Ok, thank you for the information, that has been really helpful. I will get back to you with results once I am able to get the inline attenuator and run the tx_waveforms test.
Best Regards, Jerrid From: Marcus D. Leech <patchvonbr...@gmail.com> Sent: Thursday, March 25, 2021 4:07 PM To: Jerrid Plymale <jerrid.plym...@canyon-us.com> Cc: USRP-users@lists.ettus.com Subject: Re: [USRP-users] Strong noise added to signal transmitted by X310 with a UBX 40 daughterboard On 03/25/2021 07:01 PM, Jerrid Plymale wrote: We currently are not using a 30dB attenuator in the coax line, I was not aware that one was needed until you brought it up. We are looking into getting one ordered at the moment. Could this be the cause of the problem? Best Regards, Jerrid The problem is that if you transmit directly into an RX with the TX power turned up, you can fry the RX. It also provides a much more realistic simulation of an over-the-air link. ANY receiver that is designed for "Over The Air" reception will have a pre-amp that doesn't like to receive power levels above about -15dBm (which for over-the-air signals is very very loud). With the noise figure of your RX (if it isn't damaged already) being only about 4dB, it's going to be fairly sensitive to any low-level noise that happens to get coupled out of the X3xx electronics when the mixer and RF amplifier are "brought up" when a flow-graph is started. Even when you're transmitting with a "0dB" setting, it can still be producing up to about -20dBm power output--the attenuator is 31.5dB in 0.5dB steps, and the max Pout of the UBX is about +10dBm as I recall. Now, with zeros going to the ADC, the IF port of the mixer isn't going to be seeing much power, but it won't be seeing exactly zero power because of the inevitable DAC noise. It will be instructive to see your results with a 30dB attenuator in-line and using the suggested tx_waveforms test. From: Marcus D Leech <patchvonbr...@gmail.com><mailto:patchvonbr...@gmail.com> Sent: Thursday, March 25, 2021 3:40 PM To: Jerrid Plymale <jerrid.plym...@canyon-us.com><mailto:jerrid.plym...@canyon-us.com> Cc: USRP-users@lists.ettus.com<mailto:USRP-users@lists.ettus.com> Subject: Re: [USRP-users] Strong noise added to signal transmitted by X310 with a UBX 40 daughterboard Could you confirm that you’re using at least 30dB of attenuation in the coax link? Sent from my iPhone On Mar 25, 2021, at 6:27 PM, Jerrid Plymale <jerrid.plym...@canyon-us.com<mailto:jerrid.plym...@canyon-us.com>> wrote: Yes, moving TX out of the RX band has no affect on the noise level increase. Best Regards, Jerrid From: Marcus D Leech <patchvonbr...@gmail.com<mailto:patchvonbr...@gmail.com>> Sent: Thursday, March 25, 2021 2:29 PM To: Jerrid Plymale <jerrid.plym...@canyon-us.com<mailto:jerrid.plym...@canyon-us.com>> Cc: USRP-users@lists.ettus.com<mailto:USRP-users@lists.ettus.com> Subject: Re: [USRP-users] Strong noise added to signal transmitted by X310 with a UBX 40 daughterboard If you move the TX out of band with respect to the RX do you still see the noise when the TX graph starts? Sent from my iPhone On Mar 25, 2021, at 3:57 PM, Jerrid Plymale <jerrid.plym...@canyon-us.com<mailto:jerrid.plym...@canyon-us.com>> wrote: So I was able to test setting up both USRPs with the same 10 MHz reference signal, and there was no improvement to the noise being added. If the flowgraph just emits a CW tone, we see a similar response, the added noise is still there and at the same level, we just have the added spike of the CW tone at the frequency we set it to. Best Regards, Jerrid From: Marcus D Leech <patchvonbr...@gmail.com<mailto:patchvonbr...@gmail.com>> Sent: Wednesday, March 24, 2021 5:01 PM To: Jerrid Plymale <jerrid.plym...@canyon-us.com<mailto:jerrid.plym...@canyon-us.com>> Cc: USRP-users@lists.ettus.com<mailto:USRP-users@lists.ettus.com> Subject: Re: [USRP-users] Strong noise added to signal transmitted by X310 with a UBX 40 daughterboard You don’t need 1PPS for this. Having an external high quality reference may help, assuming my guess about your issue is correct. What happens if your flow graph just simply emits a Cw tone, using the mute function perhaps tied to a GUI control or some such. Sent from my iPhone On Mar 24, 2021, at 7:13 PM, Jerrid Plymale <jerrid.plym...