Forum: GNU Radio questions about calibration examples in UHD utility

Posted by Gary Lee (gary1943)
on 2013-01-28 21:23
(Received via mailing list)
Hi, guys,

I am reading the codes for uhd_cal_rx_iq_balance.cpp.
When doing calibration by setting the RX and TX antenna as "CAL", is
the signal from TX directly switched to RX?
And I noticed that in the function set_optimum_defaults(), the tx_gain
is 0 but the rx_gain is 25! Is that too high? I heard that when
directly connecting TX and RX, we should use attenuators. So why does
it set 25 rx_gain? Thanks.

Best,
Gang
Posted by Mike Jameson (Guest)
on 2013-01-28 22:16
(Received via mailing list)
Hi,

Calibration uses full duplex and should only be done without anything
plugged into the 'tx/rx' or 'rx2' ports. There is no direct connection,
just measurement of the residual rf escaping.

Mike
M0MIK
Posted by Gary Lee (gary1943)
on 2013-01-31 16:48
(Received via mailing list)
Hi, Mike,
Thanks very much for your reply. I also searched  online about the
usrp calibration. I found a post and some information:

"During cal, the TX is switched away from the TX antenna into the
"closed" switch port, where it leaks into the RX which is listening to
the TX/RX antenna. So termination on RX2 should have no effect as that
port is already unused in calibration."

Here is the link:
http://lists.ettus.com/pipermail/usrp-users_lists.....

I am a little confused. Since calibration uses full duplex, why is
there no direct connection between RX2 and TX/RX? I thought they are
connected inside the usrp during calibration.

Best,
Gang
Please log in before posting. Registration is free and takes only a minute.
Existing account (Switch to SSL-encrypted connection)
NEW: Do you have a Google/GoogleMail or Yahoo account? No registration required!
Log in with Google account | Log in with Yahoo account
No account? Register here.