On Monday 07 December 2009, dave wrote: >Hi all, >For once I'm going to crow rather than eat crow. > >Discussions with Matt Shaver yielded an idea for a probe that was not >the classical Renishaw ball and bar design. > >To wit: >Drill and ream 0.187 holes in a disk on a 2" bolt circle and press >0.1875 ground dowel pins into the hole. Arrange so they protrude >approx .3". Drill a hole in the center of the disk and tap 1/4-28 to >attach the probe. You will note set screws in the pic since I didn't >have an undersize reamer. > >Mill holes 0.8" dia in another disk on the same bolt circle. >Make cups in .75 dia x .5" cylinders by drilling with a #2 >drill/countersink. Put teflon washers in the holes in disk and float the >cups in epoxy using the dowel pins as a jig to align the cups. >Tap the cups with a 4-40 to attach electrical connections. > >The logic utilizes the inverting inputs on a hal logic 3 input OR to >digital inputs on the stg card. A wire off the disk with the pins goes >to digital ground. No pullups were necessary. > >see > > >http://imagebin.ca/view/UHmqcPz.html
Needs one or 2 more pix Dave, one with the probe tip installed, and one of it installed one the mill ready to work. The logic I can get, and using a 3 input gate certainly simplifies the probe. Not clear in this pix but I assume the pins being held by the set-screws are also brass? Tip profile? A silver plate kit might make it even more accurate, but this is certainly tolerable accuracy indeed. :) >Now for the proof of the pudding. > >Used G38.3 probe a ground surface to check Z repeatability. G code was >provided by a C program that simply looped. (PROBEOPEN dataset_name) and >(PROBECLOSE) were used to accumulate the data. >AFIK if one uses a path in the dataset name then spacing between the >PROBEOPEN and the dataset name is not critical; otherwise extra spaces >i.e. > 1 will be prepended to the data set name. Ask me how I know. ;-) >For each test reps = 30. > >In like manner I probed the bed at 1" intervals and also used movements >in X to probe the side of a ground bar. > >Note: resolution of the axes are 1 um in X and Y and about 1.5 um in Z. > encoders are on the ball screw for X and Y and on the servo motor >for Z. > >Results: std dev. >X probe: 0.00022" >Z probe: 0.000267" >Bed probe: 0.000634" > >I think I can live with these numbers for a shop-built probe. ;-) And they are better than the optical, z only gizmo I made following directions from someone here (Matt Shaver?), that seemed to have a repeatability of maybe a fat thou. I had figured to use math to smooth the curve if and when I got around to finishing the outside face of the floor plate I have about 2//3rds finished for my old Akley-06. Using a 1/4" ball nose, I was going to retrace the map of the old one on about a .005" pass increment. But I've miss laid my round tuit. The idea was to replace the whole magazine well, trigger guard and floor plate with alu parts eventually, which would lighten the whole thing by half a pound or more. As is, it's nearly 14 pounds loaded and hanging on my shoulder. Laminated walnut is dense & P14's never were noted for being light by the time a medium heavy 26" Douglas is screwed into them and a big variable scope mounted. >Many thanks to those hard working (and I might add bright) people that >created the various components that made this this implementation >possible. > >Dave > I'll second that. Loudly. -- Cheers, Gene "There are four boxes to be used in defense of liberty: soap, ballot, jury, and ammo. Please use in that order." -Ed Howdershelt (Author) The NRA is offering FREE Associate memberships to anyone who wants them. <https://www.nrahq.org/nrabonus/accept-membership.asp> "How many people work here?" "Oh, about half." ------------------------------------------------------------------------------ Join us December 9, 2009 for the Red Hat Virtual Experience, a free event focused on virtualization and cloud computing. Attend in-depth sessions from your desk. Your couch. Anywhere. http://p.sf.net/sfu/redhat-sfdev2dev _______________________________________________ Emc-users mailing list Emc-users@lists.sourceforge.net https://lists.sourceforge.net/lists/listinfo/emc-users
