August 18, 2019 at 9:24 am #110040
I noticed these lines during a facing operation (using a 1/2″ straight router bit):
I printed out Ryan’s DW 660 Perpendicularity Tester, and got these results:
With the tester pointed to the front of the MPCNC:
With the tester pointed to the back of the MPCNC:
I tried playing with the tightness of the bolts on the gantry, but they didn’t seem to do anything.
Everything “seems” to be plumb, but I’m probably not measuring correctly. The back supports are 1mm high, but I don’t think that would account for these symptoms as my build is for a 24″x24″ work area.
Any thoughts on how to troubleshoot this? Would lowering the back supports help? (Also, I wonder what caused the bit to leave that one circular shape.)
Attachments:August 18, 2019 at 9:26 am #110045
TimParticipantAugust 18, 2019 at 3:13 pm #110065
I shim with masking tape usually. I like to get thinks real nice and even. I can;t see it but you might want to make sure you mounts are the right distance apart and the top one is facing the right direction, every once and a while someone gets it backwards.August 19, 2019 at 10:00 am #110166
It turns out that the z is off. The reality is a little more pronounced than how it looks in the photo.
I’ve tried fiddling with the tightness of the bolts. I had them very loose. The bearings in most places barely rode on the tubes or didn’t touch them at all. Is that the way it should be? Maybe my prints are off?
In any case, I didn’t see much of a difference with different tightness of the bolts.
Any suggestions on twisting the bottom of the z towards the front of the machine?
Attachments:August 19, 2019 at 10:04 am #110169
Tightening the lower two 2.5″ bolts will push out the bottom. Start very small 1/4 turns.August 25, 2019 at 8:03 am #110723
I think I discovered a bigger problem in that there’s some play between the bearings and the tubes on my parts. Tightening the 2.5″ bolts don’t do much because of this, other than help engage the bearings with the tubes. Thinking this was a setting on my printer, I printed a new XYZ_C_Burly to test with a spare bearings and a piece of conduit. On that one, the bearings just engage the tubes–which is correct, right? I’m not sure what the issue is. I would hope I would have noticed the gaps when I originally put the MPCNC together. The other possibility is cheap PLA which likes to move away from electrical conduit, which seems unlikely.
In any case, where’s the best place to shim? Between the 660_Low_Mount and the Tool_Mount_C?
Thanks.August 25, 2019 at 8:48 am #110726
It is better to completely take it apart and swap some of the parts positions and put it back together, it really does seem to work. Not sure it if loosens things up in a good way or the second time around the assembly just goes smoother but with a bearing not touching it is extremely far off.August 25, 2019 at 2:11 pm #110766
I removed the XYZ_C_Burly off the bottom, since it was the easiest piece to take off. The rollers were loose against the conduit–so much so on one side that I could hear clicks against the bearings when I shook a spare piece of conduit in it. I absentmindedly futzed about with the bolts and bearings while on a long phone call and… Magic? Now the XYZ holds the conduit snugly on both sides (though tighter on one). I have absolutely no idea what happened, but the “take it apart and put it back together” advice definitely worked for this part. I’ll try the others.August 26, 2019 at 9:32 am #110866
I have absolutely no idea what happened, but the “take it apart and put it back together” advice definitely worked for this part. I’ll try the others.
I wish we knew but from day one that tends to be the magic fix.August 26, 2019 at 9:43 am #110869
Putting things back together, a test z tube seems to be perpendicular along the x axis, but not the y. (The bottom of the tube is tilted towards the back of the machine.) Any more magic to try?August 26, 2019 at 6:56 pm #110947
This is a little embarrassing, but I’m fairly sure the “magic” I’m looking for are rails that are exactly parallel to the table.August 27, 2019 at 4:50 am #110983
You must be logged in to reply to this topic.