@canyon-us.com<mailto:jerrid.plym...@canyon-us.com>> wrote: Ok, if that’s the case, would it help to have both USRPs connected to the same 10 MHz reference signal and PPS? In this situation, would I need to provide a secondary source for the PPS, or can that use the 10 Mhz reference as well? Best Regards, Jerrid From: Marcus D. Leech <patchvonbr...@gmail.com<mailto:patchvonbr...@gmail.com>> Sent: Wednesday, March 24, 2021 2:23 PM To: Jerrid Plymale <jerrid.plym...@canyon-us.com<mailto:jerrid.plym...@canyon-us.com>> Cc: USRP-users@lists.ettus.com<mailto:USRP-users@lists.ettus.com> Subject: Re: [USRP-users] Strong noise added to signal transmitted by X310 with a UBX 40 daughterboard On 03/24/2021 05:10 PM, Jerrid Plymale wrote: The devices are operating using direct connection via coax cables. You absolutely need to put a 30dB attenuator in-line to prevent RX destruction in the case of "ooops" moments from setting the TX power output too high. The target frequency is 1.57542 GHz, and our current bandwidth is 4 MHz. We will need to increase the bandwidth to 20 MHz soon for further system development. The TX and RX gain are maxed on the receiving hardware. The TX gain of the transmitting hardware is set to 0, as at max the noise strength increases to ~ 20 dB. Attached are images of the FFT provided by the Frequency Sink in Gnuradio. Hopefully these give a visual aid for the problem at hand. When I have the transmitter USRP turned off (image 1), it would seem like the noise floor coming into the USRP is around -94 dB. When the transmitter is turned on and the flowgraph is started with the transmitted signal muted (I am using a python block with code to create custom signals that is then connected to a mute block that then connects to the UHD USRP sink block to be able to mute the signal during runtime), the noise floor increases to around -78 dB. Best Regards, Jerrid You are likely just seeing the LO leakage, along with the inevitable phase-noise curve of the LO. From: Marcus D Leech <patchvonbr...@gmail.com><mailto:patchvonbr...@gmail.com> Sent: Wednesday, March 24, 2021 11:58 AM To: Jerrid Plymale <jerrid.plym...@canyon-us.com><mailto:jerrid.plym...@canyon-us.com> Cc: USRP-users@lists.ettus.com<mailto:USRP-users@lists.ettus.com> Subject: Re: [USRP-users] Strong noise added to signal transmitted by X310 with a UBX 40 daughterboard Is the j B is over-the-air or direct connection? What frequency? Bandwidth? Do you have TX and RX gain turned all the way up? Can you share your flow-graphs, or minimum Graphs that show the problem? Sent from my iPhone On Mar 24, 2021, at 12:20 PM, Jerrid Plymale <jerrid.plym...@canyon-us.com<mailto:jerrid.plym...@canyon-us.com>> wrote: Hello All, I have been running tests in which I am transmitting a signal from one USRP X310 that’s using a UBX 40 daughterboard, and that signal is being received by another USRP X310 using a UBX 40 daughterboard. I have noticed that when I have the receiving USRP running with the Gnuradio flowgraph active, as soon as I start the transmitting USRP’s Gnuradio flowgraph, there is a jump in the noise floor as it is seen by the receiving USRP, and its roughly a 14 dB increase. This occurs even if I have the signal being transmitted muted. Does anyone have any idea what the source of this added noise could be? It is something that I need to mitigate as much as possible for the tests I am running using these USRPs. Any feedback will be greatly appreciated, thanks! Best Regards, Jerrid _______________________________________________ USRP-users mailing list -- usrp-users@lists.ettus.com<mailto:usrp-users@lists.ettus.com> To unsubscribe send an email to usrp-users-le...@lists.ettus.com<mailto:usrp-users-le...@lists.ettus.com>
_______________________________________________ USRP-users mailing list -- usrp-users@lists.ettus.com To unsubscribe send an email to usrp-users-le...@lists.ettus.com
