WBX power level

GNU Radio
1

Hi All,

Has anyone measure the power level
coming out of their WBX board?

Measured on a Gigatronics 8541 microwave
power meter connected directly at the
back of the USRP2 I am getting (with a CW carrier)

100 MHz about 35 mW
1.3 GHz about 15 mW

Ettus quote

50 - 100 mW < 1.2 GHz
30 - 70mW > 1.2 GHz

I know measuring power is notoriously inaccurate
but I was wondering what others have found.

Setup USRP2 WBX UHD

  • Charles
2

On 06/12/2010 11:15 AM, Charles B. wrote:

Ettus quote

50 - 100 mW < 1.2 GHz
30 - 70mW > 1.2 GHz

I know measuring power is notoriously inaccurate but I was wondering
what others have found.

Setup USRP2 WBX UHD

Is your flowgraph delivering the maximum amplitude to the USRP?

Philip

3

On 6/12/2010 11:54 AM, Philip B. wrote:

100 MHz about 35 mW
Setup USRP2 WBX UHD

Is your flowgraph delivering the maximum amplitude to the USRP?

Philip

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

It looks like your operating at half power, which I believe is default
if you don’t supply the proper amplitude and gain values.

Tim

–

4

----- Original Message -----
From: “Philip B.” [email protected]
To: “Charles B.” [email protected]
Cc: [email protected]
Sent: Saturday, June 12, 2010 4:54 PM
Subject: Re: [Discuss-gnuradio] WBX power level

Is your flowgraph delivering the maximum amplitude to the USRP?

Philip

I am not using gnuradio I am using
C++ code that talks directly to the USRP2 using
the UHD interface and I have it writing 0x7FFF
to the I channel and 0 to the Q channel. I also
tried 0x7FFF to both I and Q with similar results.

  • Charles
5

Ettus quote

50 - 100 mW< 1.2 GHz
30 - 70mW> 1.2 GHz

I know measuring power is notoriously inaccurate but I was wondering
what others have found.

Setup USRP2 WBX UHD

Please try the following command with non-UHD for comparison:

usrp_siggen_gui.py -x 250k -i 64 --tx-gain=25 -A 1.0 -f 1.3G

You should see about 50mW (17dBm) at 1.30025GHz

If you change the final argument to 100M you should see around
80-100mW (19-20dBm) in the tone at 100.25MHz

If you can confirm those numbers, perhaps we have a constant scalar
missing in either your flowgraph or the UHD code someplace.

Thanks
Jason

6

On Mon, Jun 14, 2010 at 11:03 AM, Jason A. [email protected] wrote:

Please try the following command with non-UHD for comparison:

usrp_siggen_gui.py -x 250k -i 64 --tx-gain=25 -A 1.0 -f 1.3G

My mistake here, please use -i 16 …
usrp_siggen_gui.py -x 250k -i 16 --tx-gain=25 -A 1.0 -f 1.3G

7

----- Original Message -----
From: “Jason A.” [email protected]
To: “Charles B.” [email protected]
Cc: [email protected]
Sent: Monday, June 14, 2010 7:03 PM
Subject: Re: [Discuss-gnuradio] WBX power level

Ettus quote

50 - 100 mW< 1.2 GHz
30 - 70mW> 1.2 GHz

I know measuring power is notoriously inaccurate but I was wondering
what others have found.

Setup USRP2 WBX UHD

missing in either your flowgraph or the UHD code someplace.

Thanks
Jason

Hi Jason,

I got the test done this morning (after I re-installed gnuradio 3.2.2 )

Using

usrp_siggen_gui.py -x 250k -i 16 --gain=25 -a 1.0 -f 1.3G

60mW @ 1.3 GHz
126mW @ 100 MHz

Using my code and UHD (my code modified to write I = 0x7FFF, Q =
0x7FFF)

14.4mW @ 1.3 GHz
33.8 mW @ 100 MHz

Using my code and UHD ( my code modified to write I = 0x7FFF, Q = 0x0)

14.8mW @ 1.3 GHz
24.8 mW @ 100 MHz

I ran the power meter’s self calibration routine before taking the
measurements
this time and applied the correct frequency correction.

Looks like we have a bit missing :frowning:

  • Charles
8

Does your code set the transmit gain on the WBX?

Eric

Yes to 25. I checked the range and set it to maximum.

  • Charles
9

missing in either your flowgraph or the UHD code someplace.
usrp_siggen_gui.py -x 250k -i 16 --gain=25 -a 1.0 -f 1.3G

http://www.g4guo.blogspot.com/
Does your code set the transmit gain on the WBX?

Eric

10

Well I am glad to say my low power problem is fixed.

In the end I had to go into my uhd directory do a git pull
followed by a make and then a make install.
I now have 60 mW on 1.3 GHz.

Thanks for the help guys, especially to Jason A…

  • Charles
11

On Tue, Jun 15, 2010 at 06:48:39PM +0100, Charles B. wrote:

Does your code set the transmit gain on the WBX?

Eric

Yes to 25. I checked the range and set it to maximum.

  • Charles

Charles,

In my test with USRP2 and UHD here in the lab, I am able to get similar
power output numbers to when I used usrp_siggen_gui.py.

Here are my best guesses:

Try my attached grc/python code for transmitting with gr-uhd, assuming
you have built a gnuradio from branch next, which includes gr-uhd.

Probe the voltage at R532, it should vary from 1.4V to 0.5V with
different transmit gain settings. Ok, R532 is on the bottom, so instead
try probing at the via between the “2” and “1” in the C521 label (which
is near the C523 label).

I am using my own code that writes directly to the UHD and I forced it to
set the I channel to 0x7FFF
and the Q channel to 0 and the gain to 25.

Its also possible your app is not sending the correct type data for that
DC tone. (ie, sending int on an interface expecting float)

see host/include/uhd/types/io_type.hpp

Jason