Numerically, over a 1/4 of a million does sound a lot, I have to say!
It was for 4 toolpaths totalling 110 minutes - until it got to CM, when it turned into 160 minutes!
I’m not sure CM’s estimation is terribly good. I’m guessing, but it appears to be based on the number of particular instructions and not the nature of them. So 10 instructions that each move 1mm seem to be estimated the same way as 10 instructions that move 400mm each.
1 Like
I’ve had the same issue. Purchased an USB-Isolator and have not had a problem since
2 Likes
Thanks, Jim. that’s good to know!
Welcome to the forum, by the way 
I use cc and cm. CC vastly over estimates cut times by a factor of 2. CM loads gcode file and is usually about half of CC. CM is incorrect usually by under estimating. CM likely calculates cut time correctly but not total time. Seems like rapids and retracts are not counted as well as bit chamges.
2 Likes
I found this to be the other way around!
Either CC under-estimates, but CM seems to take the amount of time it shows - although it does seem to be a bit slow, counting down the minutes. I’ll check, next time!
1 Like
The behaviour of the feed-hold button, a physical button wired back to the feed hold pin on the controller PCB is as you have described. Feed-hold causes spindle stop and Z retract, the second press of the button to re-enable CM on-screen buttons is described as ‘for safety reasons’, ie: you stopped the machine for good reason, you should have to make a conscious step to resume. Maybe so, maybe just that the controller PCB state when ‘self held’ like this isn’t visible to CM, either way it does work.
Pressing Start or Stop in CM then proceeds as you would expect.
On the topic of file/line sizes in the JOB, I have run a 40-ish MB file of (if I recall) something like 500,000 lines, a complex 2.5D finishing carve without issue - both in 4xx and now 5xx versions. I have once seen CM ‘just stop’ and put it down to the static gremlins as no other culprit could be dreamt up.
I recall an article on hacking a .nc file to safely resume following a forced or elected stop on a job, and (I don’t recall who, but think it might have been @Julien) the advice was to note the line number, read back upwards from that line number to a statement that is clearly not a cut, such as a retract and reposition and trim from BEFORE this line otherwise you interfere with cut state and the machine might resume ‘cutting’ before the proper Z positioning. It comes down to how comfortable you are reading, interpreting and then tinkering with the GRBL commands… as with all things, it is described as ‘easy if you know how’
1 Like
Thank you for letting me know! I went from CM3xx to CM5xx at the beginning of this year (had been on CM3 since 2017) so I haven’t yet had to test this out. Before, I remember at least twice the file from fusion was rejected by CM3, I believe it was due to # of lines…
This topic was automatically closed after 30 days. New replies are no longer allowed.