On 2026-06-12 05:14, Rognskog, John-Harry via USRP-users wrote:
Hi Christian,
No, what you need is a good external clock. The internal clock will drift with
temperature and age etc and is not deterministic.
Depending on your price point, there are multiple options, from cheap GPSDO's
do the highest-end Fugro Atomichron ones (Sparkfun GNSSDO for example).
Kind regards,
John
Cheap GPSDOs are available from places like Leo Bodnar, and they have a
good reputation.
I've also sourced reasonably-priced 10MHz OCXO modules from Amazon of
all places.
The spec for the B200 is +/-2PPM, and 5kHz at 2.6GHz is within this
spec. Crystal oscillators are subject to thermal
effects, and *aging*, so may not always provide acceptable accuracy
over their lifetimes.
The whole reason for the external clock inputs on USRPs is for
applications where higher clock accuracy is required.
-----Original Message-----
From: Christian Zäske <[email protected]>
Sent: Friday, June 12, 2026 10:50 AM
To: Rognskog, John-Harry <[email protected]>; usrp-users
<[email protected]>
Subject: Re: [USRP-users] B200 frequency shifted
Hi John,
yes we are using the internal clock. It is just a bit weird that the
exact same setup worked in the past. We then had the SDR deployed in
another project and now wanted to try the setup again and it stopped
working because of the shift.
But if it is technically within spec and we can't really do anything to
fix this, I guess we just need a new SDR
On 6/12/26 10:42, Rognskog, John-Harry wrote:
Hi,
I assume you are using internal clock? 5kHz is within its spec (barely). My
units vary in that range, although not as much individually.
Kind regards,
John
-----Original Message-----
From: Christian Zäske via USRP-users <[email protected]>
Sent: Friday, June 12, 2026 10:35 AM
To: usrp-users <[email protected]>
Subject: [USRP-users] B200 frequency shifted
Hello,
(Sorry if this is a duplicate, I think there went something wrong with my last
email to the mailing list)
In the past, we used a B200 and srsRAN to build our own LTE cell, which always
worked flawlessly. After some time, we wanted to build this setup again but we
ran into an issue: The phone never found the cell.
After looking into the problem I noticed, that the output frequency of the B200
is shifted around +5 kHz at around 2.6 GHz (Verified it with another SDR). Too
much for the phone to get the signal properly. When I manually adjust the
frequency in srsRAN to compensate for the 5 kHz shift, the phone immediately
detects the cell.
At lower frequencies, the shift also gets lower.
Is there something we can do to revive the B200? Because this exact setup
worked in the past, I think something happened to the b200 that causes this
shift.
Thanks in advance,
Christian
_______________________________________________
USRP-users mailing list -- [email protected]
To unsubscribe send an email to [email protected]
