Re: [CAD_CAM_EDM_DRO] Re: low cost linear encoders

----- Original Message -----
From: Jon Elson <jmelson@...>
> Yes, it is an electronic capacitive vernier. I haven't had a Mitutoyo
> apart, but have had several
> Chinese imports apart for cleaning. The Chinese scale has little
> squares about 2 mm square of copper
> on a PC board material, with the squares ocurring about every 2 cm. The
> head has a bunch of thin lines
> all connected to the control chip. Presumably by exciting some lines
> and sensing others, it can tell where
> the square is in relation to the lines. I haven't figured out the
> entire scheme, there must be some
> interpolation of signal strength, because the strips are only about 3 or
> 4 per mm, at the most.
>
> Well, making these things with such methods would likely cause a pretty
> inaccurate scale.
> You'd be lucky if you got better than .25 mm accuracy even over a short
> distance.

Thanks to everyone for your observations.

I haven't seen one of the chinese ones but this Mitutoyo definitely has the
copper pads cut right across in a diagonal fashion producing a series of
parallelograms such that the bottom point of one coincides with the top
point of the next (I'll try a bit of ASII art)
............ ............ ............
/ / / / / /
/ / / / / /
------------ ------------ ------------

I think perhaps the angle of the cuts is a little more but I couldn't think
how to show it - anyway, something coupling to the top of the pads would
obviously have a phase shift in respect to something coupling to the bottom
of them. Presumably this is what allows it to read to a fine resolution. The
thing I haven't quite figured out is how the thing works without any
electrical connection to these pads and, if two or more sensors are
involved, how they work without interacting - if one sensor feeds a signal /
voltage? into the pad, surely another sensor over the same pad would pick it
up so can't see how the phase shift would work. Perhaps its strobed onto
different sensors.

As to making accurate strip I don't think that would be much of a problem.
The way I would tackle it would be to make a jig which would have to be very
accurate but which could be made in separate parts that could be accurately
lapped to size. This would have a baseplate onto which would be fitted a
bridge section of 3 parts, the central of these being the most important and
requiring the greatest accuracy. There would also be two more items required
which I would make, one being a detent - just a flat piece of steel of an
accurately known thickness and with one edge formed as a dead central knife
edge. The final part would be a file, exactly like the detent but with file
teeth cut along the knife edge. I would cut these teeth with an 'etching
bar' as is used to make needle files but they could be made with an ordinary
file. The central part of the bridge would have to be machined and lapped to
give two exactly parallel diagonal edges the precise distance apart to set
the detent and file to the exact width of the required pads and at the right
angle. The two outer parts of the bridge would be then set to keep the
detent and file pressed against the central part. I would use a bridge
design as this would guide the PCB strip and would allow me to equip the
file with a depth stop, so ensuring that all the gaps between the pads were
the same width. Then it would simply be a matter of filing one notch into
the PCB, feeding it along until the detent dropped fully into the notch and
repeating until I had a long enough strip. This is just an adaptation of a
very old technique but one which is of proven accuracy if the tools are
carefully made. The fact that the strips used on the Mitutoyo verniers have
pads in excess of 3/8" long would make this quite a practical proposition.

Ian

--
Ian W. Wright
Sheffield UK