On 4/12/2015 2:39 PM, Gene Heskett wrote: > A dim led just came on in my mind, and I believe the HOME_OFFSET > setting might be able to do that. Its intended to allow a precise home > position to be set relative to the switch closure it finds, allowing > the switch to be mounted at an arbitray position, and I believe you > could use that to establish the found position to be -20.00000000 that > you quoted. I am assuming its a metric machine since -20 inches is > beyond the ability of most machines under 5 tons weight to reach. > There are not too many machines extant that equal Stuart S's big > Cincinatti. :-)
So I think this did the trick... I set HOME_OFFSET = -20.0 (yes, it's it mm). I have X & Y set for automatic homing, and Z & A will be manually homed, but I have HOME_SEQUENCE set so that they get homed so that I can use MDI commands. The command steps are as follows: - Boot up - Run LinuxCNC - F1 - F2 - Jog Z to clear the table/components (*1) - Ctrl-Home (homes X & Y, moves Z up 20mm, then sets Z to homed/0, and sets A to homed/0) - G01 X.. Y.. F.. to location where I know the Z-height. - Jog to touch the top of the part - Home Z (moves it up 20mm from touching the part and sets it to 0). From here, I can jog Z up down correctly (as per limits). *1 = If I jog too high, then when I get to jogging downwards to touch the part, I need to unhome first as -Z soft limit is reached.. What I'd really like to do next is to simplify this procedure as I'd like to have someone else setup the machine and start a job. Which is why I'd like to automate all of this. I do have a home switch on Z, but I know it will not be accurate enough. For X & Y, a camera will correct for any error. > Another thought crossed my mind too, the removes the need to > electrically isolate the workpiece so an electrical contact can be > detected. Add the thickness of a piece of PCB, copper coated one or > both sides, and connect the top surface to the z home switch input as > assigned in the breakout board. Clean well the bottom of the PCB, and > the top of the workpiece and place the PCB, active side up, on top of > the workpiece. With care about the cleanliness, that should repeatedly > get you to within a thou if the top of the workpiece is indeed flat > and level to the tables motions. I am not making new glasses for the > Hubble telescope here, so that would be "Good enough for the girls I > go with", as one (now ex) bro-in-law was famous for saying when it all > worked. I've been thinking about this and other ways (laser or vision) to accurately set Z automatically, and it would require some hardware mods, so that will come after I figure out my real-time kernel issues. FWIW, I will eventually have some spring on the PCB mounts, but the feeders (basic drag feeders for now) are finnicky about height. Thanks, -Neil. ------------------------------------------------------------------------------ BPM Camp - Free Virtual Workshop May 6th at 10am PDT/1PM EDT Develop your own process in accordance with the BPMN 2 standard Learn Process modeling best practices with Bonita BPM through live exercises http://www.bonitasoft.com/be-part-of-it/events/bpm-camp-virtual- event?utm_ source=Sourceforge_BPM_Camp_5_6_15&utm_medium=email&utm_campaign=VA_SF _______________________________________________ Emc-users mailing list [email protected] https://lists.sourceforge.net/lists/listinfo/emc-users
