In the posts on CNCZONE to which I linked earlier the conclusion seems to be that the force to actuate the Tormach probe actually bends the "stiff" ceramic staff and causes significant measurement errors. Using a larger diameter carbide staff seems to work much better.
> -----Original Message----- > From: Kirk Wallace [mailto:kwall...@wallacecompany.com] > Sent: Sunday, December 13, 2015 12:48 PM > To: emc-users@lists.sourceforge.net > Subject: Re: [Emc-users] Probe > > On 12/13/2015 07:36 AM, Dave Caroline wrote: > > On a cmm the probe can be calibrated around a reference ball in > > multiple directions to get the effective diameter of the stylus ball, > > this should be done whenever the stylus is changed. > > > > This can also calibrate out the variation in different directions > > caused by the variation in contact force needed to break the contact, > > some probes have a contact force adjustment at the rear which forces > > the stylus downwards, while lightening the force reduces stylus > > bending it also makes the electrical contact less good so is a > > compromise. > > At first, we had trouble with the Tormach economy probe. The contact can > electrically bounce on the make or break motion. The slower the motion, the > more chance of bouncing. This is not a huge problem on the make motion, but > I found that retracting from the edge would sometimes cause enough bounce > to trigger a probe triggered on non probe motion error. > Rogge made a new probing g-code just for retracting which masks the trigger. I > use it just long enough to retract then use normal motion. > > Keeping the probe and retract speed up also helps. The accuracy doesn't suffer > until the distance traveled between servo cycles gets bigger than the desired > accuracy, or the encoder or step size, so slower probes often are not any > better than moderate probe speeds. > > Debouncing can prevent probe errors but the signal time delay hurts accuracy, > so as far as I know, we don't debounce. > > Centering is done with six screws that adjust the tilt between the housing and > the arbor. Three push and three pull so there should be no slop between the > housing and arbor. > http://www.tormach.com/uploads/474/TD10088_Passive_Probe_Install_0515 > A-pdf.html > > There is a set of probing routines in the Tormach UI that automate common > probing actions, including centering the probe tip. I have found that the > centering is quite stable if one doesn't leave any push or pull screws loose > when adjustment is done. > > Tormach (and others) can provide (practice) ceramic probe styli that are stiff > but break easily enough to protect the probe and work. > > > > > > More modern probes use a force sensor to get around the kinematic > > design problem. > > > > Dave Caroline > > > > ---------------------------------------------------------------------------- -- > > _______________________________________________ > > Emc-users mailing list > > Emc-users@lists.sourceforge.net > > https://lists.sourceforge.net/lists/listinfo/emc-users > > > > > > > -- > Kirk Wallace > http://www.wallacecompany.com/machine_shop/ > http://www.wallacecompany.com/E45/ > > ---------------------------------------------------------------------------- -- > _______________________________________________ > Emc-users mailing list > Emc-users@lists.sourceforge.net > https://lists.sourceforge.net/lists/listinfo/emc-users ------------------------------------------------------------------------------ _______________________________________________ Emc-users mailing list Emc-users@lists.sourceforge.net https://lists.sourceforge.net/lists/listinfo/emc-users
