====== PDFB4 as Spectrometer ====== The DFB can be operated by logging in to the DFB control computer and it's IP address is 134.104.64.137. Log in as user - corr and password - h1pass.no You need to use ssh -Y/-X to get X11 forwarded. Once you are logged in, open two xterms and run these two commands the two xterms: ''sdfb4'' and ''tkds''. These interfaces are described briefly below. __NOTE__ : Start sdfb4 first and then tkds. ====== Setting Clock on the DFB ====== * Log on to computer (e.g. pulsar@psrfb0 or observer3) running ntpd (check with ps -aef | grep ntp) * Either run ''top'', then type ''s'' and ''0.5'', or run ''xclock -digital -update 1'' on the above computer * on the DFB, run atdc, choose options in the following order (1 for TCP/IP socket , locahost , 2 (clock control) , port 4010 , password (atdc), 2 (set time only) * Enter time in future as ''16 apr 2012 09 40 30'' * Once the time in step 2 reaches the above time, hit return within one second * check if the set time is correct by displaying time on the DFB (option 9) ====== SDFB4 ====== The ''sdfb4'' window is the main control of the DFB. Here are some points to note: - The button under **CYCLE** is the integration time. - The two split panels in the lower part of the window displays status messages. __BLUE__ messages are bad. You might have to restart ''tkds'' and ''sdfb4''. - Press **QUIT** buttons on both windows to exit - run the command ''corkill'' to clear off the memory and any exisiting correlator/spectrometer processes. - restart the processes as above. ====== TkDS ====== This is the user interface that talks to the DFB control program. From TkDS, you can choose the firmware to load, specify file name for the observation, and set integration time. Eventually, we can use a program that can replace TkDS and translates the telescope information to control sdfb4. Here are some steps that will be needed: - Choose a spectrometer configuration from the scrollable window on the top right and then press **CONFIG** button. The spectrometer config strings start with "sdfb4_". For instance, ''sdfb4_tb16_256_512'' works fine. To store (and display) the entire 1 GHz band, use ''sdfb4_1024_8192''. We are currently testing mode ''sdfb4_tb16_1024_2048'', which records 1024 MHz of bandwidth using 2048 channels, with 16 sub-integrations per cycle. After this point a window pops up and displays the prgramming status. Go to next step once the window disappears. - Next click on **ObsTypes** button, and choose //Normal// from the list. - Check that the **samplers** have min/max values of around -25/+25 (+/- a few), that the average is around 0 (can differ by a few), and that the rms is about 6 (+/- a few). You can adjust the attenuator levels by flicking the S13-S17 and S3/4 (ch. B/A) switches in the 50cm 'cabinet' behind the glass door (to the left of the main control room). - On the lower right corner, there is a small window named **Command**. You can type in the commands to open/close files and set integration times. For example, here are the commands that you need to use: * cycle 4.0, 0.01, 0.04, 0.04 * fo TestFile.fits ; leaving out the name results in a file yyyy-dd-mm_time.rpf - Now click the **GO** button - Wait until the observation is done. Now press **STOP** button. The DFB takes up to 30 secs to stop. Once you see that the DFB state changes to **STOPPED** in the sdfb4 window, type in "fc" in the command window of TkDS as in 3rd step. - The just written data in TestFile.fits is found under the directory ''/DATA1/SDFB4_1'' ====== SPD ====== SPD lets you monitor the output of the DFB3 in real time. Log onto the DFB control computer (see top paragraph). Then: - type in 'spd' - sel aa,bb ; select both polarizations - chan f1 5 500 ; selects the channel range - scale a 0 10 ; shows only amplitudes between 0 and 10 Jy Alternatively: - scale log ; to get a logarithmic scale - scale lin ; reverts back to a linear scale To switch time averaging on/off: use avg/noavg More information on SPD can be found at http://www.narrabri.atnf.csiro.au/observing/users_guide/html/new_atug_44.html#spd ====== Observing ====== Remember to check the focus at the start of the observations. Run a pointing. Then you can start observing your target source (on/off pointing, or tracking). In ObsInp: * **Change focus**: "Focus offset" (almost at the bottom of the l. column). Use 'Zlinear' to specify the relative change in the focus position in mm. * **Change frequency**: Frontends > Primary (50 cm receiver is in the primary focus) or Frontends > Secondary (6 cm is in the secondary focus). The frequency must be in GHz, switch 'Version' to 'Continuum/Line' * **Pointing scan**: 'Setup cont' > 'Pointing'. The following values have worked well in the past: Scantime = 36 seconds, Scanlength = -4 (this sets the distance between the target and the secondary position. A positive number indicates a number of degrees, a negative number indicates a number of beams), Scanrepeats = 2. PointCorr = 1 means that the pointing correction is applied, a negative number (or zero?) that the correction is only