On Thu, 2013-06-20 at 12:41 -0400, Gene Heskett wrote: > On Thursday 20 June 2013 11:20:45 Jon Elson did opine: > > > Gene Heskett wrote: > > > 2. Set up a choice of homing sequences, one for a normal, cuts on the > > > far face turning tool, and one that inverts the X homing sequence for > > > calibrating a boring bar. And establishes the two different sets of > > > HOME_OFFSET because a different set of contact faces on the gauge are > > > used. > > > > Why would you need a different homing sequence for different tools? > > I can't see why you need this. Is this because you have tools on > > the back side of the part, and you want CSS to work? Or, is it just > > to make the X coordinates run the right way? > > Interesting problem. I'll bet there's a better way to do this. > > > > Jon > > The cutting tooth is what is being used for the electrical contact, with > some trickery in the .hal file to derail the "probe contact while jogging > logic" and steer the probe signal pin to the homing logic ONLY while > homing. Seems super sensible to me since its the tooth that is actually > doing the cutting., so why not measure it? :) That of course doesn't work > for Z when the threading tool is mounted, but the side of the toolholder > seems to work just as well. The same comment of course applies to the end > of the boring bar as the actual tooth is, for a conventionally shaped bar, > not on the end of it, but I often use an .125 milling bit, set in a drill > chuck mounted in a boring bar holder, with the toolpost turned 5 degrees > CCW as a boring bar for small holes, like the pocket for a #209 primer, > supposedly .240" diameter but very sloppily made, I've measured some that > were .007" out of round right out of the box! Apparently shotgun hulls > don't care. :(
Hi Gene Shotgun hulls are rather forgiving; thin brass(?) and low pressure, max is slightly higher than a .22. As long as the primer doesn't leak under pressure it is probably OK. The fun of this list is the diversity of thought and approach to problems. :-) Dave > > The cutting tooth of the boring bar is of course on the near side of the > tool, whereas the normal turning tools cutting edge is on the far end. My > gauge has contact surfaces arranged to sense both front and rear points. > > That of course needs a reversal of search directions for the X only. The > gauges contact for front edged tools is a swing down & lock in place arm, > with only smallish 3/8" square contacts, but the Z is offset about 1/4" to > the right, which needs a Z axis HOME_OFFSET set to compensate. > > The rest of the file sets Z left limits about 10 thou off the face of the > chuck jaws, and X limits about 50 thou short of the end of mechanical > travel in and out. > > Except for those details, the .ini files are (or should be) identical, and > it would be a huge advantage to actually use the same file but setup 2 > buttons HOMEN(ormal) and HOMEB(bore) which would when clicked, call the > correct homing sequence file for the tool mounted. That would also have > the huge advantage of not needing to stop and restart LCNC just to fine > tune an axis's 0.0000 position. X of course is the problem child here, and > I should plow a pocket in the left end of the gauge to make room for the > test turning piece so I don't have to remove it to re-home it to the gauge. > There is sufficient stock that I can plow a groove for an inch of stickout > of some 1/2" stock for test cuts so I don't have to remove, then remount > and recenter the test piece. > > Perhaps that can be todays gauge improvement, but I intend to get to the > range and test a new bullet seating tip for the ramrod today too. This > particular bullet, a swaged 300gr HPBT that looks like the ultimate deer > slayer, but a poorly fitted seater on the ramrod can do serious damage to > the hollow point geometry, which will turn one hole groups into a a > cylinder bored shotgun pattern. The bullet maker supplies a plastic rod > tip but its stem is easily broken off in the ramrod face and I have one > that needs some time on the cook stove to burn the remains out to where I > can finish cleaning it out with the properly sized tap, an 8-32 IIRC. I > need to find my round tuit at the same time the missus is gone for long > enough that the stink will dissipate, which likely would not be doing her > COPD any good at all. :( > > So that is the reasoning behind this need for a separate files containing > the homing instructions. And it seems to me that this could, rather than > scratching just my itch, be a worthwhile general improvement in how we > treat and calibrate a lathe, usable by everyone running a lathe with > LinuxCNC once they 'get the hang of it', because there would be lots less > time wasted making test cuts on scrap material. > > I just found two spare pins in the parport, the hal config assigns an A > axis I don't have, so pins 8 nd 9 are available, so I can now go ahead and > make the braking R switching control out of a wcomp module looking at > encoder-velocity. That could speed up the G33.1 stuff by short circuiting > 2/3rds of the 3 or 4 turns of overshoot I'm getting now as it virtually > coasts to a stop with a 16 ohm 40 watt load R. Below 400r's or so, 4 ohms > at 40 watts should be safe for the PM magnets I'd think. > > Thanks Jon, I'll get me coat now :) > > Cheers, Gene ------------------------------------------------------------------------------ This SF.net email is sponsored by Windows: Build for Windows Store. http://p.sf.net/sfu/windows-dev2dev _______________________________________________ Emc-users mailing list [email protected] https://lists.sourceforge.net/lists/listinfo/emc-users
