I cut a bunch of arches into the spoil board. I’m patching it, but have a few questions:
I’m heading out to get some more Durham’s and a squeeze bottle for more precise application after I mix it.
Those override what is in your file. If properly set the machine will never cut below the maximum depth.
A possible explanation is the bit slipped. Were you using an up-cut end-mill?
Brian,
What in the picture suggests to you that the bit slipped? Personally, I would expect less deep cuts transitioning to deeper cuts if the bit was slipping down in the collet. I’ve personally had similar results to @tomthm in the past and couldn’t give a great reason why either. But, my issue were before all the fixes to the router.
The upcut wants to dive in and lateral forces could keep it there.
Another possibility is missed steps. If the Z axis misses steps while moving up, all future cuts will be deeper.
Cut below the maximum depth.
@tomthm strikes me as honest, thoughtful, and well instructed. He claims to have set the maximum depth. Based on what I believe about him I suspect that he did correctly set the maximum depth.
I believe the machine itself is reasonably well built. If MultiCam claims the machine will never cut below the maximum depth then I suspect the machine will never cut below the maximum depth.
That leaves just one remaining possibility: the end-mill changed positions after the maximum depth was set.
When the spindle shaft, collet, collet nut, and end-mill all expand from warming would you expect the slipping to continue or stop? Also bear in mind the spindle rotates so the nut naturally tightens.
I assume the end-mill slipped early in the job then never again.
@tomthm, did you blow out the collet and nut?
Did it cut into the spoil board the exact height of your material in vcarve?
Thanks for the response Brian,
I’m just quoting to respond to specific statements, not trying to getcha, if you know what I mean.
Not sure this is always the best approach, to just write off the machine as working perfectly and always expect the issue to be user. I agree the machine is seemingly well built. I personally really like it. But, my issues with cutting spoil board when I believe everything was correct make me wonder sometimes, if the machine actually did over feed into the spoil board.
That seems like a possibility, but hard to tell from the picture. my guess from the picture it seems like every cut was even depth, but deeper than intended and as all the depth setting are done with a freshly tightened end mill in the collet, I wouldn’t expect it to slip that far at the start and suddenly stop slipping during the first cut.
I do wonder if there are possibly better collet options if this slipping is as common as it seems.
This is what I wonder.
Thanks for the post Tommy. I am planning to be at DMS this evening for the board meeting. Will come in a little earlier to take a look. Hopefully it can be milled and usage continued. If not, we have some spares on hand. If you are around, we can look at the file. If not, PM the folder & file name to me and I’ll look at it as well. Anyone who has spent much time with CNC router work has run into mysteries. Unfortunately, it’s not usually the machine provided the collet is good and the mill shaft is good. The machines have a maddening capability to do exactly what they are told to do … and not what was intended to be done! Where is AI when you really need it!
Is that what you get when you reverse the order of top and bottom depth measurements ?
If the maximum depth is not set.
The other possibility is an “air cut”.
It depends on the model and where Z = 0 is set.
Is he multicam not closed loop?
No, it is not.
After checking his design and all, would you be able to let us all know what happened if you do figure it out?. I have not cut anything as of yet because I have been nervous about messing anything up. Thanks!
The attached image is when I truly realized the ramifications of missed steps. See how it cut the outline of the sign (1/32" if I remember correctly) then it moved to cut the first letter and scratched the material while doing that? What almost certainly happened was the Z became bound for some reason and didn’t actually retract to the safe height during the G0 moving to the letter. The theory is further strengthened by the fact Z was definitely bound up and could not move up afterwards.
Kinda ironically you can just barely see one of the signs in the background which is a reminder to always set your surface depth. I liked the look of the white with red letters and that one is still around somewhere I think.
Thanks Ken for sharing and having alternate examples. This seems much more consistent with my issues in the past than just the bit slipping. I have had the bit slip though and it was blatantly obvious at the time. The biggest tell tell sign, was the bit being left in the material when the router retracted.
The missed steps is an issue only a very experienced user would notice, probably even with a great example like the one above.
Just to be clear, I have seen what looked like bit slippage too. You can tell the bit is further out from the collet than when you started.
Overall I still think all 3 happen. Not setting top and bottom is obviously the number one reason (not saying this is the case, just obviously the most common). I am not sure if bit slippage or missed steps happens more often.
In my case I shouldn’t have been staring at that scratch like a dog with my head tilted sideways and hit the E-Stop quicker. Will mostly never be that obvious something is going wrong though 
In your file, did you have the Multicam selected or the HAAS setting? Both pieces of equipment use V Carve, but in my lecture class, I distinctly remember @cghaly stating that if the HAAS setting was selected, the spoil board would ALWAYS get damaged.
Actually, if you selected Haas and ran it on the Multicam, assuming it would even parse the G-Code correctly (guessing it would not), it would probably not cut anything and just immediately crash on a rapid.
Haas has a positive Z up in the air while MC has a positive Z into the board.
Thanks to all for your responses. Thank you @bertberaht for coming in and starting the job to resurface the spoil board. If only all mistakes in line were so efficiently remedied. Based on how the bit looked and felt when he removed, Bert seemed to think a slipped bit was the culprit.
Thanks again to all for your comments, and an extra thanks to Bert for his generous and timely rescue.
Ok, there did not appear to be a program problem. Tommy still had the bit in place and the setting as used in the project in question.
Two observations, 1) the existing Z’s were both about an 1/8" too low. So, the machine cut the path it was told to cut which was too deep. 2) The bit was not very far up in the collet. In addition, it had a little scarring on the shaft. Hard to tell if it was there originally or if it slipped. It was a downcut bit so that wasn’t the problem. The cut was three trips through a 3/4" sheet of dense MDF (the dark brown material). It is quite possible that by the time the cut got to the third pass, the bit had worked out of the collet enough to be a problem. As mentioned by others, the slippage could have started and then stopped as the tooling heated up.
Knowing for certain is not likely, but bit slippage as discussed above is my best guess.
The board is being resurfaced and should be ready to go soon.
