[time-nuts] PPS delay on rockwell

[Francesco Messineo]

On Sun, Feb 15, 2015 at 10:34 PM, Hal Murray <hmurray at megapathdsl.net> wrote:
>
> francesco.messineo at gmail.com said:
>> I measured the PPS output with an HP-5328B, PPS of the thunderbolt as start,
>> rising edge, rockwell PPS as stop, rising edge. The delay is 406.3 ms +/- 30
>> uS.
>
>> Are those numbers normal for a rockwell chipset? Any idea why the two PPS
>> are so far apart? PPS offset on the thunderbolt is set to 0.
>
> Do you have a scope?  What's the width of the Rockwell PPS?

yes, I have looked at the PPS shape, it's a 20ms positive pulse

>
> Are you triggering on the correct edge?  Is there an inverter in there
> someplace?
>
> I would try all 4 combinations of rising/falling edges and see if any of them
> match cleanly.  That could happen if the Rockwell PPS is 500 ms wide.

I tried all 4 combinations (besides, the thunderbolt is set for
positive PPS), the positive thunderbolt to positive rockwell is the
closest of all at 406 ms.
I can only suspect it was unlocked, but I need to setup all the test
in another place closer to the window, since I don't have a splitter
to use the same antenna of the thunderbolt.

Thanks
Frank