[USRP-users] [Discuss-gnuradio] Recording continuous I-Q stream and frequency offset with an external reference clock

Nazmul Islam mnislam at winlab.rutgers.edu
Wed Jun 13 14:29:12 EDT 2012


Thanks a lot for your email. My previous USRP2 modules had some old
daughterboards. I am currently working with USRP N210. I am not sure even
this one is getting the external reference, either. Couple of things:

1. I have the tools to probe the 10 MHz reference signal. But I did not
find R523 :S. I can see R521, R522, R524, C523, etc. But I did not see R523
in USRO N210.

2. I have kept the J510 jumper at 1-2. Therefore, it should take signal
from external reference clock instead of GPS.

3. When I use the "external" clock source option in UHD:USRP Sink block of
gnuradio-companion, the E led turns off. On the other hand, if I transmit
with "default" clock source option in UHD: USRP sink block, the E led turns
on !!! It's very surprising !

4. I did another experiment to check if the USRP is locked to the external
reference. At first, I gave an input of 10 MHz sine wave as an external
source and transmitted a signal at 700 MHz carrier frequency. The carrier
frequency option was given at gnuradio-companion software. Thereafter, I
varied the reference clock frequency slightly (from 10 to 10.3, 10.4 MHz,
etc.). During this period, I observed the signal at my laboratory spectrum
analyzer. If there was a fixed gain that converted the 10 MHz reference
source to 700 MHz carrier frequency, the transmitted signal's center
frequency would have shifted from 700 MHz because the fixed gain would
multiply 10.1 instead of 10 now. However, the center frequency did not

I suspect that my USRP is still locked to its internal clock source. Since
this happened to two different USRP's, the issue might be in software. Has
anyone used external clock source from a gnuradio-companion generated
python code? I am changing the external clock reference source in GRC. Do I
need to make any other change?

Thanks a lot for your help.


On Fri, Jun 8, 2012 at 12:53 PM, John Malsbury <john.malsbury at ettus.com>wrote:

> Nazmul,
> Do you have the tools(o-scope) and capacity to probe the 10 MHz reference
> signal at various places?  Looking at the schematic, it looks like R523 is
> a good place to determine if the 10 MHz reference is a good place.  If you
> don't see the external reference across this resistor, there may be a
> problem with the reference input or conditioning circuitry.  If we
> eliminate this as a possibility we can investigate some others...
> -John

Muhammad Nazmul Islam

Graduate Student
Electrical & Computer Engineering
Wireless Information & Networking Laboratory
Rutgers, USA.
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://lists.ettus.com/pipermail/usrp-users_lists.ettus.com/attachments/20120613/63502e7b/attachment-0002.html>

More information about the USRP-users mailing list