On Thursday 24 January 2013 11:05:42 Todd Zuercher did opine: Message additions Copyright Thursday 24 January 2013 by Gene Heskett
> Add a spring to your contact, to give it a degree of compression. > I did, in a previous design, mount the pcb on grommets so there was room for crush. but found it near impossible to get a consistent spring back. Or if off center, a consistent restoration to level. a 4 grommet mount would have solved that, but the frame of that gage didn't have room. So I wound up insulating the pcb being carved in a pocket of micarta, and g38.2'd the pcb I was carving, which worked very well indeed. Like T. Edisons & his light bulb, he found 10,000 ways it didn't work. All my current problems regarding this would go away if the switch hysteresis backup move would just do another 5 thou of back away anyway if it found the switch open when it checked instead of freezing in place and fussing with a big red error advisory because the switch was found to be already open as it started the back away move. HINT, HINT. IMO, the only error here is in the assumption that the home switch is mechanical, with several tens of thousandths on an inch, or in metric, possibly a whole mm of hysteresis. Perhaps even a per axis keyword to make it work either way? Would seem to give maximum flexibility at any rate. > ----- Original Message ----- > On Wednesday 23 January 2013 21:02:47 Stephen Dubovsky did opine: > Message additions Copyright Wednesday 23 January 2013 by Gene Heskett > > > Gene, > > Nowhere does it specify in the graphic that the two HAVE to be > > separate. The drawing is arbitrary/universal and just showing which > > edges are used on which steps. If you set the SEARCH_VEL reasonably > > high enough the machine must over travel the switch due to > > deceleration. Have you tried setting your search velocity high > > enough that the over travel is greater than your backlash > > compensation? > > > > Stephen > > No I haven't Stephan, because its backed up against the end of the > workpiece in the chuck in normal use, is a solid black of alu with a > sheet of double sided pcb materiel glued to the right face for the > contact. If I hit it that hard and fast I would be crushing the pcb > material. So my initial search is at about 2"/min, with a final search > vel of about .2/minute. Repeatability is someplace in the 4th digit to > the right of the decimal point even with 23 thou of backlash. So we > aren't talking about a switch at all, but a contact with hysteresis of > perhaps .0001" max. > > That backlash is going away, I now have a 16mmx5mmx675mm ball screw > laying on a chair cushion and am milling the rest of the parts to > install it as I sit here yakking. > > And wonder of wonders, we actually have enough snow to turn the ground > white! We are about 40" of rain or equ behind since last spring. Got about 3", DW staying home I guess. Had Dr's apt this morning. > > On Wed, Jan 23, 2013 at 4:11 PM, Gene Heskett <ghesk...@wdtv.com> wrote: > > > On Wednesday 23 January 2013 15:59:55 Bruce Klawiter did opine: > > > Message additions Copyright Wednesday 23 January 2013 by Gene > > > Heskett > > > > > > > I am curious, how does LinuxCNC back off of a limit switch? > > > > I have a bridgeport with hand wheels so it is easy for me, just > > > > wondering how this is done on the big machines. > > > > > > > > Bruce > > > > > > As shown in the gfx on the wiki under homing, there is an implicit > > > assumption that it is a mechanical switch, with a bit of hysteresis > > > space between the close and the open going the other way. It's > > > fairly well explained, but the lack of hysteresis in an electrical > > > contact such as I am using, means I cannot set my backlashes for > > > total compensations, because the backlash takeup move may reopen the > > > contact, and its a show stopper error if the switch is already open > > > when the backup move starts. I have x home working about 90% of the > > > time by setting backlash to half a thou in the .ini file for x on my > > > lathe, and around 3 thou less than real on my old sloppy z. When x > > > works, it does all the motions in just a few milliseconds once the > > > initial contact has been made. The old Z was the same except for > > > the 300 milliseconds the backlash moves of 20 thou take. > > > > ---------------------------------------------------------------------- > > -- ------ Master Visual Studio, SharePoint, SQL, ASP.NET, C# 2012, > > HTML5, CSS, MVC, Windows 8 Apps, JavaScript and much more. Keep your > > skills current with LearnDevNow - 3,200 step-by-step video tutorials > > by Microsoft MVPs and experts. ON SALE this month only -- learn more > > at: http://p.sf.net/sfu/learnnow-d2d > > _______________________________________________ > > Emc-users mailing list > > Emc-users@lists.sourceforge.net > > https://lists.sourceforge.net/lists/listinfo/emc-users > > Cheers, Gene 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) My web page: <http://coyoteden.dyndns-free.com:85/gene> is up! My views <http://www.armchairpatriot.com/What%20Has%20America%20Become.shtml> "Just think, with VLSI we can have 100 ENIACS on a chip!" -- Alan Perlis I was taught to respect my elders, but its getting harder and harder to find any... ------------------------------------------------------------------------------ Master Visual Studio, SharePoint, SQL, ASP.NET, C# 2012, HTML5, CSS, MVC, Windows 8 Apps, JavaScript and much more. Keep your skills current with LearnDevNow - 3,200 step-by-step video tutorials by Microsoft MVPs and experts. ON SALE this month only -- learn more at: http://p.sf.net/sfu/learnnow-d2d _______________________________________________ Emc-users mailing list Emc-users@lists.sourceforge.net https://lists.sourceforge.net/lists/listinfo/emc-users
