[eagle] 70 cm Receiver Phase Noise

Juan Rivera juan-rivera at sbcglobal.net
Mon Sep 17 18:42:51 PDT 2007


Guys,

 

While you wait for the USRP files here's some background info using the
SDR-IQ:

 



The screen shot above is the IF output of the 70 cm Receiver with an RF
input of -60 dBm.  FFT averaging is set to 100 and the span is 200 Hz.

 



I then matched the level of the Receiver's IF output using a step attenuator
and fed a signal from a battery-powered HP reference oscillator I salvaged
from an old signal generator into the SDR-IQ.  The SDR-IQ settings remain
exactly the same.  You can see the difference.

 

 



Since the reference oscillator runs on batteries I was able to take it to
work last month and test the phase noise using a hugely expensive Agilent
E5500 phase noise measurement subsystem.  The shot above is the results of
the phase noise test on the oscillator so you can correlate the SDR-IQ
against the Agilent.  So for example, the battery-powered oscillator phase
noise is down -130 dBc/Hz at an offset of 100 Hz.  The SDR-IQ says it's down
about -115 dBc.  That's not bad since we're comparing a $400 box against
something that probably cost over $100,000.  If we assume that adding
another -15 dB to the SDR-IQ brings it into line with the Agilent then the
70 cm Receiver has a phase noise of about -85 dBc/Hz at 200 Hz offset.  I'll
be curious to see what the USRP has to say about this.

 

There are some external sources of phase noise that need to be considered -
The SDR-IQ has its own issues of course, and the Receiver was using my GPS
disciplined clock as an external reference.  The clock is very clean but
nothing is perfect.  I can't use the internal reference since it jumps all
over the place and will be changed out in the next rev anyway.

 

IMPORTANT - Last night I think I noticed that the receiver phase noise was
greatly affected by the quality of the external reference.  If this is true
then we need to consider how that external reference will be generated and
distributed on Eagle and what sort of quality we can expect - or the
Receiver needs to be designed so it can function with a dirty external
reference.  I'll look into this as time permits.  We hope to get the USRP
over here this weekend and rerun these tests with it.

 

73,

 

Juan

 

 

-------------- next part --------------
An HTML attachment was scrubbed...
URL: http://amsat.org/mailman/private/eagle/attachments/20070917/13637790/attachment.html
-------------- next part --------------
A non-text attachment was scrubbed...
Name: not available
Type: image/gif
Size: 19148 bytes
Desc: not available
Url : http://amsat.org/mailman/private/eagle/attachments/20070917/13637790/attachment.gif
-------------- next part --------------
A non-text attachment was scrubbed...
Name: not available
Type: image/gif
Size: 19049 bytes
Desc: not available
Url : http://amsat.org/mailman/private/eagle/attachments/20070917/13637790/attachment-0001.gif
-------------- next part --------------
A non-text attachment was scrubbed...
Name: not available
Type: image/jpeg
Size: 77850 bytes
Desc: not available
Url : http://amsat.org/mailman/private/eagle/attachments/20070917/13637790/attachment.jpe


More information about the Eagle mailing list