[time-nuts] I'm all of a dither.
bruce.griffiths at xtra.co.nz
Mon Sep 21 21:46:22 UTC 2009
Joe McElvenney wrote:
> A fellow amateur and I have just put together the GPS-based frequency
> reference published in the "Every day Practical Electronics" issue of
> April/May 2009 and are both experiencing the same problem. The type-3
> phase comparator of the 74HC4046 has both 1-pps and 50kHz DC-coupled
> inputs designed to produce an error pulse due to the difference between
> their leading edges. However, the longer pulse dominates the output
> because it holds the latch in thrall despite the 50kHz pulse trying to
> grab a little of the action.
> My question is then, are there variants of the 74HC4046 that have a
> type-3 detector which is edge-drive only? Natch, I can AC-couple and
> clamp them but I just wondered. I know that in the past that there have
> been differing versions of the old CD4046 and now, of course, there are
> also the '7046 and '9046 types around.
> Cheers - Joe G3LLV
Found the 3 relevant issues and as I suspected without the correct
divider connections the 50kHz output may misbehave due to glitches on
the 74HC160 TC output.
The correct circuit actually eliminates the inverter between the
74HC160s by cascading them synchronously.
More information about the time-nuts