Use LightBurn on Pi with a small touchscreen

I’m using a Raspberry Pi to control my CNC system. Running it from an actual workstation is a royal pain, since it means leaving a long USB cable on the workshop floor. (For various reasons I don’t want to run the cable through or along the ceiling.) I do my design in LightBurn on my Mac, then save the G-Code and go to the Pi (with a 7" touchscreen) connected to my Shapeoko 3 and load the G-Code in UGS to do the work. (I’ve tried gSender, which has been recommended, but it crashes on my Pi - I’ve reported the issue and am waiting for them to compile a new version.)

Has anyone tried using LightBurn on a Pi and with a smaller touchscreen? I’m thinking if I installed an instance of LightBurn on my Pi, I could change the size of the panes in the main window so the screen would be filled with the toolbar and the Laser pane. That would let me control the CNC with LightBurn directly without the cable hazard.

The only issue I see is that I can’t find a way to load the G-Code file into LightBurn to run it. I’d have to save the project and load the project. That doesn’t seem like too much of an issue, but it’d be nice if I could just work with the G-Code on the Pi.

Has anyone tried this and found that using LightBurn on a Pi, mainly just as a controller, works? Or is the Pi underpowered for LightBurn? And if I use just the Laser pane and toolbar as a user interface, is there anything critical I’m losing there?

There is a “run gcode” in Laser window. It immediately will send gcode to the connected laser.

LightBurn won’t run on a Pi.

That’s for the project already loaded. I was hoping to load just the GCode to run, but since LightBurn won’t run on a Pi, that won’t be a problem anyway.

That’s not the case. It will run whatever gcode file that you load. It’s not related to the existing project. And again, it runs it immediately. There’s no opportunity to “review” the gcode or otherwise visualize it.

There are other miniature x86 architecture boards that you could potentially use for this. Or else using some other gcode sender from the Pi would also work.

But how do I load a GCode file? I’ve tried “Import” and the GCode files in the folder are greyed out.

I’m also looking at ways to do some of the things I’d like to easily without having to change the hardware I have.

(I’m also looking into USB over IP. There are commercial apps for this. The problem is they take over the USB port completely, so I couldn’t have that working on the Pi, but still use that port locally for UGS or Carbide Motion. I’m trying to find out if I could start and stop USB/IP as needed.)

This morning I sat down and looked over things and thought, “What can I do, that I need to do, with LightBurn directly connected to my CNC that I can’t do otherwise?” After looking everything over, I think the answer is the one function I asked about last night (in the link above), and that’s being able to outline the image so I can use that for placement of the work materials. Otherwise, I can use UGS to run GCode, just like LightBurn, to turn the laser on low and off (with a macro), and to do what I need to do for doing this kind of work.

So, for now, I’m thinking if I can just get that problem solved (drawing the outline), then I can do what I need to do with things as they are.

Push the “Run gcode” button in Laser window. Then pick the file.

What Pi model are you using? Could you dedicate specific USB ports for different devices possibly?

I like this approach. It’s outcome focused rather than getting caught up in the process itself.

Ah! I had just thought that button did the same as the Start button did - that it ran the GCode from the current project. Thanks!

A Pi4. Can’t remember - is there an A and B for the Pi4? If so, it’s a B. (So frustrating, these days, with Pis costing so much and so hard to get!)

I know very little about how Linux handles USB devices or how things work in the /dev directory. I have wondered if it’s possible to create an alias for a device and do that, but I would think if the original device was claimed and locked, that using an alias would not help, since it’d connect back to the locked device.

I’m setting up two 3D printers (one took 5 months to just print a clean first layer!) and the CNC with the laser. Part of the problem is linear footage - not square footage, but the total wall space. I have tools like a drill press and more, plus a workbench, computer table, and the scrap pile against the walls. I put the CNC in an island in the middle, which means no USB cables to it. The networking cables to it have to be fiber because we’ve had lightning issues with cables above the ceiling getting blown out. So it’s an unusual situation.

