Don, When you say you see about 3 nSec standard deviation, where is this number coming from? As reported by a program like GPScon or LH, or an actual measurement of the 1pps or 10 MHz output against a lab standard? /tvb