Using external clock with USRP2/WBX

Hi all,
I’m trying to use the external clock input of the usrp2 with WBX in
receive
mode. Unfortunately after processing the recorded signal there is still
a
clock drift despite the fact i’m using a GPS reference clock. So i
suppose
that i’m making something wrong.

  1. I’m using a square 10 Mhz reference from a GPS receiver (Novatel
    OEM4)
    which has a 300 mV peak-to-peak amplitude → this amplitude is
    sufficient ?
  2. I activated the external clock by selecting the external SMA option
    in
    the usrp2_source block in GRC. Could somebody confirm that is method
    works ?

Damien

View this message in context:
http://old.nabble.com/Using-external-clock-with-USRP2-WBX-tp28871938p28871938.html
Sent from the GnuRadio mailing list archive at Nabble.com.

On Sun, 2010-06-13 at 08:12 -0700, Damien S. wrote:

Hi all,
I’m trying to use the external clock input of the usrp2 with WBX in receive
mode. Unfortunately after processing the recorded signal there is still a
clock drift despite the fact i’m using a GPS reference clock. So i suppose
that i’m making something wrong.

Can you specify how much offset you see?

I presume that the transmitter is also using a precision timebase?

  1. I’m using a square 10 Mhz reference from a GPS receiver (Novatel OEM4)
    which has a 300 mV peak-to-peak amplitude -> this amplitude is sufficient ?

I’m not familiar with that unit, but 300mV should work.

I’ve used a Trimble Thunderbolt and a Jackson Labs Fury GPSDO without
issue.

  1. I activated the external clock by selecting the external SMA option in
    the usrp2_source block in GRC. Could somebody confirm that is method works ?

Yes, it works. I just tested the GRC source and sinks with a signal
generator and my USRP2 definitely tracks the “External SMA” clock.

–Eric

Eric S.-4 wrote:

Can you specify how much offset you see?
I see an clock offset of about 2-4 ppm (almost the same with or without
external clock).

Eric S.-4 wrote:

I presume that the transmitter is also using a precision timebase?
I’m recording a signal from a DTV transmitter which is, according to its
owner, synchronized with a GPS receiver. But unfortunately i can’t be
100%
sure of that.
And as i’m using a GPS reference in the external clock input i hoped
that
the observed clock drift would be null but it’s not the case.

Eric S.-4 wrote:

Yes, it works
Ok now i eliminated a possible source for my problem. Remains:
transmitter
not synchronized on GPS time, GPS reference not good, error in my
post-processing to find the clock drift… I continue the investigations
to
find the probelm source.

Eric S.-4 wrote:

Can you specify how much offset you see?
issue.


Discuss-gnuradio mailing list
[email protected]
Discuss-gnuradio Info Page


View this message in context:
http://old.nabble.com/Using-external-clock-with-USRP2-WBX-tp28871938p28891119.html
Sent from the GnuRadio mailing list archive at Nabble.com.