Greetings all; This could even be considered off topic, but here goes: A lot of us use a slicer like cura to turn OpenSCAD output into additive gcode. Works quite well in fact. But now I have a need to invert that gcode. It would need to strip out the thermals and such that LinuxCNC would not understand, but would read cura's output backwards, from the top down IOW, issuing the gcode commands to remove in cura's layer sized depths of cut, that which is in a solid starting block but which is not printed, going around anything that cura deposits. Has anyone considered, or knows of such a beast which could turn OpenSCAD's output into gcode for a subtractive machine is what I'm trying to discover?. Recommendations please. I have run into a fan failure on my last working printer, and I have a reducer if not an eliminator of the cause grokked, a piece of printing debris was sucked into the hot end fan and stopped it, and the hot end in turn got warm enough to destroy the carriage. But now I need to duplicate the carriage I designed and put on my ender5+, made from PETG+CF, but carved this time from alu. I have the machine to do the carving. Thank you everybody. Cheers, Gene Heskett. -- "There are four boxes to be used in defense of liberty: soap, ballot, jury, and ammo. Please use in that order." -Ed Howdershelt (Author, 1940) If we desire respect for the law, we must first make the law respectable. - Louis D. Brandeis Genes Web page <http://geneslinuxbox.net:6309/>

I've been working on a system for making G-code for subtractive processes: https://github.com/WillAdams/gcodepreview Might be useful as a starting point. William

HI Gene, If you have the original stl, I may have something that will help. If you are thinking of converting existing fdm type gcode, in the way in which you say, you would need a milling cutter, more or less the diameter of your extruded filament. If you designed the shape for the carriage, then most likely it is 2.5d, or could be made so, then easy enough to work with dxf and profile/pocket as required. Cura and others can load g-code files and attempt to export as stl, which may or may not work. DeskProto, for example can work, sort of, from stl files, but i do not think that is linux, I do not use linux, so none of my software will help you, but if a simple shape, and a carriage for an ender should be simple, I could have a go at processing your files, and send you the g-code. Best wishes, Ray On 27/11/2022 21:08, gene heskett wrote: > Greetings all; > > This could even be considered off topic, but here goes: > > A lot of us use a slicer like cura to turn OpenSCAD output into > additive gcode. Works quite well in fact. > > But now I have a need to invert that gcode. > It would need to strip out the thermals and such that LinuxCNC would > not understand, but would read cura's output backwards, from the top > down IOW, issuing the gcode commands to remove in cura's layer sized > depths of cut, that which is in a solid starting block but which is > not printed, going around anything that cura deposits. > > Has anyone considered, or knows of such a beast which could turn > OpenSCAD's output into gcode for a subtractive machine is what I'm > trying to discover?. > > Recommendations please. > > I have run into a fan failure on my last working printer, and I have a > reducer if not an eliminator of the cause grokked, a piece of printing > debris was sucked into the hot end fan and stopped it, and the hot end > in turn got warm enough to destroy the carriage. But now I need to > duplicate the carriage I designed and put on my ender5+, made from > PETG+CF, but carved this time from alu. I have the machine to do the > carving. > > Thank you everybody. > > Cheers, Gene Heskett.

