Stopping a Run and then jogging should not break software

If i start a cnc run, then stop it mid run, then use the remote controller to move around the CNC, the state gets logged in jogging and I can no longer play or upload, I have to restart.

Seems like a bug. Not sure where to report these.

I haven’t yet used the 1F/Buildbotics controller, but on the Mach3/Mach4 controller on other CNC machines, once you Stop the program it ends execution and cannot be resumed. In order to pause a program, you hit the Feed Hold button. It can then later be resumed from the paused point. Looking at the Buildbotics documentation, it looks like you accomplish the same thing with the Pause button. Hitting Stop halts execution and exits the program.

The best rule of thumb is to never stop or pause a cutting run unless it’s an emergency.

Oh this is useful; thanks.

Did you stop it or pause it?

That is ideal, but the machine should be able to stop and continue. Don’t you think.
This feels like a software issues.

I tend to agree, for instance - what if a bit breaks, what if your DC craps out, what if you have to leave the machine for any reason, what if your material shifts. There are 1000’s of ā€˜what if’s’. I wish we were able to stop/pause, jog the machine, even re-home z, if necessary. It should be in our control.

Exactly. Sometimes i just want to be in the room when it’s running for safety reasons.

If a bit breaks you can’t stop the 1F quick enough. It’ll execute more steps without the bit so you would need to go back to the last step before the bit broke. Then you would need to understand why the bit broke. If it’s your G code has the depth too deep then there’s no point going forward.

If you want to pause a job then you just need to press pause and wait for the buffer to clear out. Even if 1F can make a way to lock out any commands going to the processor until the buffer clears it’s still a wise idea to wait.

Rehoming isn’t really needed unless you turn the machine off. Zeroing can be a problem if you’ve removed material from the point where you dis the initial zero. This is how I’ve done projects that are going to take hours but I don’t want to let it run over night while I sleep. I’ll mount a square block on the work table that’s not part of the project off to the side and then mount a piece of 1/4" plywood that’s an L shape near it at a known offset. Since neither is going to be machined I can shut down the 1F if I want and even remove the project and set it back up exactly where it was.

How exactly can you do this? Go back to the last step…or any point in a program?

The only way I know how would be to edit the Gcode and reload it. Not ideal and in the case of a broken bit I wouldn’t know of how to know exactly how far back to go.

Honestly I’m thinking of getting rid of my Onefinity due to this ridiculous bug. I’m so tired of this software creating headaches for me. why can’t this be fixed?

I think the OF bb controller is ancient news now. Within 6 months of my purchase they came out with the masso. I was thinking I should have waited, I never expected them to switch to a completely different system. I’m still happily using the 1.6.3 firmware.

I’m in the process of planning the switch to Fluidnc on an ESP32-S3. For the same reason - being able to pause/jog/resume. You can apparently keep ā€œstall homingā€ too. Still need a computer, phone or raspberry pi5 to access Fluidnc via web browser. I’ve been holding off hoping they would update support for the newer C5 with 5Ghz wifi, but they have no plans to.

It’s that or upgrade the raspberry pi to ver5 and run Linuxcnc. Linux means adding homing switches/sensors too, not to mention the learning curve, sadly. Either way it means new internals for the controller. I’m starting with trying the ESP32-S3/Fluidnc on my cheap laser burner, if that works well, I’ll figure out the stall homing and try it on my journeyman. I’m in no rush… unless the bb dies. :slight_smile:

Upgrading to the realtime controller with redline HMI through our IPP program is a great option as will!