Description

    Test of DBBC2 fullband mode to be used when co-observing with ALMA.

    Participating stations:

      Log Prc
    Ef t162aef.log t162aef.prc
    On t162aon.log t162aon.prc
    Yb t162ays.log t162ays.prc
    Mh t162amh.log t162amh.prc

    Notes

    - Ys had trouble with flexbuf recording in the beginning of the test. Data was recorded for scans 7-9 only. Inspection of the transferred files with vsum and other VDIF tools revealed problems in decoding the data. Reason needs to be investigated. So far no valid data from Ys.
    - Mh had very bad weather during the whole test.

    - Efffesberg did parallel recording with RDBE system (Eb). VDIF data is stored on module MPI+1621. RDBE was configured for DDC mode (2nd Nyquist zone).

    Results


    Effelsberg zero-baseline correlation

    O-baseline correlation using zerocorr

    Some test data was taken using the t162a setting prior to the observations. Test data was named full_ef_no006 (DBBC full band mode) and full_eb_no0006 (RDBE 4x128 MHz). The channel settings are:

    $FREQ;
    *
    def 86140.00MHz4x128MHz;
    * mode =  1    stations =Eb
         sample_rate = 256.000 Ms/sec;  * (2bits/sample)
         chan_def = : 86140.00 MHz : U : 128.00 MHz : &CH01 : &BBC01 : &U1Cal; *Rcp
         chan_def = : 86140.00 MHz : U : 128.00 MHz : &CH02 : &BBC02 : &U1Cal; *Lcp
         chan_def = : 86268.00 MHz : U : 128.00 MHz : &CH03 : &BBC03 : &U2Cal; *Rcp
         chan_def = : 86268.00 MHz : U : 128.00 MHz : &CH04 : &BBC04 : &U2Cal; *Lcp
    enddef;
    *
    def 86524.00MHz2x512MHz;
    * mode =  1    stations =Ef:On:Ys:Mh
         sample_rate = 1024.000 Ms/sec;  * (2bits/sample)
         chan_def = : 86524.00 MHz : L : 512.00 MHz : &CH01 : &BBC01 : &NoCal; *Rcp
         chan_def = : 86524.00 MHz : L : 512.00 MHz : &CH02 : &BBC02 : &NoCal; *Lcp
    enddef;


    The data from the DBBC was recorded on the Mark6 combining both VSI data streams are combined and written as interleaved single threaded VDIF data with 8000 bytes packages. VDIF_8000-4096-2-2.

    The data from the RDBE was recorded on the Mark5C in multi-threaded VDIF data with 5032 bytes per frame. Below an example how to convert this with vmux to single threaed data VDIF_20000-1024-4-2.

    oper@eff-mark5c-1:~$ vdifbstate full_eb_no0006.m5a 5032 1024 0,1,2,3 500
    Executing: vmux full_eb_no0006.m5a 5032 6400 0,1,2,3 - 0 | m5bstate - VDIF_20000-1024-4-2 500
    Mark5 stream: 0x92960c0
      stream = File-1/1=<stdin>
      format = VDIF_20000-1024-4-2 = 3
      start mjd/sec = 57660 29686.000000000
      frame duration = 156250.00 ns
      framenum = 0
      sample rate = 128000000 Hz
      offset = 0
      framebytes = 20032 bytes
      datasize = 20000 bytes
      sample granularity = 1
      frame granularity = 1
      gframens = 156250
      payload offset = 32
      read position = 0
      data window size = 524288 bytes
    10000000 / 10000000 samples unpacked

    Ch    --      -     +     ++        --      -      +     ++     gfact
     0 1698169 3306628 3311096 1684107      17.0   33.1   33.1   16.8   1.05
     1 1694346 3311571 3311592 1682491      16.9   33.1   33.1   16.8   1.05
     2 1698207 3309868 3304658 1687267      17.0   33.1   33.0   16.9   1.05
     3 1693271 3305839 3319599 1681291      16.9   33.1   33.2   16.8   1.05

    The convereted single threaded data can then be correlated with zerocorr in zoom band mode. Below some plots and the zerocorr parameter files to process the data.

