It's not ideal, but it might work out fine. You can think of the PPS signal
as a marker that declares which 10 MHz flank is the "first" one per second.
So the timing accuracy does not come from the PPS directly, but from the
combination of PPS and 10 MHz.
Quick back-of-the-envelope: Let's say your velocity factor of your cable is
2/3, and I'll approximate 1ft as 1/3m. Then your additional delay is
approx. (1/3m)/(2/3 * 3 * 10^8 m/s) = 1.7 ns (probably better, I think 2/3
was lowballing it for good RF cables but the math is easy). Your 10 MHz
flanks are 100 ns apart. So your PPS is shifted relative to your 10 MHz
flank by about 1.5%... I would guess it works.
You could also have meant that the 2 cables going to the first USRP are
1ft, and the 2 cables going to the second USRP are 2ft. In that case, the
second USRP will be delayed by 1.7 ns relative to the first, but as long as
temperature stays constant, the relative delay will be constant. Here, your
signal bandwidth also matters. At 2 Msps sampling rate, such a delay
corresponds to approx. 3 samples. But if you have a low bandwidth, then the
constant delay will look like a (constant) phase offset, which you will
have anyway and need to account for.
--M
On Sat, May 17, 2025 at 1:13 PM Steve Hamn stevehamn311@gmail.com wrote:
Hello,
I have a pair of x440's I'm trying to phase cohere together. I wanted to
understand how critical it is to have equal length 1PPS and Ref In cables?
Right now I have a 2 ft RF cable going to one and a 1ft RF cable to
another. Will that be a problem?
Thanks
Steve
USRP-users mailing list -- usrp-users@lists.ettus.com
To unsubscribe send an email to usrp-users-leave@lists.ettus.com
It's not ideal, but it might work out fine. You can think of the PPS signal
as a marker that declares which 10 MHz flank is the "first" one per second.
So the timing accuracy does not come from the PPS directly, but from the
combination of PPS and 10 MHz.
Quick back-of-the-envelope: Let's say your velocity factor of your cable is
2/3, and I'll approximate 1ft as 1/3m. Then your additional delay is
approx. (1/3m)/(2/3 * 3 * 10^8 m/s) = 1.7 ns (probably better, I think 2/3
was lowballing it for good RF cables but the math is easy). Your 10 MHz
flanks are 100 ns apart. So your PPS is shifted relative to your 10 MHz
flank by about 1.5%... I would guess it works.
You could also have meant that the 2 cables going to the first USRP are
1ft, and the 2 cables going to the second USRP are 2ft. In that case, the
second USRP will be delayed by 1.7 ns relative to the first, but as long as
temperature stays constant, the relative delay will be constant. Here, your
signal bandwidth also matters. At 2 Msps sampling rate, such a delay
corresponds to approx. 3 samples. But if you have a low bandwidth, then the
constant delay will look like a (constant) phase offset, which you will
have anyway and need to account for.
--M
On Sat, May 17, 2025 at 1:13 PM Steve Hamn <stevehamn311@gmail.com> wrote:
> Hello,
>
> I have a pair of x440's I'm trying to phase cohere together. I wanted to
> understand how critical it is to have equal length 1PPS and Ref In cables?
> Right now I have a 2 ft RF cable going to one and a 1ft RF cable to
> another. Will that be a problem?
>
> Thanks
>
> Steve
> _______________________________________________
> USRP-users mailing list -- usrp-users@lists.ettus.com
> To unsubscribe send an email to usrp-users-leave@lists.ettus.com
>
