USRP not properly configured or HW broken?

I am trying to make a setup to capture GNSS signals.

I bought a very cheap antenna, a power supply and I am using a USRP1. The setup more or less looks like this:

Antenna <-> DC_Blocker <-> USRP_Daughterboard (DBSRX2 USRP) <-> USRP1

A PSU is connected to the DC_Blocker providing with 5V DC. I checked with a voltmeter; everything was fine (and the DC is not getting to the USRP).

The Daughterboard is connected to the RXB socket in the motherboard (see the pictures).

I am using the GRC to record the signal, and you can find the diagram in the links below.

I created a small Python program to read the recorded signals, as I was unsure whether I had a problem in the reader I did the same with a Sine signal generated by the GRC. As you can see in the picture, the Python programme can read and plot the sine signal. But when it comes to the recorded signal from the USRP, it plots something strange. It seems like either I did not configure the USRP source signal properly or I have an HW problem (which could perfectly be the case).

The signal I see (and the QT time Sink shows something inline with it) with the USRP source is some kind of transient signal and then 0. It does not matter what "Mb0: Subdev spec" I set (I think it should be B:0) and "CH0: Antenna" is "TX/RX", the other option was RX2, I tried with both.

Signal "from" the antenna and the USRP:

The signal generated with the GRC sine source then recorded to a file and plotted with the same Python programme (so I guess the Python programme is fine).

You can see the pictures of the recorded signals and the configuration in the GRC USRP source.

Any idea on whether I configured the GRC USRP source wrong or shall I look into the HW part?

Thanks! Fran

GRC USRP source configuration 1: https://drive.google.com/file/d/1IQaOqLtZGeoXpD1stHZQ1dPb8_UJqT2z/view GRC USRP source configuration 2: https://drive.google.com/file/d/1yQ41eD51suVBg2Ahzu6yWanXCmYzq80z/view GRC USRP source configuration 3: https://drive.google.com/file/d/1dT37br1MpMp7c1wwGROBLt7C47MZIsas/view GRC USRP source configuration 4: https://drive.google.com/file/d/1gQrivWyLzu5xEk-z-zGF-j7xVvkMt5NC/view

GRC diagram: (GRC file) https://drive.google.com/file/d/1PICfljNcf2TF-lLb01rHUA2AQvyznHAO/view?usp=sharing

Picture of the USRP with the daughterboard disconnected: https://drive.google.com/file/d/1FvUju99H_Zm0rjZpJ5y4DREX5dFzVFbU/view Picture of the USRP with the daughterboard connected: https://drive.google.com/file/d/14MP5uofLP1sz08-mBDeVDHoH9DAWaMsU/view

Signal recorded with the antenna (A:B and Chain RX2) https://drive.google.com/file/d/1LGB-ErfakBUFWnEqMflxj1mznFsqvmAi/view

Signal recorded with the antenna (B and Chain TX/RX) https://drive.google.com/file/d/1xXy8pTOv9mmCll851CC8fzgiTTI-xH9p/view

Sine signal generated with the GRC and plotted with the same python program: https://drive.google.com/file/d/1kRBr5HFlEmYAe2OgKFwkkHl8GGQCxUDT/view

EDIT: This is the SC of the Python reader: https://github.com/fgallardo/sandbox/blob/master/USRP_recorded_signal_reader.py

Ok, after asking in the Ettus mail list. See the discussion here:

http://lists.ettus.com/pipermail/usrp-users_lists.ettus.com/2020-April/061794.html

It seems like my gain in the daughterboard was too low and the sampling rate was not correct.

Now, the sampling rate is 8 Msps. (Be aware it has to be a strict-integer fraction of the master clock rate, which is 64 MHZ).

Now I see the signal in the time plot. I am still unsure about it because the signal is not clearly visible in the search space, but I think I am going to find a better antenna to continue.

Thanks to you all and kudos to Marcus Leech from the Ettus forum who pointed me to the mistake in a matter of minutes!

Time plot of the signal, noise is now visible (as expected).

Search Space with GAL E24, there seems to be a peak, although it is not very stable, nor clear. I am going to upgrade my antenna.