Hi Robert, Thanks for the follow up. One question I have is related to your comment that you needed 3.15 for Tx but you didn't mention a version requirement for Rx. In my own tests, I found that I had to use 3.15 even for Rx or else I would get varying phase behavior. If you are able to get constant results with say 3.14.1.1, then I wonder if perhaps I had something else wrong and fooled myself into thinking that the UHD version fixed my issue. Rob
On Tue, Feb 18, 2020 at 8:17 AM <[email protected]> wrote: > Hi Nate, Rob, Sammy, > > > > thanks for pointing out that the external LO should be set to 5 GHz for > the QEC init calibration. So far I had omitted this calibration, as it > wasn’t workin. This is important, please add it to the documentation (maybe > it’s there and I missed it?). > > > > Happy to share some of our experiences regarding coherent operation of the > N310: > > We are successfully using the N310 with coherent RX for DoA estimation. > Our setup is: > > - external LO with 4-way splitter -> 2 TX 2 RX LO inputs > > - tracking_cals=OFF > > - At the beginning we run a flowgraph with force_reinit=1 and > the desired init_cals settings > > - Later we use exactly the same settings without force_reinit, > this ensures that no reinit is performed > > - The RX phase offset between the 4 channels is now constant, > running different GnuRadio flowgraphs is no problem > > - When the device is power-cycled or a full re-init is > performed, the phase of the radios can jump by 180° > > > > Very recently we added coherent TX, aiming at TX beamforming. Not much > experience yet, but it seems to work. > > - Same setup as above. > > - Important: UHD 3.15.0.0 is required, some DUC bug has been > fixed with this release > > - TX phase appears to be stable until power-cycle / re-init > > > > This even works with RFNoC (the RX part, TX not tested yet), although > quite some modifications are necessary to gr-ettus to get synchronized > streams. > > > > Regards, > > Robert > > > > > > > > *From:* USRP-users [mailto:[email protected]] *On Behalf > Of *Nate Temple via USRP-users > *Sent:* Tuesday, January 28, 2020 9:03 PM > *To:* Sammy Welschen > *Cc:* usrp-users; Rob Kossler > *Subject:* Re: [USRP-users] DOA with N310 or X310+TwinRX > > > > Hi Sammy, > > >Can I turn it off and come back the next day and still have the same > phase offset between the channels that I had the day before? > > Yes. This assumes that you are running with the same frequency, gain, > sample rate and system temperature that your calibrations were made with. > Also unless you have phase stable cables, if you move your cables at all, > it can cause phase variation. > > > Regards, > Nate Temple > > > > On Mon, Jan 27, 2020 at 9:29 AM Sammy Welschen <[email protected]> > wrote: > > Hi Nate, thank you for the information. > > > > I'm still a bit unsure what repeatable phase offset means exactly. Suppose > I have a system with 8 channels with X310+TwinRX and shared LO. Can I turn > it off and come back the next day and still have the same phase offset > between the channels that I had the day before? > > > > Sammy > > Nate Temple via USRP-users <[email protected]> schrieb am Mo., > 27. Jan. 2020, 18:04: > > Hi Rob, Robert, Sammy: > > Generally for this type of application we would recommend the X310+TwinRx. > With the TwinRX, you'll be able to have repeatable phase offsets with a > given gain, frequency, sample rate and temperature of a device/system. The > N310 will have a 180 degree phase ambiguity due to the /2 LO architecture. > > It is possible to share the LO across multiple motherboards for a > X310/Twin setup, and with the NI branded X310+TwinRX setup (NI-2955) the > LO's are provided out of the back panel. The chassis for currently shipping > and Rev C, F, G X310's back plate has the holes for the LO cables, but the > sticker needs to be removed. This application note covers the process: > https://kb.ettus.com/Modifying_an_X310_Chassis_for_External_LO_Sharing > > You'll also need to provide a splitter and most likely an inline amplifier > to overcome splitter losses. A splitter such as the ZFRSC-4-842+ will work. > https://www.minicircuits.com/pdfs/ZFRSC-4-842+.pdf > > > @Rob: With the current init process of the N310, yes it is required to > first set the external LO to 5 GHz. > > With regards to the offsets you're seeing, I believe you should only see a > possible phase difference of 180* within the two channels on the same DB. > Are you issuing a tune request at the start of streaming? > > Regards, > Nate Temple > > > > On Mon, Jan 27, 2020 at 8:20 AM Rob Kossler via USRP-users < > [email protected]> wrote: > > Robert, Sammy, > > I am presently running some tests which compare the X310/TwinRx and the > N310 with regard to channel-to-channel phase. In my setup, I have a signal > source that is split 8 ways (1:8 splitter) to feed the 4 channels of my > TwinRx and 4 channels of my N310. I have seen some strange behavior of the > N310 that perhaps Robert has experienced? Take a look: > > - For the TwinRx (for which I am a more experienced user with LO > sharing), I get consistent channel-to-channel phase difference among all > channels. This is true regardless of power cycles, re-starts of UHD, etc. > - For the N310 (for which I am a beginner when it comes to external LO > operation) > > > - it seems more complex to run in this mode (as compared to TwinRx). > In order to get it to work, I have had to disable startup QEC > calibration > because it seems that the N310 initial cal occurs at 2500 MHz RF such > that > I would need to have my external LO at 5000 MHz for startup (during the > UHD > deveice 'make') and then later switch my external LO to the desired > RF*2. > Is this true? > - when