calculated, but is not applied. Run a pointing scan every (couple of) hours, and at least at the start of the observations. * **Position switching**: in "Setup Spec" select "PSwitch". Select "Vmode = Fixed" and "ScanRepeats = 1". * **Tracking**: in "Setup Cont" select "Tracking". * Hit the start button on TKDS once the correlator shows the message "Measure received" (just after a message about the BEACON) * See [[:50cmproject|50 cm Polarisation Project]] for tools that help you plan the observations ====== Switching between 50 cm and 6 cm receivers ====== * The signal path in the MultiFiBa needs to be changed to 192 (50 cm receiver) or 161 (VLBI). Type in s99161z / s99192 (capital 'S'??) + check 'Choose MultiFiBa mode' in the 'XFFTS leveling' panel (this is shown on the leftmost monitor, below the channel 1/2 histograms) * Backends: XFFTS select '3000 MHz' IF, and '2000 MHz bandwidth'. Load this to the observing program. * Primary/Secondary (left column): select the right receiver * Set the right frequency (Frontends > Frequency), select 500 MHz bandwidth (50 cm receiver, not 6 cm receiver?) * set focus to +0.9mm, do a pointing, observe, ... ===== Troubleshooting ===== * ** No signal**: Check that the MultiFiBa uses the correct signal path. In ObsInp, go to "MultiFiBa", S08 and S16 should have "192" in their final column. To change the current setting to 192, use S08192 , or S16192 . * **Bad pointing scan**: check the attenuation in the XFFTS, and check that you're using the correct focus position. The HPBW displayed in ALON/ALAT should typically between 1750"-1950" at 472 MHz. * **XFFTS only show 2 bins with signal in each channel** or **All lights in the attenuation control panel are off**: send the position of the focus again to the receiver system: Frontends > Primary for the 50cm receiver and Frontends > Secondary for the 6cm receiver. Then specify the observing frequency in GHz, and switch 'Version' to 'Continuum/Line' ===== Various items ===== * Accessible frequencies: 300 - 505 MHz |filter| 608 - 685 MHz |filter| 790 - 920 MHz; see also [[http://www3.mpifr-bonn.mpg.de/div/electronic/content/receivers/50cm.html|technical information on the 50 cm receiver (Effelsberg website)]] which includes also block diagrams. * Focus positions: -10/-80/-140/-180 mm at 308/472/653/852 MHz * Set attenuation: First toggle the H/R switch from "R" to "H" (H=hand; R=remote). Switches S3/S4 select channel B/A (note: reversed!), S13-S17 add 1-2-4-8-16 dB of attenuation. Don't forget to flip the H/R switch back to "R" when you're done. * Manual switches on the 50cm attenuation panel: S7 should be blinking when a cal signal is used (e.g. for pulsar observations). S10 switches the LNA on. S11 and S20 must be lit up. More information can be found in the 50cm manual (blue book in the Effelsberg control room), section 4.5, p17/26. * For the 50cm system you should check the counts in a panel named '2F-Polarimeter' in the '50cm cabinet' (that also houses the panel with attenuation switches). Standing in the same room, there are 2 further counters in the '2F-Umsetzer' panel on your left hand side (bottom half of one of the cabinets there) that you should check. These counters saturate at 2000 counts, when they jump to -1 (50 cm receiver; for the VLBA receiver they saturate between 5000-5500 counts). Don't forget to have a look at the spectral display on the oscilloscope, to see that the RFI peaks are not saturated; this shows up as all peaks having the same max. level. The 2F Polarimeter typically samples a 2 MHz band, while the 2F Umsetzer samples a 100 MHz band; both bands are centred on the frequency that the observer specifies in the control room using 'Frontends > Primary' or 'Frontends > Secondary'. If you want to check the bandwidth used by the 2F-Polarimeter: look at the panel '3 Kanal Filtereinheit' that is just below the '2F Polarimeter'. If the 2F Polarimeter uses a bandwidth of 2 MHz then the '2 MHz' light should be lit up. * To display the full band on the XFFTS window: go to ObsInp > XFFTS, then select "3000 MHz" and "2000 MHz" - the rightmost buttons. * Pointings use the same band as the '2F-Polarimeter' (see the previous bullet point), which is typically 2 MHz. You can use the XFFTS to select a 3-4 MHz wide band that is free of RFI before you can do a good pointing. An RFI peak of up to 0.5 dB can still give a usable pointing solution (it might need to be stable in flux?) * The strong (up to 50 dB!) line at 390 MHz is caused by the 'Tetra' police communications band, and the GSM 900 system is producing the peaks >~ 900 MHz. ===== Future observations ===== * Peter Mueller warned that the bandpass at 50cm might vary during the observations; we need to test this. At 6 cm the bandpass is stable. * The 50 cm receiver can provide pulse profiles for LOFAR; the LOFAR people can then test their calibration strategy. Also, the RFI generated by LOFAR can be measured with the 50 cm system (we could not conduct that test back in January) Both are excellent reasons to liase with the LOFAR people, and work out an observing plan with them.