[time-nuts] HP 115 BR Clock 60 Hz output option: how HP did it...

[John Vendely]

Upon a closer look at its innards, I realized the method HP used for the 
115BR Clock 60 Hz output option is simpler than I initially 
thought--essentially an opto-mechanical frequency divider.   Here's a 
brief description for those interested.

The standard 115BR uses three "aperture discs", an exciter lamp, and 
photocell to generate an optically triggered 1 PPS output. The clock 
motor shaft, running at 1000 RPM, drives a 300 RPM gear with 12 
apertures which in turn drives a 60 RPM gear with 1 aperture.  The third 
disc has three apertures and is driven directly by the motor shaft at 
1000 RPM.  The apertures of all three discs are in alignment once per 
second, transmitting a light pulse to the photocell, which gates through 
1 pulse of the 1 kc regenerative divider output to produce an accurate, 
stable 1PPS.

The 60 cps aux. output option simply uses a second photocell mounted on 
the side of the 300 RPM gear opposite that of the first photocell.  The 
second photocell is thus excited by optical pulses from the 300 RPM gear 
and since this gear's angular frequency is 5 cycles per second, its 12 
apertures directly produce 60 cps pulses at the second photocell 
output.  Photocell output is applied to a transistor switch which drives 
a lowpass filter and output transformer, providing a fairly sinusoidal 
250 mV RMS 60 Hz output.  Since the 60 Hz output is motor-derived, its 
phase angle is adjustable by the front-panel induction resolver setting. 
Since it's mechanically generated, the output is slightly jittery, but 
its frequency does average out to 60 Hz and is adequate for driving 
synchronous motor clocks, which will themselves provide a mechanical 
averaging effect.

The parts list implies the requisite aperture gears are standard; if so, 
it wouldn't be too difficult to install a 60 Hz aux. output on a 
standard 115BR by adding the second photocell, a comparator, active low 
pass filter, and an output transformer, if desired. Alternatively, for a 
60 Hz output independent of the resolver setting, or if lower jitter is 
desired, output from the first tripler of the 1 kc regenerative divider 
could be buffered, divided by 50, and shaped as desired.

p.s., the only thing wrong with mine was a bad coupling capacitor...

73,

John K9WT