Re: which commercial linear scale for EMC & S2G card?





Ray Henry wrote:

> Doug
>
> Perhaps some of the confusion comes from the difference between absolute
> and incremental encoders or scales.
>
> An absolute encoder or scale spits out enough information to locate its
> exact position anywhere while an incremental encoder or scale only spits
> out a change in position.

Theoretically true, but now many manufacturers of linear encoders are
advertising
them as "absolute".  What they have done is one of two things.  The first one
is quite old, it is to put many "index" or "home" marks on the scale, but have
a different number of incremental ticks between any two.  With ticks approx.
1" apart, that is real easy to do.  All you have to do is move a little more
than 1",
and the readout will see two of them, and by reading the distance between them,
it can calculate where one the scale it is.

The newer technique uses coded index marks, which have some bit pattern that
the readout can decipher.  These can be every .1" or so, and the readout only
needs
to see one of them to know the absolute position.

With either of these techniques, the axis must be moved, at least a small
amount, for
the readout to read the index mark(s).

So, they are not true absolute encoders, which can report absolute position
without
being moved at all.

If you figure out how many bits you need for a true absolute encoder of any
size,
it is clear that it is impractical for linear measurement.  You'd need dozens of

bits!

> I know that STG cards work with quadrature type incremental signals
> regardless of the specific device (scale or rotary) but I don't know if
> those boards can be made to work with absolute signals.  I suspect not.

Since there are no TRUE absolute linear scales, that's no problem.
But, one of the above mentioned pseudo-absolute scales could cause
a problem.  If the first type was qualified with a clost-to-home switch, then
it should work fine.  The second type would not work, because the index
consists of a series of closely spaced pulses, too close for a close-to-home
microswitch to qualify exactly one of those pulses.

Jon




Date Index | Thread Index | Back to archive index | Back to Mailing List Page

Problems or questions? Contact