It’s just taking time to work out the best control methods for everything. My original plan was to use USB<->ethernet converters, but it turns out those use an ethernet cable but NOT ethernet protocol, so they won’t work over fiber cables. So I’ve gone through a lot of steps to deal with this and have had to back up and re-evaluate continuously. One factor has been that Pis are hard to get ahold of now and are quite expensive, so I was hoping to set up some kind of remote connection that didn’t involve USB wires on the floor or running things through the ceiling where, in the past, lightning has been an issue.

With trying to set up all these devices, I was hoping for a one-size-fits-all solution, but it’s taking a lot of time to go through and work out what I can do for each one individually.

I’ll warn you again just so you know, as soon you select the file gcode is going to get executed. This catches people off guard as they’re expecting a multi-step process.

No joke. I’ve been literally trying to source a Pi4 for over a year. Not willing to pay scalper markup.

From what I remember each usb port will be assigned a distinct device so is uniquely addressable. This should allow you to dedicate a different port per task but I haven’t tried this myself.

Can you go vertical?

I suspect wireless to Pi with USB as a general solution could work if they were readily available. This would work with any portable computer if you have any lying around. Maybe as a stop gap until Pis become more readily available.

Thanks! Yes, that’s important to know!

I got one a few months ago because I needed another Pi4, but, boy, was it expensive. On the other hand, it was one of the few I could find and it was a much better case/heatsink/fan combo than most I’ve seen. When they’re no longer in short supply, I’ll probably be going with that case and kit when I order.

What I find frustrating is that it’s hard to find a “critical part” kit without extras. One friend has said he’s seen so many fans burn out he doesn’t add them anymore, he just uses heat sinks. I suspect I could save something on them if I could get just the Pi, power supply, and heatsinks, but all the kids I’ve seen also include fans and cases. I’d rather just print a case, but I haven’t seen heat sinks and power supplies bundled.

Right. But I only have one USB connection to my CNC, so I’d need some way to create an alias. The other possibility is that I might be able to activate and deactivate USB/IP with a simple shell command or script. That’d let me turn that on and off as needed and I could link a script to an icon so it’s easy to run on a touchscreen. But I would also need something that would indicate to me whether it’s on or off. (Maybe a simple Python script that opens a window saying, “USB/IP is ON” and then gives me the option of turning it off - something like that.) (And that’s why it helps me a lot to talk about issues on forums - if I weren’t explaining it to you, I would not have thought of that option.)

That’s an excellent suggestion. I can’t really. Maybe, if I end up with a larger number of 3D printers to produce products, I might make multi-level cases so I can stack 'em, as if they were on shelves. One thing I’ve done, which I love and which has been a major organizational help, is using peg boards on the walls. That takes up some space, but I can usually put things at a good working height and still have the peg boards in place.

There is a USB/IP package out there in Debian, so it’s in Raspbian. The problem is it’s only Linux to Linux. There are a few companies out there that have created their own. That makes me suspect it’s not too hard to do - if you know what you’re doing. I’ve asked about this in the Raspberry Pi forum, but there’s one user there who keeps saying, “It won’t work, it won’t work,” even though he doesn’t have any clear evidence of it. He keeps saying, “Do this instead,” and I’ve had to give him even more details than I’ve given you about my setup to make it clear why his suggestions won’t work. I find a lot of that thread is more about telling him to back off than actual discussion of the issue. I dread his posts because he doesn’t listen to what I’ve said and I hate having to keep giving what I consider boring details of why I’m trying to do things the way I am.

I’m going to waste a day or so when I can just to get USB/IP working so I can say, “It works. Works just fine. But then again, I made my evaluations on evidence, not on jumping to conclusions without my critical thinking skills or ignoring what people have told me multiple times.” He’s done this kind of thing on several of my questions and it’s hard to ignore him, so I don’t even use the Raspberry Pi forum anymore.

This topic was automatically closed 30 days after the last reply. New replies are no longer allowed.