How do digital calipers work?

[Bob Mueller]

Seeing a request for an interface to these reminded me of my inability to
understand how the digital calipers measure position. What kind of sensor
is able to give this accuracy while running a long time on the tiny battery?
Thank you to anybody who knows and takes the time to spread the word!

Sincerely,
Bob<iff161@djukfa11>

[Paul Kennedy]

I seem to remember that one method of implementing digital calipers used
two overlapping precision-printed gratings. I think the grating lines were
at a slight angle to each other and so produced Moire (sp?) fringes as the
two gratings where slid past each other. Hook this up to a couple of
photocells, some schmidt triggers and some counting logic to drive the
display.

+-------------------------------------------------------------------------+
| Dr Paul Kennedy - Paul.z.h.k....@zhflur.ubs.ubs.ch ? |
| Union Bank of Switzerland, ~~~ |
| Bahnhofstrasse 45, 8021 Zurich, Switzerland ( . . ) |
+------------------------------------------------------------ooO-(_)-Ooo--+

[Robin Kenny]

Another method is to fabricate metal plates on the PCB along the length of
the unmoving caliper part and the moving caliper piece so the
moving and stationary boards form air spaced capacitors.
Shifting the vernier piece changes the amount of overlap of the plates and
hence the capacitance. It is now possible to wire a PAIR of capacitors into
a Wheatstone bridge for detecting the voltage difference sum.
Small linear movement will produce different voltage sums - analog position
output!
When the two capacitors in a pair are exactly equal, the sum of an a.c. voltage
passed equally through the caps equals zero; anything else is some
repeatable voltage sum.

You'll likely need many small capacitors along the length of caliper travel
to keep the small size of the instrument and probably because the accuracy will
not increase linearly with linearly increased capacitor plate size.
Obviously this involves several Wheatstone like arrangements and some logic,
but today that's trivial to fabricate on ICs.
Also; with several carefully placed capacitor null points along the length of
travel it may be possible to directly read out the gross position using vernier
principles.

All this is from electronics done ten years ago (*GASP*) so the details may
be off in places but the principles should be right.
Note that your electronics reduces to a.c. voltage measurement, A-D
conversion and driving an LCD/LED display... i.e. no large power drains.
For better theory read up on Linear Differential Voltage Transformers, LDVT,
an unbelievably simple way for precise linear motion measurement.

Robin Kenny - rob...@hparc0.aus.hp.com "New ideas in status quo"
(everything in this message is PERSONAL OPINION ONLY and has no connection
with my work or my employer, the Hewlett-Packard Company Australia)
--------------------------------------------------------------------------

Paul Kennedy (zh...@zh014.ubs.ubs.ch) wrote:

[John Lundgren]

I've got a dial calipers (helios), that uses a few gears in the dial to
make the pointer move enough to be seen easily. It doesn't seem too
difficult to me to substitute a disk for the pointer and put codes
around it, and use some method of sensing the codes. If the idea is to
eliminate the mechanical parts to save money, and substitute electrical
sensing, then I can see where it gets a bit more interesting, and one
would have to be more innovative. But whatever method, it would have to
be resistant to metal shavings, oil, dust and dirt, and rough handling
and wear to be successful.

The cheap little 9-pin printers seem to do a fairly good job of
determining where the print head is accurately along a linear length of
about 8 inches, and the 24 pin's do it even better. And it's a pretty
dirty environment, too.

--

[Jarek Lis]

John Lundgren (jlun...@news.kn.PacBell.COM) wrote:
: I've got a dial calipers (helios), that uses a few gears in the dial to
....
: The cheap little 9-pin printers seem to do a fairly good job of

: determining where the print head is accurately along a linear length of
: about 8 inches, and the 24 pin's do it even better. And it's a pretty
: dirty environment, too.

Most of them now utilize stepper motor and single photosensor to calibrate
column 0. There is no actual measuring of position.

But HP Deskjest 510 has very suspicious element (belt from foil with
stripes, located behind the carriage) which probably measures cartridge
position. Up to 300 dpi < 0.1 mm ?

: --