[time-nuts] 5 MHZ PIC PPS Divider?

Bruce Griffiths bruce.griffiths at xtra.co.nz
Wed Apr 16 20:35:52 EDT 2008

SAIDJACK at aol.com wrote:
> Hi Bruce,
> did you or anyone every build this LTC application?
> I wonder about the tempco matching of the reference 1N457 diodes  
> compensating the 2N5771 etc...
> bye,
> Said

Not this particular implementation but have built similar circuits 
decades ago.
Tempco isnt an issue.
There's plenty of time to interleave calibration cycles of say 100ns, 
200ns, 300ns, 400ns duration between PPS measurements.
It should even be possible to correct residual low order non linearity.

With some elaboration is easy to construct a BJT current source with a 
residual tempco of around 100 ppm or so due to the BJT's hfe tempco.
Replacing the polystyrene cap with a C0G/NP0 cap is probably worthwhile 
especially if one wishes to avoid hand soldering the polystyrene cap 
using heat sinks on the cap leads.

This technique and variants thereof have been in use for decades in 
nuclear instrumentation.
Another widely used technique is the Wilkinson converter wher the 
capacitor is charges by a large current during the time interval to be 
measured and the discharge time with a current perhps 1/1000 of this is 
measured using a comparator and a counter clocked at 10 or even a 100 MHz.


More information about the time-nuts mailing list