once again I would like to give you a short overview of the latest developments on the programme.
This time I have been working on the production of wings for aircraft modelling.
I confess that I don’t really know anything about this and therefore depend on tips and suggestions for improvement from the modellers among you.
In the short video I show how to simulate the GCode of the program JediCut in GrblGru.
But I have also made a “WingCreator” with which you can construct wings yourself. I would like to tell you more about this in the next video.
I am just discovering GRBLGuru searching for “jedicut” in this forum.
I have the feeling that you just made a beautiful piece of code
I am a RC flyer and builder and I can have access to an old foam cutter using “MM2001” controller board (with parallel port controller).
My intention would be to retrofit this machine with a more modern controller board “FluidNC” dongle.
Coming back to your software. It seems that you can read gcode files generated by jedicut and send them to GRBL controllers ? So it should also work with FluidNC controllers right ?
Foam cutters have a particularity to handle “wire temperature”. I don’t exactly know how this is done into Gcode !
A few information xan be found here (but in french) :
Connection is Ok, the FLuidNC board seems to be seen. But imediately I get this error
I can’t find this line into the gcode file ?
So probably something is missing for FluidNC board to be recognized ?
I know that there are solutions to add “missing parameters” but I don’t know what GRBLgru is waiting for during initialization !
FluidNC does not support the U option on G10 commands. Hence the error message.
Maybe your example config is for a different Firmware?
See: Supported Gcodes | Wiki.js
You will need to edit the job start gcode and correct the G10 line. Maybe also sanity chack the rest of the output too. FluidNC is /essentially/ GRBL 1.1 compatible, so look for GRBL examples if looking at example configs online.
Something I’d not seen before… I’m guessing the ‘G10’ line is designed for that… It seems unlikely that GrblGuru is generating that line since it’s ‘non standard’ ?? (or maybe it is a Grbl Mega thing?
I dont know GrblGuru at all (Laserweb is my thing) but it sounds like you need to configure the Axis mapping somehow? Sorry I cant help further; needs an expert.
I am a little late in this discussion, and perhaps you have already found what you were looking for in the “Grbl hotwire GCode” program.
Regardless of that, I just want to briefly explain what my goal was in developing the Foam Cutter machines and what the current status is.
Originally, it was only intended as a graphical simulation of GCode, which was calculated by other programs (e.g. JediCut). But then it turned out that there was also a need for ‘normal’ cutting of Styrodur plates. (e.g. cutting out letters)
I then started to generate the corresponding GCode from normal DXF /SVG files.
As I don’t have such a machine myself and can’t carry out any tests, I am dependent on external help. Unfortunately, no one has been able to help me so far. The current status is therefore an unfinished construction project.
Now to FluidNC. Here, too, I do not have a suitable controller. The control of GRBL, Mega-5X and GrblHAL controller works, so I assume that FluidNC will not cause any unsolvable problems.
The GRBL error message 20 states that an unknown axis has been controlled. I assume that this is the U axis. Axis names can be mapped under GrblGru in the MachineManager in the “Mapping” chapter.
So if you are interested in more GrblGru details, I would be happy to contribute anything you need. However, it would be even better (for me) if you would introduce me to the ‘secrets’ or at least the requirements of the Hot Wire Cutting.
Long story short:
But it would also be ok if you have already found the “Grbl hotwire GCode” program in a roundabout way.
And sure, I can try to explain to you what hotwire cutting secrets are (although I am quite new also to this hobby ! (but my brother is skilled in foam cutting ))
Once again thanks for your reply and your software
Ok, I would suggest that we first try to move the axis manually.
Please select “GRBL-Mega-5x” under Controller and press the small “R” button to the right of the input. This will reset the controller. It should then be ready in the bottom right-hand window as shown in Pic 1.
Then go to the left window and set “Command line” on. Now you can enter commands in the left window and FluidNC will answer you in the right window.
Please enter “$$”. All parameters should now appear in the right window. (Similar to pic2 ) Please send me a screenshot of this.
Now please set "Command line"off, so that the normal buttons are active again.
Now try to move the individual axes with the arrow keys. pic 3
Unfortunately, there are still problems with the mapping of the U-axis.
Thanks to your help, I was able to fix some of the errors in the new version V5.2.6.
However, the A(U) axis cannot yet be moved in the simulation. But it should work in real mode.
I am working on the error in the simulation.
In the meantime, could you please check whether the axes and the model now move with V5.2.6.
Thanks for the info
Mapping a translational axis into a rotary axis has turned out to be more difficult than I initially thought. I’ll get back to you here as soon as I have a new version.
Do you have the necessary hardware to carry out real practical tests? Or are you still missing something ?
Currently I don’t have the hardware totally ready. The hotWire machine I use is a very old MM2001 board connected via a // port to an XP PC…
My plan is to replace all the electronics by FluidNC board (I know it can work !) But it’s a matter of time… This machine is a “standard” 4 axis foam cutter
The hardware for this is almost ready as I have a 3 axis CNC driven by FluidNC and I can easily add a rotary axis. This CNC is already equiped with a hotwire cutter (vertical cut fixed on the Z axis).
But I would like to use this turn table to shape “fuselages” of my planes. More or
less like the video BUT with a non constant profile… The profile should be changed anytime the rotary axis moves. I could cut 16 or more faces to approximate the shape of the fuselage, then finish by sanding.
The workflow could be:
project the shape of the fuselage on horizontal plane, select half of this profile and send Gcode to follow it by a vertical hotwire
turn the rotary axis and repeat
I could be sure that no collision would occur as a fuselage is more or less a “bottle shape” but with a kind of ovoid section !
BUT I can’t find a software able to extract the profiles from a 3D model and then generate Gcode… If any idea ?
I have just uploaded a new version V5.2.7. The mapping should now also work. However, I have changed the functionality a lot. The GCode is now always generated with the axes of the model. The conversion of the axis letter now only takes place when the GCode is sent to the controller. I hope this has no other disadvantages.
By the way, I have already included the control of a turntable in the program. I am very confident that I can also calculate the GCode for this.
For now, however, it would be important for GrblGru to be able to control the 4 linear axes properly. With version V5.2.7 it should now also be possible to load DXF files and cut them out.
There are certainly still some bugs, but the essentials should actually work. I would be very interested to know if this works. Can you please check this?