     

    DBBC_2x512MHz.png

    RDBE_4x128MHz.png

    DBBC_2x512MHz+RDBE.png

    Fig. 1 DBBC auto correlation Fig. 2 RDBE auto correlation Fig. 3 RDBE and DBBC autocorrelation combined


     

    RDBE-BBC01_DBBC2-512MHz-Mark6_RCP.png

    RDBE-BBC02_DBBC2-512MHz-Mark6_LCP.png

    RDBE-BBC03_DBBC2-512MHz-Mark6_RCP.png

    RDBE-BBC04_DBBC2-512MHz-Mark6_LCP.png

    Fig. 4 DBBC and RDBE BBCs zoom band cross correlations

     

    0-baseline correlation using DiFX

    The 0-baseline data was also processed with the MPIfR DiFX correlator yielding the same results as shown above.

    EfEb_LL.png EfEb_RR.png
    EfEb_LR.png EfEb_RL.png

     

    Correlation (On,Eb,Ef,Mh,Ys)

    No fringes found except on the zero-baseline between Eb and Ef (see above). The Ys data could not be decoded at all.

    See the memo by Dave Graham for details.

     

    Conclusions

    The fullband mode of the DBBC2 has been demonstrated to work properly. Also the required bitstream reordering done by the FiLA10G works as expected.

    Prior to the March/April 2017 run a working setup needs to be established at all the DBBC2 stations so further test runs are needed.

    Tag page (Edit tags)
    • No tags
    Pages that link here
    Page statistics
    1045 view(s), 16 edit(s) and 9263 character(s)

    Comments

    You must login to post a comment.

    Attach file

    Attachments

    FileSizeDateAttached by 
     DBBC_2x512MHz+RDBE.png
    combining both data
    46.22 kB14:36, 18 Nov 2016ubachActions
     DBBC_2x512MHz.png
    DBBC auto correlation
    30.2 kB14:36, 18 Nov 2016ubachActions
     EfEb_LL.png
    No description
    446.53 kB16:58, 18 Nov 2016rottmannActions
     EfEb_LR.png
    No description
    449.12 kB16:58, 18 Nov 2016rottmannActions
     EfEb_RL.png
    No description
    455.52 kB16:58, 18 Nov 2016rottmannActions
     EfEb_RR.png
    No description
    446.75 kB16:58, 18 Nov 2016rottmannActions
     RDBE-BBC01_DBBC2-512MHz-Mark6_RCP.png
    DBBC - RDBE BBC01 cross-correlation
    48.79 kB14:36, 18 Nov 2016ubachActions
     RDBE-BBC02_DBBC2-512MHz-Mark6_LCP.png
    DBBC - RDBE BBC02 cross-correlation
    43.09 kB14:36, 18 Nov 2016ubachActions
     RDBE-BBC03_DBBC2-512MHz-Mark6_RCP.png
    DBBC - RDBE BBC03 cross-correlation
    43.51 kB14:36, 18 Nov 2016ubachActions
     RDBE-BBC04_DBBC2-512MHz-Mark6_LCP.png
    DBBC - RDBE BBC04 cross-correlation
    41.69 kB14:36, 18 Nov 2016ubachActions
     RDBE_4x128MHz.png
    RDBE auto correlation
    25.64 kB14:36, 18 Nov 2016ubachActions
     t162a.pdf
    Memo by Dave Graham
    115.51 kB10:07, 23 Nov 2016rottmannActions
     t162aef.log
    No description
    127.56 kB11:43, 17 Nov 2016rottmannActions
     t162aef.prc
    No description
    1760 bytes11:44, 17 Nov 2016rottmannActions
     t162amh.log
    No description
    81.8 kB11:44, 17 Nov 2016rottmannActions
     t162amh.prc
    No description
    1730 bytes11:44, 17 Nov 2016rottmannActions
     t162aon.log
    No description
    43.29 kB11:44, 17 Nov 2016rottmannActions
     t162aon.prc
    No description
    1736 bytes11:44, 17 Nov 2016rottmannActions
     t162ays.log
    No description
    167.92 kB11:44, 17 Nov 2016rottmannActions
     t162ays.prc
    No description
    1706 bytes11:44, 17 Nov 2016rottmannActions
     zerocorr_zoom_mk6-BBC01.par
    zerocorr parameter file for RDBE BBC01
    146 bytes14:36, 18 Nov 2016ubachActions
     zerocorr_zoom_mk6-BBC02.par
    zerocorr parameter file for RDBE BBC02
    146 bytes14:36, 18 Nov 2016ubachActions
     zerocorr_zoom_mk6-BBC03.par
    zerocorr parameter file for RDBE BBC03
    146 bytes14:36, 18 Nov 2016ubachActions
     zerocorr_zoom_mk6-BBC04.par
    zerocorr parameter file for RDBE BBC04
    146 bytes14:36, 18 Nov 2016ubachActions