Lost a chunk of aluminum because I got impatient mid-cycle

Nov 10, 2026 Last reply: 1 day ago 4 Replies
Jump to replies (4)

So there I was, running a macro on the mill to face off a stack of 6061 blanks, and I decided to use the cycle time to punch in the next job's edits. Big mistake. Somewhere in the middle of me poking around the editor, the loop counter in the running macro lost the plot and threw in an extra pass. The tool walked right past my clearance point and took a bite out of the fixture plate. Luckily it was aluminum and not something expensive, so it cost me a 20 minute setup and a bad word or two. But the weird part was the repeat. Ran the exact same program untouched the next morning and it was perfect, every pass where it should be. Only difference between the two runs was me sitting at the keyboard while it was cutting. Did it three more times over the week to prove it to myself, editing during the cycle almost every time, and sure enough the pass count came up wrong twice. Hands off until the program ends and it behaves every time. Now I just load the next job after the cycle finishes and go get coffee while it runs.


Same thing bit me on my ST-20 a couple years back. I was running a batch of drill cycles on some aluminum brackets and got bored waiting, so I started typing G-code tweaks for the evening job right there in the editor while the spindle was turning. Partway through the batch the counter skipped ahead and it skipped two holes. No crash, just two missing holes and a confused me holding a caliper. I did basially what you did, ran it back with edits and without. Hands off, perfect every time. Editing mid-cycle, it would glitch maybe one run in three. Never the same spot twice either, wich made me think it was something the control was doing on its own while I typed.

My theory was that the editor redraw and whatever the background was doing to the macro variables just stepped on each other, but I never dug deeper than that. I asked our Haas guy about it and he just laughed and said don't do that, treat a running program like it's radioactive.

So now I keep a printed copy of the next job's changes on a clipboard by the machine and mark it up with a pencil while it runs, then type everything in once the cycle ends. Feels old fashioned but it costs me nothing and the control stays happy. That machine has run hundreds of macros since without a single repeat of the problem.

Funny, I edit during cycles on my VF-2 all the time and never had a loop counter misbehave. What did bite me once was saving the file itself while the program was running. I had a facing macro with the counter stored in a variable, wrote my changes, hit save, and the control aparently choked on reading and writing the same file at once. Next pass the counter reset to somthing dumb and the tool dove in like the stock was still there. Scrapped a vise jaw that day, which hurt more than your fixture plate did. After that I got in the habit of saving edits under a new program number instead of over the one that's running. Haven't had a repeat since, and I still sit there typing away while chips fly. Might be worth trying that next time you want to punch in the next job mid-cycle. If the weirdness follows the saving and not the typing, at least you know what to blame. Could also just be your control is an older software rev than mine and it handles the editor differently, no way to know from here.

I wonder if it depends on which file you're editing. On my VM-2 I got in the habit of opening a copy of the next job under a diffrent program number and typing into that while the macro runs, and in three years of doing it I've never had a counter skip. The one time I got lazy and edited the running file itself, same thing happened, extra pass and a gouged vise jaw. Could be coincidence but I stopped testing that theory after the vise. Try the copy trick and see if it still misbehaves.

Count me in the same club. My fix was even lazier than a clipboard, I just keep the next program on a USB stick in my pocket and transfer it over after the cycle ends. Used to stand there editing on the control during long facing runs on my old knee mill and yeah, weird stuff would happen. Since the USB habit started, not one repeat in over a year.

The thing I noticed too was that it wasn't just editing. Even flipping through pages on the screen, like looking at offsets or positions, seemed to raise the odds of a glitch. Hands in my pockets, machine runs like a rock.

Join the Discussion

Have something to add? Share your thoughts — no account required.

Didn't find your answer?

Ask the community — no account required