On Monday 27 August 2018 09:49:50 Les Newell wrote: > After a few more hours head scratching I figured out most of what is > going on. In some cases LinuxCNC is calculating the work envelope > incorrectly. The axis limit warnings are based on the work envelope. > From my testing it looks like it is sometimes messing up the tool > offsets when it does a simulation run. > > I can now replicate this bug using a sim configuration. Create a > sim.axis, axis_mm configuration and drop in the attached ini file > (only the Z axis extents and home position are changed from standard). > Start Axis and home it. In MDI: > M6 T1 > G43 H1 > Jog down to the Z minimum limit then jog up a bit (say 10mm or so). > Tool touch off Z and set it to 0. Load the attached ngc file and run > it. The file will run correctly. Turn on View->show extents. Pay > attention to the Z axis size. It should be around 80 - 90mm. > In MDI: > M6T1 > Now reload the program and run. You should get a warning 'Program > exceeds machine maximum on axis Z'. If you click 'Run anyway' it will > run correctly. Look at the work extents. Z is now around 170- 180. > > Restart Axis, home and load the program. It will run correctly. In > MDI: M6T1 > Reload the program. Now you get the error. > > This configuration is specifically set up to show the error as clearly > as possible. However you can probably see it in most configurations. > Set up a fairly long tool offset and use code similar to my ngc file. > The important parts are the G43 followed by one or more G1 moves, > including the Z axis. Load the file and look at the Z extent. Now in > MDI M6Tx (use whatever tool number is relevant) > Reload the file and you will see the Z extents have changed. > > Les
A me too post: Les, I am glad you looked into that, because I've had what sounds like the same problem for a long time, years in fact. I could always reboot, restart, or rehome and eventually get it to run. But cumulatively, half an hour at a time, once, maybe twice a week, its wasted days of my time. So I've even gone to setting a -z in the .ini file that will drill clear thru the table w/o fixing the z exceeds minimum warning/stop. Having to click to clear the error for every step when single stepping looking for the error is also a pain. When single stepping, we ought to be able to click a step until the next move is aborted because it sees the limit error during that move. Thats likely a different fix, but it sure would be a welcome improvement. Put that on the wish list. :) So would having an axis/pyvcp display of what the machine thinks it has for clearance, so with the ability to park a just mounted drill bit, on the table, and set that z limit dynamicly with a spin button to stop it before it drills yet another hole in the table. I have several of those already. A better way perhaps might be to measure the contact height above the table, then G38.2 over the contact pad and use that data to move the z co-ord. Since I use different sizes of drills when making the tap hats, that is exactly what my code does. And that has worked well, but no tool goes below about 5/8" above the table the way I currently have the forest of jigs mounted. So far I've made 50 some of those tap hats, but haven't made any for over 3/8", the torque to turn the bigger ones just isn't there, plus the heads backgears are plastic. Its also making me vaguely aware that my post isn't truly perpendicular. Leaning fwd about a degree, and maybe sideways a red hair. At one point in making the hat, its mounted in a 7/8" r8, and a chuck is fixed to point up, so I am lowering the brass slug to drill the hole thru it on center, that allows the tap shank thru it. But either the drill chuck isn't standing straight or the post isn't as I am getting a tapered hole. But, it works good enough to get the job done. -- 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> ------------------------------------------------------------------------------ Check out the vibrant tech community on one of the world's most engaging tech sites, Slashdot.org! http://sdm.link/slashdot _______________________________________________ Emc-users mailing list Emc-users@lists.sourceforge.net https://lists.sourceforge.net/lists/listinfo/emc-users
