This is a very interesting approach, because it involves no commer-
cially available pneumatic switches, e.g., those that were used in
magnetic tape drives with vacuum loops (in the 1970s).  Those switches
can be found in electronic catalogues for $20 each.

Could you tell us more about your design?

1.  How were the cylinders built into the pneumatic action?
2.  Were they teed in or serially integrated?
3.  Was the piston spring loaded or did it return to its
    rest position by its own weight?
4.  Which is the orientation of the sensor during use?

Perhaps you can find a rough sketch that illustrates the device in
more detail.

Regards

Walter Tenten
W.Tenten@FZ-Juelich.de