While working on a multipass workpiece the 3/8” Compression bit somehow slipped in the collet (new Amana bit and new collet). Not sure how that happened. No real damage to my work piece as I stopped it as soon as I heard it “shuddering”. I was going to check bit, re-tighten and Z-0 and resume, but when reloading the code, I noticed that the toolpath on the controller seemed offset. Fortunately, I did not just start anew after new Z-0 as the CNC lost my original XY-0 (or it’s Home). Not sure if the “shuddering” made the machine think it moved when it didn’t or what. Thoughts?
Re-setting XY-0 precisely on that front left corner was nearly impossible to get within the tolerance I needed to not mess up. Fortunalety, I had a clean wire hole already bored in the design. I set as best I could on the corner, then I jogged over to the known hole center (912x880mm) and tweaked the XY until the bit turned cleanly in the hole. Seeing I was still off by 0.5mm and 1.5mm, I backed that off of my XY-0 position and reset XY-0. Upon resuming, it was nearly perfect, perhaps still off by 0.1-0.2mm.
Oh, by the way, I also turned controller off and back on again to completely Re-Home the machine before resetting XY-0.
Is there a better way to do this? Ie, tell it a known point and let it calc true XY-0?
Shuttering usually means the machine has skipped steps, and since it is a stepper based machine with no encoders you can never really get it back perfect even if you used machine home.
Thanks. Yes, I think it skipped steps, Next time I do a critical piece I will:
record the machine XY prior to setting XY - to try to get back close after a reset. I’ve seen some data stating that Laguna Swift re-homing “can” be within 0.01.
drill a small reference hole in my jig at (-20,0) after setting XY-0 so I can re-reference in the event something similar happens.
For what it’s worth, I thought the little display on the new controller was “silly”, but that is actually what alerted me to something being off as it was offset from the workpiece on the screen.
On the bit slipping, it was a new coated Amana industrial bit and a brand new collet. Perhaps the coating affected the hold?
.01” repeatability on homing is going to be pretty hard with this machine hardware/
To get the absolute most repeatable machine homing, I would also say this is super important to do when the machine skips steps as there is only one non-contact limit switch on one side of the gantry. This means the gantry can get crooked and make your parts unsquare.
Turn the machine off when the gantry is close to home. Manually grab the gantry and pull it onto those hard stops. while still holding turn the machine on and wait for the steppers to engage then home like normal.
Good to know the tip on manually pulling to hard stops. Thanks
It was the 2nd 8-9mm pass with 3/8” EM on this piece (36mm Baltic Birch). It flawlessly cut previous profiles and grooves, including a 31mm deep pocket (total depth after 4 passes). Previous run had just finished bumping down to the 31mm (+1/-0) spec of the pocket by increasing depth 0.1mm each pass ultil pocket was in spec across the entire length. This pocket toolpath performed perfectly. I then moved on to finish piece with final profile cuts, same settings and per pass depth. That is when I suddenly saw a ~4-6 mm depth change that stopped it in it’s track. Actually glad it did before bit continued downward!
Adding reference holes is a great technique. On machines like the Tormach that have probes to re-find them, it’s my favorite re-locating technique, so I move my bores up so that they happen early.
I’ve also used a trick for when I know I’m going to need to work in multiple sessions: cut a hole pattern in the stock for the workpiece. Then the next time I need to work on that workpiece, recut the same hole pattern in a sacrificial fixture and use pins (tool shafts work) to locate the work on fixture. No need to precisely zero coordinates before cutting the fixture, a zero in any convenient area will establish the location of the part.