On 11/28/22 10:56, Raymond West wrote: > HI Gene, > > If you have the original stl, I may have something that will help. If > you are thinking of converting existing fdm type gcode, in the way in > which you say, you would need a milling cutter, more or less the > diameter of your extruded filament. If you designed the shape for the > carriage, then most likely it is 2.5d, or could be made so, then easy > enough to work with dxf and profile/pocket as required. > > Cura and others can load g-code files and attempt to export as stl, > which may or may not work. DeskProto, for example can work, sort of, > from stl files, but i do not think that is linux, I do not use linux, so > none of my software will help you, but if a simple shape, and a carriage > for an ender should be simple, I could have a go at processing your > files, and send you the g-code. > > Best wishes, > > Ray Hi Ray; Yes, I have the original in OpenSCAD so I can "export" whatever would work best. But the translator would need to adjust for the size of the tool, which in this case would be a 1/16" flat or round nosed carbide cutter, and I would have to devise, for some shapes, a reliable method of registering the other side, or perhaps even a 90 degree turn of the work piece. That is my problem to solve as the 6040 is presently equipt with a 7 degree wedge under the spindle mount to carve buttress threads in hard maple with the side of that tool. Changeable of coarse but there is a half degree or so of backlash in the B axis I built. Not enough to affect thread cutting, but would this for sure, so limiting me to simply turning the work piece over to do the other side. I've done that on another mill accurately enough to make holes halfway thru a pcb meet in the middle of the boards thickness with no offsets but its historical, involved a pallet with a touchoff target in one corner and a bottom fin to engage the table slot for axis alignment. Unforch, the 6040's alu extrusion bed is neither flat nor level. So a pallet for the pallet would probably be required. This mill has ceramic spindle bearings, so that scheme would require I fab a brush to ground the spindle while running the location gcode. Sorta a pita. Another project for OpenSCAD. And capable of being articulated by gcode so it is engaged or gotten out of the way for tool changing access. ER-20 spindle with a dirt slinger above it Question, have not tried it. How about I tell cura the filament is 1.5875mm in diameter, not 1.75mm? that way the filament and the mill would be the same size. Or cura has, IIRC, an offset correction that might work, IDK. This thing is basically single sided, the risers are 5 total, various heights, and one would require later drilling and hack sawing open the adjustment slot for an inductive ABL sensor. However, that alu is only 12,6mm thick so that riser would have to be bolted. I can remove it from the part That however is my problem and I could remove that from the file I send before the final F6. If I render in .12mm layers, I'm pretty sure the tool would survive digging that deep per pass in the alu I have. I have mist cooling available on that mill, and that tends to blow away most of the swarf. 2 pieces of it would be thicker than the stock, but I can drill and bolt that stuff easy enough. And, with the fixed length of the bolts thru the POM wheels, a thinner remainder would be fixable in the stl before the F6. So the alu piece could be reduced in complexity quite a bit. I could take the finished piece to the other, bigger mill to drill and tap the holes. In that case, I'd make 3 so I could drill and mount it on a linear rail slider as I've carbon fiber tubing on a rowboat in the middle of the big pond to reduce the weight the Y motor has to throw around by 3 or 4 hundred grams. We should probably take this to PM though. This is not directly an OpenSCAD problem other than its lack of subtractive mode gcode export. The exact opposite of cura's bottom up, we need a top down approach, and that is a whole new can of worms. Subject to that rule that says once opened, it takes a 2 pound can to hold what did weigh a pound before looking into it. ;o)> Thank you Ray. > > > On 27/11/2022 21:08, gene heskett wrote: >> Greetings all; >> >> This could even be considered off topic, but here goes: >> >> A lot of us use a slicer like cura to turn OpenSCAD output into >> additive gcode. Works quite well in fact. >> >> But now I have a need to invert that gcode. >> It would need to strip out the thermals and such that LinuxCNC would >> not understand, but would read cura's output backwards, from the top >> down IOW, issuing the gcode commands to remove in cura's layer sized >> depths of cut, that which is in a solid starting block but which is >> not printed, going around anything that cura deposits. >> >> Has anyone considered, or knows of such a beast which could turn >> OpenSCAD's output into gcode for a subtractive machine is what I'm >> trying to discover?. >> >> Recommendations please. >> >> I have run into a fan failure on my last working printer, and I have a >> reducer if not an eliminator of the cause grokked, a piece of printing >> debris was sucked into the hot end fan and stopped it, and the hot end >> in turn got warm enough to destroy the carriage. But now I need to >> duplicate the carriage I designed and put on my ender5+, made from >> PETG+CF, but carved this time from alu. I have the machine to do the >> carving. >> >> Thank you everybody. >> >> Cheers, Gene Heskett. > _______________________________________________ > OpenSCAD mailing list > To unsubscribe send an email to discuss-leave@lists.openscad.org > . Cheers, Gene Heskett. -- "There are four boxes to be used in defense of liberty: soap, ballot, jury, and ammo. Please use in that order." -Ed Howdershelt (Author, 1940) If we desire respect for the law, we must first make the law respectable. - Louis D. Brandeis Genes Web page <http://geneslinuxbox.net:6309/>