[time-nuts] 1pps signal presence with no GPS signal.
phk at phk.freebsd.dk
Sat Jul 9 06:25:56 EDT 2005
In message <42CFA26E.4060007 at onetel.net>, David Kirkby writes:
>However, having thought about it more carefully, 3 of the 4 LEDs could
>probably be driven from the 9-pin RS-232 port on the box, without the
>need for an optical detector. The 1pps appears on pin 1, the TXD on pin
>2 and the RXD on pin 3, with pin 5 common. I assume just using those
>lines to switch a transistor would work fine - it might even be possible
>to drive a low current LED directly with a resistor.
Let me just insert a caution here: RS-232 ports are heavily EMI protected
and therefore the flanks of the signal are pretty slow already. If you
add further load to the signals you will make the slopes even worse
and therefore get even more jitter on your PPS signal.
In the first GPS I got from Synergy there was a capacitor from the PPS
signal to GND that reduced the slew-rate to less than 1V/us, but I
hope I managed to persuade them to remove it in subsequent versions.
Poul-Henning Kamp | UNIX since Zilog Zeus 3.20
phk at FreeBSD.ORG | TCP/IP since RFC 956
FreeBSD committer | BSD since 4.3-tahoe
Never attribute to malice what can adequately be explained by incompetence.
More information about the time-nuts