Thanks to Tom, Jean-Louis and Chris for all of the helpful advice. Problem is now fixed.
First problem was obvious - the 74HC390 I use as a divider clocks on negative-going transition. Added inverter, and the problem appeared to still be present (!) Next problem was not so obvious - my very cheap benchtop chinese frequency counter (a Zhaoxin HC-2700L, bought on ebay, and retrofitted with a divided-by-two Pletronics 26 MHz VCOCXO as timebase, which were for sale cheaply on ebay a few years ago) turns out to have a bug/defect in the "totalizer" mode; when count transitions from 99 999 999 to 00 000 000, the counter ignores its input for a few milliseconds while it processes something, so my original plan to count for 10 seconds (expecting to see 30 000 000, which is 130 million with the hundred-million digit wrapped off) didn't pan out. Small change to the counter circuit and I now count a series of 1-second intervals, each of 13 MHz reference oscillator in the counter. I adjusted the counter's timebase VCOCXO bias input so that counting for seven 1-second intervals shows 91 000 000 (13 000 000 * 7). There seems to be a +/- 1 count uncertainty, likely due to absence of synchronization between the counter's timebase and the 1 PPS signal, but this accomplishes what I set out to do, which was a cheap-and-dirty (but certainly not quick!) calibration of the counter's timebase. Thanks for everyone's advice and support! Dave >To: jl.on...@free.fr, Discussion of precise time and frequency > measurement <time-nuts@febo.com> >Subject: Re: [time-nuts] Beginner question - unexpected possible > jitter in 1 PPS output of Motorola ONCORE UT+ module >Message-ID: > <CABbxVHvpyUYZfP7=zztTKRXGr+F=k7DBYCzNLwAG04Tr=3_...@mail.gmail.com> >Content-Type: text/plain; charset=UTF-8 > >I posted here some days ago an admission that I did exactly this by >accident once. It is a very hard problem to debug because everything works >just fine, except for the very high jitter. So I go looking for noise and >what not. The problem was a wrong number of inverters. > > >On Sat, May 31, 2014 at 7:58 PM, Jean-Louis Oneto <jl.on...@free.fr> wrote: > >> ..... So if your divider is clocked on the falling edge of the 1PPS, you >> can expect a 1 ms jitter... > >-- > >Chris Albertson >Redondo Beach, California _______________________________________________ time-nuts mailing list -- time-nuts@febo.com To unsubscribe, go to https://www.febo.com/cgi-bin/mailman/listinfo/time-nuts and follow the instructions there.
