You might try the FADAL mill method. They use two arrows (markers). One on the moving member and one on the unmoving member. You physically align the arrows and run the home routine. The next found index mark on the encoder is the home position.
regards Stuart On Sun, Dec 4, 2022 at 1:25 AM gene heskett <ghesk...@shentel.net> wrote: > On 12/3/22 22:56, John Dammeyer wrote: > >> > >> From what I understand you're talking about a normal parallel lathe so > here > >> are my thoughts: > > Yes. And thank you. > >> > >>> The Z axis is more interesting. Before homing I imagine the tail stock > >>> has to be loosened and moved all the way to the right as far away from > the > >>> headstock to ensure finding a home switch. Or a home switch could be > >>> somewhere in the middle but then which direction to search? > >>> > >> > >> I think the safest setup in this case (given your Z axis can crash with > >> the tailstock if you forget to move it all the way to the right) will be > >> having the home switch towards the chuck side (with a proper > >> independent limit switch right at the left of the home switch to avoid > >> crashes when homing). This way you can set up your homing sequence to > first > >> home the X axis to move it to a safe place and then home the Z axis > towards > >> the chuck. > > > > I'm guessing you are suggesting something like what Gene was that the > home switch isn't at the end of travel but somewhere else and once > activated never goes inactive in that same direction. So if you start a > home sequence and the switch is ON you know you have to go to the right > until it goes OFF. It can't over travel past the OFF position and go ON > again. That approach lets the Home Switch be pretty well anywhere. > > > > Unless the limit switch is movable it's useless. My carriage can go > much further to the left to turn near the part held in the 5C collet than > it can with the part held in the 3-Jaw. So a crash into the spinning 3-Jaw > can happen without ever touching the limit. And if the limit does prevent > that one can't turn a part close to the collet. > > > >> > >> How does one determine, with that tool tip, where the lathe centerline > is > >>> and set that so G54 X is 0.000? > >>> > >> > >> I think you're asking about tool setting. If you have tool fixtures that > >> ensure that whenever you change your tool you get the exact same tool > >> position then it's just like a CNC turret. You just take a skim cut on > the > >> diameter (or maybe use some fine paper to gauge the tool against the > >> workpiece) and then measure and input the diameter (or radius depending > if > >> you're in G7 or G8) in the touch off popup. > >> > > Thanks. I'm going to have to study that a bit further. > > I understand that the machine coordinate for x at the home switch is > 0.000. > > > > And measuring the tool tip relative to how it mounts into the AXA holder > can be done creating an offset from the home switch.v Now each time the > tool holder +tool is put into the AXA the distance from the tool table is > offset from that home position and LCNC knows where the tip and can turn a > diameter. Theoretically. > > > > However as soon as the holder is pivoted even slightly the distance of > the tool tip relative to the home switch changes. And it changes > differently depending on how far the tool protrudes out of the AXA holder > right? > > > > So the tool measurements are only useful with the AXA in one position > and that can change really easily in the home shop. We're not talking a > commercial production operation where the setup is the same over a long > period of time. > > > > I suppose if you can touch of or measure the new position of the tool > tip you can then use simple trig to determine the angle that the AXA holder > has pivoted. Unless it also slid in the T-Slot which now moves it in more > than one dimension. > > > > Doe my explanation make sense? A small shop lathe where the operator > can shift the tool bit around relative to the X home switch and even on the > carriage relative to the Z home switch suggest a tool table for a LCNC > lathe is virtually impossible. > > > > So I can manually turn that shaft. Leave the X axis where it is and > measure the diameter; say exactly 1.000". Then click on the "Touch Off" > on the display and enter half the diameter (0.5") for the offset. > > Now a request to move the X to 0.480" to do a 0.020" deep pass should > work right? > > > > But the tool table and for that matter the home switch and machine > coordinate system is pretty well useless. Or have I missed something? > > > > John > > > > > Don't adjust your antenna, you got it, John, the reception is perfect. > The main reason for a home sw is the let the machine know where its at > in the absolute sense, I've surveyed my bed for wear, and corrected it > with a lincurve/offset module combo. Rather satisfying to see the X > motor turn a bit as the Z is scanned. But, its effect is determined by > the absolute z position established by the home switch. It does not > respond to z touch off, only to the real position. > > Threading to any std, or to one you've invented is nothing but arguments > to the G76 routine. I've even abused G76 to make compression fittings > > The main advantage to cnc'ing a lathe is that its far faster at > executing the move codes than you can squint at a dial and do it by hand > in between cuts, it turns the crank in milliseconds and gets on with the > job. > > With no compound, mine was broken beyond redemption by a fall over no > one mentioned before I bought it, so LinuxCNC IS my compound, and far > more accurate than you can turn a compound and set it, fraction of a > degree several digits right of the decimal point with submicron errors > can be done by LinuxCNC. W/o a compound. > > It is all done with 2 stepper/servo motors and a raspberry pi4b, and w/o > a PID module anyplace. The driver does it all, including stopping > LinuxCNC in a millisecond if a motor can't get to where it was told to > go. I can set a stopped chuck jaw in the way, and program a .2" a second > move into the chuck jaw, it touches the jaw and springs back about 10 > thou as the drivers shut down, without marking the jaw, or damaging the > carbide chip in the tool. Tested many times doing that, but it has yet > to actually shut down a job. > > And if you want to step out of the way of burning swarf, the programs > linear speed can be turned way up, w/o worrying about getting holes > burnt in your clothes by white hot swarf. Carbide tooling is right at > home with those temps, human skin isn't. Coolant would be nice, but this > 80 year old Sheldon has no way to collect the coolant to reclaim it. > > 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, 1940) > If we desire respect for the law, we must first make the law respectable. > - Louis D. Brandeis > Genes Web page <http://geneslinuxbox.net:6309/> > > > > _______________________________________________ > Emc-users mailing list > Emc-users@lists.sourceforge.net > https://lists.sourceforge.net/lists/listinfo/emc-users > -- Addressee is the intended audience. If you are not the addressee then my consent is not given for you to read this email furthermore it is my wish you would close this without saving or reading, and cease and desist from saving or opening my private correspondence. Thank you for honoring my wish. _______________________________________________ Emc-users mailing list Emc-users@lists.sourceforge.net https://lists.sourceforge.net/lists/listinfo/emc-users
