Michael,
The Tony Griffith description suggests that the pitch correcting idea was superceded in the thirties. In the late sixties I was involved with using a trio of Pratt and Whitney Jig /Drill positioning tables as multi spindle drilling machines for pcb's They were driven by NC controls from 8 hole paper tape and the 40" x 20" table traverse was through recirculating ball screws. It was early days for this type of screw and P&W used exactly the same pitch correcting method to acheive a +/- .001" positional toleance over the whole table travel, each machine having its own calibrated cam track to turn the ball nut. Needless to say they didn't have additional proportional pitch variable feature.
Bob D

Michael,
There is little evidence now of the P&W machines. I have just one photo of the prototype as we were building it but it shows most of the features of the machine. It was designed to drill upto 4 holes simultaneously in each of two multilayer motherboards destined for 1960/70s mainframe computers
Holes on a .100” pitch were drilled for mounting the sockets to take 4 rows of plugin boards plus additional holes for intersect connections that were through hole plated on the buried pcb layers.

Referring to the photo the Tape-o-matic was fed an input off 8 hole teleprinter punched tape to define X and Y position data and also the sequence of the two banks of 4 high speed drills to deploy at each stop position. These were powered by the 8 solenoid valves on top of the arms feeding an air cylinder attached to each drill head. Heads typically run at 20,000 rpm for 0.042” carbide drills but had a max speed of 45,000rpm.
The frame design bore some resemblance to a railway buffers structure and seemed to be quite stable but in the end the machines proved to be distorting with the arms swinging side to side on a random basis, making the overall +/- 0.003” positional tolerance very difficult to achieve.
Eventually this was found to be caused by movement in the building. The machines were sited on the first floor of a steel framed building, one corner of which was on a clay bed that moved with the seasons and the weather. It applied sufficient twist to the floor plane to influence the drilling machine frame and caused the movement of the arms carrying the drill heads.
The solution was to mount the entire machines on pneumatic cushions that isolated them from the twisting floor. Each machine had three reference levelling pads to keep them stable and in normal incremental table movements this was fine, however they had also got a high speed jog mode with a supplementary AC drive and magnetic braking. This movement would cause the machines to rock to and fro in an alarming manner on the very flexible air cushions for a few moments but fortunately without detriment to the machine accuracy as it resumed drilling.
A further machine was taken from P&W that had a smaller table and came with a rear pillar to which we attached a custom designed head to carry the multiple drill spindles.
Bob D