On 1/24/2012 4:52 PM, John Prentice wrote: > ----- Original Message ----- > From: "Kent A. Reed"<knbr...@erols.com> > > >> Several items were called out recently as being show stoppers for >> LinuxCNC. I do not aspire to learn the inner workings of LinuxCNC well >> enough to contribute to discussion of the first item, "No jog on >> feedhold". >> >> However, the second item "Taper thread pitches are measured along the >> hypotenuse ???" is an issue I think this bear of limited brain ought to >> be able to understand without being a LinuxCNC guru. >> > (a) The jog in feedhold really is a significant pain. You do need to move > the tool away from the work when milling or turning "stringy" materials. > Even something simple like deep-drilling with pecks can snarl up the tool > with swarf. Of course the general solution could be very complex (e.g. if > offsets were to be changed while feedheld) but some simple rules would cover > a lot of cases without obvious risks to a thinking user. Implementing this should actually be pretty easy if you implement a jog at the hal level. In the case of a servo machine, just add an offset to the target position of the PID control. Of course, you would probably want to have some sort of switch to enable it and cause the "normal" jog signals to be used here.
You would also want to save the current position so you could go back to where you started. Most of this is probably doable without writing a line of C code. Of course, if you want to change tools and use a tool with different offsets, that would be a different problem. If I were implementing this, I would probably build a new component. The new component would "memorize" the jogging that was done so that hitting a single button could reverse the motion and have to tool return to the original location following the same path in reverse. I can understand that not everyone is a C programmer. Many EMC^H ^H ^H LinuxCNC users are integrators, though, and should be able to build something like this using hal. As far as the threading is concerned, some people just seem to want to complain. Alternatives include: 1 -- Do a little trig and convert the numbers. 2 -- Modify the source to (use the same trig) do it the way you like. Be generous and add a configuration flag that lets the code do things either way. 3 -- Find a friend to do #2. 4 -- Hire someone to do #2. (Don't even ask -- I have two prices for my work on LinuxCNC, the first is free and you probably can't afford the second.) Regards, Ken > > (b) For most practical tapered pipe threads no one will notice the pitch > error. On one hand I think it is unusual CNC behaviour in threading (so > possible difficulties for CAM users without a special postprocessor). But on > the other hand the case of the angle not being small but being 90deg does > appear to allow cutting of scrolls - has anyone ever tried this? > > John Prentice > > . > > > ------------------------------------------------------------------------------ > Keep Your Developer Skills Current with LearnDevNow! > The most comprehensive online learning library for Microsoft developers > is just $99.99! Visual Studio, SharePoint, SQL - plus HTML5, CSS3, MVC3, > Metro Style Apps, more. Free future releases when you subscribe now! > http://p.sf.net/sfu/learndevnow-d2d > _______________________________________________ > Emc-users mailing list > Emc-users@lists.sourceforge.net > https://lists.sourceforge.net/lists/listinfo/emc-users ------------------------------------------------------------------------------ Keep Your Developer Skills Current with LearnDevNow! The most comprehensive online learning library for Microsoft developers is just $99.99! Visual Studio, SharePoint, SQL - plus HTML5, CSS3, MVC3, Metro Style Apps, more. Free future releases when you subscribe now! http://p.sf.net/sfu/learndevnow-d2d _______________________________________________ Emc-users mailing list Emc-users@lists.sourceforge.net https://lists.sourceforge.net/lists/listinfo/emc-users