I run with either external LO or internal LO, I see > inconsistent channel-to-channel phase results even between the two > channels > of a given daughterboard that share the same LO. I do not understand > how > this is possible. My results over 16 captures (with some re-starts of > UHD, > device reboots, and switching between internal/external LO) show the > following channel-to-channel phase difference between channels 0 and 1 > which share the same LO: (values in degrees) -77, -19, -77, -19, -77, > -19, > -19, 39, -19, -19, -77, -19, -77, 39, -19, -19. Note that there are > only 3 > unique values and the delta happens to be 58 deg, but I don't know what > that implies... > > Rob > > > > > > On Mon, Jan 27, 2020 at 9:57 AM Robert via USRP-users < > [email protected]> wrote: > > With external LO its 300 MHz – 4 GHz, check footnote [3] from > https://www.ettus.com/all-products/usrp-n310/. LO has to be supplied at > twice the carrier freq. > > > > Currently we use 4 channels. You can find an example how to do the > calibration here: https://github.com/EttusResearch/gr-doa > > gr-doa was written for TwinRX, but can be adapted. > > > > Phase noise behavior of N310 and N320/1 could be different, as N310 uses > an RFIC and N32/1 use discrete components. This could be important if you > want to operate in the small sample regime. > > > > > > *From:* USRP-users [mailto:[email protected]] *On Behalf > Of *Sammy Welschen via USRP-users > *Sent:* Monday, January 27, 2020 3:40 PM > *To:* [email protected] > *Subject:* Re: [USRP-users] DOA with N310 or X310+TwinRX > > > > Thank you for the information Robert! Isn't it 6 GHz? However, 4 GHz would > also be sufficient for me. > > > > How many channels does your system have? I suppose you use some algorithm > for phase calibration after power cycling? I plan to do the same, so the > 180 deg ambiguity should be manageable. > > > > I looked at the N32x, however, they cost twice as much and I dont't plan > on using 200 MHz of bandwidth. If I have an external LO signal I can feed > it to the N310, so the only difference between N310 and N32x in this regard > would be that I need to generate the LO externally when using the N310, > right? > > > > <[email protected]> schrieb am Mo., 27. Jan. 2020, 14:42: > > We use the N310 for DoA estimation, however: > > - you are limited to 4 GHz > > - after power-cycling you get a 180° ambiguity between the two > radios (I do not know if this could also happen when you just change the LO > frequency) > > > > If you want to have >4 channels, have a look at the new N320/N321. No > experience with those, but apparently they can do LO distribution. > > > > Also take into account if maybe later in the project you want to be able > to transmit, which you cannot do with TwinRX. > > > > Regards, > > Robert > > > > *From:* USRP-users [mailto:[email protected]] *On Behalf > Of *Sammy Welschen via USRP-users > *Sent:* Monday, January 27, 2020 2:19 PM > *To:* [email protected] > *Subject:* Re: [USRP-users] DOA with N310 or X310+TwinRX > > > > Thank you Marcus! So the N310 would be the way to go? I was unsure since > the TwinRX is recommended for phase coherent applications. > > > > Marcus D. Leech via USRP-users <[email protected]> schrieb am > So., 26. Jan. 2020, 18:57: > > On 01/25/2020 11:43 AM, Sammy Welschen via USRP-users wrote: > > Dear all, > > > > I am planning a system with 5-10 channels that is capable of DOA > > estimation. > > > > Concerning the calibration of the resulting array, would there be a > > difference between a system made up of N310 and one made up of X310 > > with TwinRX boards? Would there be other important differences that > > influence estimation performance? > > > > As I understand it, the TwinRX allows LO sharing between the boards in > > a single X310, but this would not help me if I have two or three X310. > > On the other hand, the N310s could be connected to a shared LO. > > > > Are the following thoughts correct? > > > > Suppose I turn on my system. Then I have to calibrate phase offsets > > between channels in any case. Now I change the center frequency. If I > > have N310s without shared LO, I have to recalibrate. Same for the > > X310s, since LOs are shared only internally. If I have N310s with a > > shared LO, I do not have to recalibrate. > > > > If I restart the system, I have to recalibrate in any case. > > > > The devices would by synchronized with PPS in any case and with the 10 > > MHz reference if no external LO is used. > > > > What is the better choice for DOA estimation (N310 or X310 with TwinRX > > or something different)? > > > > Thank you very much > > > > Sammy > > > > > Sammy: > > Your characterization of the two scenarios is correct. > > There may be some folks on this list who have implemented DOA schemes, > but likely few-to-none who have done it on both X310 and N310 > and can comment on the differences they encountered. > > > > > _______________________________________________ > USRP-users mailing list > [email protected] > http://lists.ettus.com/mailman/listinfo/usrp-users_lists.ettus.com > > _______________________________________________ > USRP-users mailing list > [email protected] > http://lists.ettus.com/mailman/listinfo/usrp-users_lists.ettus.com > > _______________________________________________ > USRP-users mailing list > [email protected] > http://lists.ettus.com/mailman/listinfo/usrp-users_lists.ettus.com > > _______________________________________________ > USRP-users mailing list > [email protected] > http://lists.ettus.com/mailman/listinfo/usrp-users_lists.ettus.com > >
_______________________________________________ USRP-users mailing list [email protected] http://lists.ettus.com/mailman/listinfo/usrp-users_lists.ettus.com
