bottom of the sled and slowing the machine way down I have a totally acceptable result. It ended up 1/16" too tall and 1/16" too skinny on a 21" x 23" cut, but for our purposes that is acceptable.
However after this I attempted to cut 2 - 24" diameter circles, but they again were not quite right.
Look in the settings for fractal mode and turn it off.
Sorry, I couldn’t resist. I’m assuming those extra arcs were not what you intended. My guess is that is something 1) in the gcode, or 2) how the Maslow interprets the gcode. There was a discussion recently about some unwanted behavior when using a certain arc command. I bet you found that. M4 crashes whilst trying to cut gcode
Sorry I just re-read my post and it’s a bit confusing, that job actually did run correctly with good accruacy, the face was digitally printed on vinyl and mounted to the PVC blank I cut with the maslow.
