RX channel 1 out of range for configured RX frontends

GNU Radio
1

Hello,

We are using a TVRX2 board with a N200 USRP.    Until recently, the

system was working properly with two channels using the attached
flowgraph. For reasons that are unclear, it stopped working and is no
longer capable of two channel operation (if we set the Num channels
parameter to 1, the system works, but with only one channel). So far
as we can
tell, nothing in the S/W changed.

Version details and the error message appears below. Suggestions what
has gone wrong would be most appreciated.

–Dan

H/W: Dell Optiplex Dual 2.4 GHz, 2 GB / USRP N200 with TVRX2 GPSDO
boards.
OS: Ubuntu 14
gnuradio: 3.7.6

Traceback (most recent call last):
File “/home/marlow/gnuradio/TVRX2/top_block.py”, line 177, in
tb = top_block()
File “/home/marlow/gnuradio/TVRX2/top_block.py”, line 94, in init
channels=range(2),
File
“/usr/local/lib/python2.7/dist-packages/gnuradio/uhd/init.py”, line
122, in constructor_interceptor
return old_constructor(*args)
File
“/usr/local/lib/python2.7/dist-packages/gnuradio/uhd/uhd_swig.py”, line
1753, in make
return _uhd_swig.usrp_source_make(*args)
RuntimeError: LookupError: IndexError: multi_usrp: RX channel 1 out of
range for configured RX frontends

http://www.hep.princeton.edu/~marlow/gnuradio/TVRX2.png
http://www.hep.princeton.edu/~marlow/gnuradio/TVRX2.grc

2

On 11/03/2014 05:55 PM, Daniel M. wrote:

Version details and the error message appears below. Suggestions
what has gone wrong would be most appreciated.

–Dan
There’s a fix for this problem in the most recent “master” of GR, which
doesn’t appear in the “maint” branch.

It was a wee problem in gr-uhd that was introduced by wanting to be more
“strict” about master-clock LOCK, and, well, that provoked
other underlying problems.

What are you using the TVRX2 for, BTW. I’ll note that it has
cannot-be-defeated hardware AGC, so it’s of only limited utility for
radio astronomy, which I think you’re probably using it for… :slight_smile:

3

On Nov 3, 2014, at 5:59 PM, Marcus D. Leech wrote:

There’s a fix for this problem in the most recent “master” of GR, which doesn’t
appear in the “maint” branch.

It was a wee problem in gr-uhd that was introduced by wanting to be more
“strict” about master-clock LOCK, and, well, that provoked
other underlying problems.

What are you using the TVRX2 for, BTW. I’ll note that it has cannot-be-defeated
hardware AGC, so it’s of only limited utility for
radio astronomy, which I think you’re probably using it for… :slight_smile:

Hi Marcus,

Thanks. Does that mean we need a new install of GR? The one we
have is quite recent.

Regarding the TVRX2, I did not know about the AGC issue. We are using
it for RA (pulsar searches). Since we are doing synchronous
detection, it’s not clear to me that the AGC will be a fundamental
problem, since we don’t need quantitative power measurements. We wanted
two channels since we are using crossed Yagi’s. I don’t see any
alternatives from Ettus.

–Dan

4

On 11/03/2014 06:15 PM, Daniel M. wrote:

Hi Marcus,

Thanks. Does that mean we need a new install of GR? The one
we have is quite recent.
build-gnuradio by default installs the “maint” branch–since that is,
generally, more likely to be somewhat-more-stable. If you use the -m
option to
build-gnuradio, it’ll fetch “master”.

This fix is quite recent.

Regarding the TVRX2, I did not know about the AGC issue. We are
using it for RA (pulsar searches). Since we are doing synchronous
detection, it’s not clear to me that the AGC will be a fundamental
problem, since we don’t need quantitative power measurements. We
wanted two channels since we are using crossed Yagi’s. I don’t see
any alternatives from Ettus.

In terms of multi-channel daughtercards for N2xx/USRP1/X3xx/E1xx, yeah,
the TVRX2 is it.

But the B210 is dual-channel, strictly-coherent (they share the same
LO). Ken Tapping and I have been doing C-band correlation
interferometry with
one.