PPS over USB

[Dan Drown]

Quoting Hal Murray <hmurray at megapathdsl.net>:

> esr at thyrsus.com said:
>
>>> fudge 127.127.22.1 refid PPSU time1 0.100700
>
>> *Trailing* edge? That's disturbing. I thought previous observation had shown
>> that the Adafruit hat reports rising edge.  If this so, whether we catch the
>> rising or falling edge is variable based on more things than just the
>> hardware type and life gets much more complcated.
>
> The Adafruit HAT does the right thing.
>
> We need to know the details of Dan's hardware setup.  I think the context of
> his fudge offset was with a non-HAT device.
>
> You can flip the rising/falling aspect of the leading edge with an inverter.
> The standard RS-232 level shifters include an inverter.  If that's what's
> going on, I would have fudged flag2 to use the falling edge rather than
> fudging the time offset.

Yes, I had the wrong edge setup on my USB PPS configuration.

This is my new configuration:

fudge 127.127.22.1 refid PPSU time1 0.000550 flag2 1

I put a picture of the hardware, the timeline for the changes along  
with the impact of those changes here:  
https://dan.drown.org/rpi/usb-pps/

Quoting Gary E. Miller:
> Or easier yet, you can change the pps_gpio parameters at boot to
> look at the other edge.

This specific PPS is the USB UART (pps_ldisc), so it picks up both edges.