[time-nuts] Atomic Clocks: It is important that they keep good time, Part 1

[ed breya]

Paul, what do you mean by the display "looking pretty ratty?" As I 
recall, the original buck regulator had regulated output voltage around 
5V for the LEDs. and the PMOS clock IC needed something around 12V. 
Whatever the LEDs run from, it should be regulated and well filtered. If 
the LEDs are dim, it could be the old displays themselves are 
deteriorated, or the regulation isn't right, or maybe a bad output 
filter cap on the buck converter. If the brightness is OK, but digit or 
segment intensities fluctuate with count, then it's probably a 
regulation issue.

I used the shunt regulator to isolate the rest of the system from the 
large variation (about 3:1) in total LED current with readout values, 
and it was possible because I had made lower supplies anyway, via DC-DC 
converters. You wouldn't want to linear-regulate all the way down from 
the main supply around 24V, to a few V for the LEDs. As I recall, the 
peak load is in the 200-300 mA range at good brightness. The efficiency 
of the buck converter makes it practical to run this from the normal 
supply or battery voltage. I think the original deal was that on power 
failure it switched to battery mode, the buck converter was shut off to 
shed the LED load, and the clock IC stayed powered up to keep the right 
time. The button below the display could force it to show when needed. 
Mine will work the same way, when/if I ever finish all the details, but 
will have adjustable brightness, and maybe the option of still 
indicating time in backup mode, with very dim LED setting.

Ed