Orientation was Re: Repeatability of Home switches

My 2p (or 3c american. ):

Why not look into using some good old-fashioned IR LED's and IR-
dependant resistors? If you're skint you can find them in an old
printer (HP printers have these nice little square units with both
contained inside, and a slit between them - between which a thin beam
slides to obscure the infra-red beam).

There's got to be a billion ways to integrate this into the table
too - either tapping a hole direct (ahh! heathen), or tapping a hole
in the side to attach a little armeture to obscure the beam. You can
leave some "slack" at the end of travel, so even if your table
overshoots (racing really fast) it won't crunch into the end of
travel.

Also, since no mechanical contact is made, it'll never wear out and
there's no danger of it being crushed to death by your machine. Also,
the accuracy is probably going to be quite good if you set it up with
only a thin "peep-hole" of a beam shining through. IR
detectors/emmitters are less $1/piece nowadays.

You probably don't need any driver circuitry either, since the way-
low milliamp requirements needed for sensing on the digital input
end - you may actually need a resistor on the sensing end to drop
down the level a bit (it may read 1 all the time).

My 2p.

--- In CAD_CAM_EDM_DRO@yahoogroups.com, CL <datac@l...> wrote:
> mrgamber wrote:
>
> >What is the recommended orientation of the switch?
> >
> >Should it be in the path of the axis and depress the lever or
button?
> >or is it better to let it "wipe" the trigger? Wiping seems safer
but
> >less accurate.
> >
> >
> The setup should always be designed in the event that a machine has
no
> idea where it really is decides to go anywhere near a home switch,
that
> it does not smash the switch to bits. I would tend to agree that
hitting
> a mechanical swtich "head on" might be more accurate, but who
knows ?
> The first really important question is what repeatability do you
need at
> the switch ? As Doug pointed out just earlier, Does ones machine
> resolution drop way lower than your switch can detect ?
>
> I'd sure hate to think that some are overly concerned about home
switch
> locations, settings and resolutions on any machine that does not
have
> the "other mechanical 'Guts'" to make a fine switch matter.
Everything
> has to extremely "tight" in this regard to even get to .001". That
> might rule out homemade antibacklash stuff that needs occasional
> adjustments.
>
> Back to your question though, My favorite technique is somewhat
twofold.
> I'll hit a switch "headon", but then back the switch up with a
small
> hydraulic dampener to absorb the impact that will occur if the
machine
> is "lost". The dampener protects the switch.
>
> The only other safe bet is to allow the switch to wipe or swipe,
and
> have a goodly amount of space for the "coasting" before a fixed
> "deadstop". I will use "swipe" techniques on a Z axis before an X-
Y, as
> it is usually a spot that is "set" for jobs anyhow.
>
> I'd be confident that a swipe technique is fine for most
applications,
> as long as the machine can transverse the contact point
effectively.
> That can be sometimes tricky. But, it depends on what your making,
how
> your making it, and if you need to be concerned about it based on
your
> needs.
>
> Chris L