First off, a "real" microswitch costs only about $0.50 if you buy them on eBay 10 at a time and on Amazon Prime, I paid about double that. What you want is repeatability and microswitches are good at that to better than 0.001 inch. Test your push buttons to see if they always click on/off at the same amount of depression. They might be really good or really poor.
Next, you need to decide how to handle contact bounce. The switch will open and close perhaps a dozen times when ever it changes state. If the goal is to stop quickly you don't want the software to see a bouncing switch. Or for the first few milliseconds, the software will see "home" and stoart slowing only to see "not home" and continue searching and flip about a dozen times before seeing a solid signal. SOme switches are MUCH better and some are much worse. I just tested a bunch of mine. I got the best performance from a small micro switch with a resistor and very small capacitor soldered across the terminals. It changed state in a few nanoseconds, others were as bad as 20 milliseconds. Yes, a factor of about 100,000 different. How to know if you are skipping steps with an open loop stepper? Measure where to table stops AFTER it stops. If steps were skipped it will overshoot the mark. But you need to find a way to measure one step of travel. Maybe clamp a block to the table and measure the distance to spindle but with spindle just above the block? Make it stop quicker un till it overshoots then divide the speed by maybe two. Do the testing with a typical weight bolted to the table to simulate the weight of a part. Also, if you are trying to stop WAY to fast, you will hear the motor skipping steps. But don't depend on that because I doubt you could hear just one missed step. How to mount the switch. A real micro switch can have a roller on a lever. Mount the switch so that a cam or "bump" on the table triggers the switch the roller reduces friction as the cam passes. If overshoot no longer matters you can mount the switch at the center of the table, not on the end. One more idea uses two switches. Switch one trips about an inch before the real home switch and simply tells the software to "slow down". The other way to just to ram the home switch at full speed. The machine stops, likely after overrunning the home position. But then it backs up about 1/4 inch and approaches the switch very slowly. I have seen systems where it repeats this backoff and then go slow perhaps three times and then takes an average of all three positions. Or if one of the three positions is not a good match to the other two raise an error. But in all cases mount the switch to the side so that a software error does not destroy the switch. Then if you like, build a "'safety stop" switch that the software never sees. It simply disables the motor if tripped. (the "step" pulse has to go through an NC switch.) This could be mounted on a spring or block of foam. Even 1mm accuracy is overkill for this safety-stop. On Sun, Mar 31, 2019 at 7:02 AM Gene Heskett <ghesk...@shentel.net> wrote: > Greetings everybody; > > I think I've got the coolant pump starting problem fixed. Ignore that > faint knocking sound. > > Now I would like to use a teeny little pushbutton (6x6x2.5mm tall) > between two solid parts of this machine for home and potentially as > limit switches. > > However the amount of available overtravel after the switch has clicked > is quite limited unless this switch is mounted on something crushable so > that the getting stopped overtravel does not crush the switch like a > Coors can. > > Is there a way to determine how much overtravel vs approach speeds is > occuring? > > I ask because a wide open x or y move can do around 220 ipm on this > machine, and that stopping distance is not an ignoreable distance when > the switch only has maybe .010" of overtravel after its clicked. > > So I first would like to determine the maximum safe SEARCH_VEL I can use > for homeing, then from that, be able to set MAX and MIN LIMITS far > enough away from the crash stop to provide crash protection in the space > between the LIMIT set in the ini file, or how much crush room I have to > build into the switch mount? > > Also, in attempting to minimize this stopping distance, what or how can > one detect a motor step slip if the ACCEL's are stopping the motor > faster than it can stop? Hopefully without pounding a $90 dial into > junk. > > Thanks all; > > 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) > Genes Web page <http://geneslinuxbox.net:6309/gene> > > > > _______________________________________________ > Emc-users mailing list > Emc-users@lists.sourceforge.net > https://lists.sourceforge.net/lists/listinfo/emc-users > -- Chris Albertson Redondo Beach, California _______________________________________________ Emc-users mailing list Emc-users@lists.sourceforge.net https://lists.sourceforge.net/lists/listinfo/emc-users
