On Saturday 23 March 2019 04:21:22 Lester Caine wrote: > On 23/03/2019 02:56, Gene Heskett wrote: > > On Friday 22 March 2019 18:52:54 Lester Caine wrote: > >> On 22/03/2019 22:46, Bruce Layne wrote: > >>> I use FreeCAD, mostly for 3D printing. It's still not ready for > >>> prime time and doesn't have a usable CAM system so it's not a > >>> possible replacement for Fusion 360, but it's a viable CAD option > >>> for me when Fusion 360 isn't. > >> > >> Sorry I have to disagree with that ... I'm running perfectly good > >> gcode for the Taig mill from FreeCAD. Yes for 3D printing one of > >> the other slicer options is required, but for 2.5D machining it > >> works well enough and is improving all the time.
How are you getting the gcode? Its not yet listed as an export option. > > If its being improved Lester, why has the downloadable been stuck at > > version .14 for several years? I take it back, the last 32 bit > > release that will run on the wheezy LiveCD install is .14, but I see > > .17 is available for 64 bit installs now. So you have to reboot to a > > 64 bit install to use it. Is it worth setting up a dual boot? > > Actually I'm running 0.17 on the main linux desktop as the SUSE > experimental load of 0.18 screws up other CAD and graphics related > stuff on Leap15.0. 0.18 is running on my windows machine with Vcarve > and TurboCAD and installed without a problem. I dropped KiCAD onto the > windows machine as well so I could play with the recent 'paid for' job > and while not totally smooth, I had models of the folded sheet metal > around the PCB along with a flattened out version for the laser ... > but it ended up being punched out on a CNC punch ... so I did not see > the gcode for that. The final box was perfect. > > My next job is to get the 3D printer hooked up to the BBB/CRAMPS board > and MachineKit driving it. Although I may get the lathe finally hooked > up first since the the proprietary control does work. Just can't add > bed levelling to it :( Thats a bummer. I'm doing a "make do" version of that on my old Sheldon. What little uneven wear there is, which has the effect of tilting the carriage to/from the center axis, I've measured and put into a screw comp file that profiles the error into the x axis. Seems to work ok, but the max correction is less than 3 thou. Some right at the mouth of the spindle isn't wear, but the carriage being pushed slightly out of square by the force the left hand bellows over the screw exerts on the carriage when its being seriously crushed. But thats effectively inside the chucks. That 1450mm Z screw should have been 1500mm with 40mm of that used to move the left thrust bearing to the left. Hindsight, always 20-05... The fix now? Would be tapered gibs on both front and rear of the carriage. But that would be quite a chore at this stage. It sure made a whole new lathe out of TLM though. One thing I did on the Sheldon that I couldn't do on TLM, was to seal up the X screw to keep swarf out of its nut. But with the motor on the rear of the carriage on TLM, theres no room for an effective cover to telescope, so I've crushed 3 attempts to make one so far. The Sheldon X motor is on the front, and the taper adapter's crossfeed extension stuff makes sealing that area up a piece of cake. I inlet a piece of 1/8" alu paneling to cover the slot above the screw, and similarly covered the bottom so an errant air hose blast can't get to it. Even had the back of the carriage covered for a while. X backlash got hugely gross a few weeks back and I thoought I was going to have to tear it down again to get at the x clamping nut screws, but it turned out the drive pulley was slipping on the shaft. Tightened the setscrews till the wrench screamed again, and gave it a coat of green locktite at the pulley/shaft joint, back to about a thou. :) The problem with the pi turned out to a a badly mis-configure G76, giving about 50,000 passes, so the poor pi was plumb out of swap trying to render the backplot. Fixed that, pi back to normal for the pi. However, that threading job disclosed that G76 needs one more parameter. Due to the length vs diameter of this 10-24 thread, G76 needs an additional parameter to cause it to cut a couple thou deeper at the right end. Because of the cutting force, it cuts the right, unsupported end of the thread larger than at the left end, very very noticeable when attempting to get a good fit of stylus into the threads in the plastic face of this cheap probe. It is possible to cheat by makeing the left end withdraw movement over one turn less than the length of the thread, you can make a nice 40 thou taper over 5/8" that way, but I haven't tried it for this small a taper. I have made intentional tapers that way on TLM. Very successfully, that is whats driving the X screw in the Sheldon. By making a socket on the rear of the new shaft I put in the X hand crank, with the socket area being tapered to match a custom 50 tpi tapered nut, the socket EDM'd to make 6 petals out of it, the screw is set into the socket, some green loctite applied, and the nuts drawn tight to crush the petals onto the screw, its effectively one piece of steel. Since that former hand crank is now the X thrust bearing, its now equipt with needle shaft bearings and ball thrust bearings for zero backlash there. But the pulleys setscrews didn't bite into the new A2 shaft well enough, hence the recent addition of some green loctite. 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) Genes Web page <http://geneslinuxbox.net:6309/gene> _______________________________________________ Emc-users mailing list Emc-users@lists.sourceforge.net https://lists.sourceforge.net/lists/listinfo/emc-users
