In many cases it is necessary to measure continuum sources (with the 80-channel continuum backend) in order to compare these with the calibration of the line observations, done with the Effelsberg autocorrelators. For the (old) 1024-channel AK that normally creates no problems, because the IF (at 150 MHz) of the continuum-filters and the autocorrelator itself are the same. Moreover the switchable continuum passband-filters are comparable to those of that correlator. Therefore it is possible to use the results of the continuum observations with the 80-channel backend for calibrating line observations with that correlator.

The (new) 8192-channel correlator not only can be used with the narrow band IF ( at 150 MHz ), but also with the broad band VLBA-IF ( at 750 MHz ), which is the normal case. Unfortunately there exists no continuum input for the 80-channel continuum backend , neither for the VLBA-IF nor for similar bandwidthes as used in the correlator. Just for that reason it is nearly impossible to use observations with the 80-channel continuum backend for calibrating line observations, done with the 8192-channel AK and broad band IF.

Therefore we introduced a "new" continuum backend, which takes the eight total-power counters of the 8192-AK sampling them like continuum data. The correlator is used like an 8-channel continuum filterspektrometer, where each channel can independendly be set at another passband by the internal oscillators, as usual. All eight channels are always written to the raw data neglecting the selected splitmode.

The data can be reduced with the TOOLBOX program CONT2 like a normal continuum observation with 8 channels.

In the observer's interface for Effelsberg ( OBS_E ) the commands for the 8192-channel correlator have not changed. These commands see the OBS_E - Addendum) are also valid for continuum observations with that backend. In parallel some other commands specified for continuum will also be necessary in this case. All the continuum procedures like POINTING or FOCUS can be observed with the 8192-channel AK. The ONLINE-display automatically plots the data and does the relevant fits.

There are necessary only a few additive commands in OBS_E.

First, one has to specify the desired continuum backend.

Now here are two possibilities:

SET   BACKEND_C   <backend_id>

• <backend_id> =    D(igital backend)      for the 80-channel continuum backend
• <backend_id> =    F (ilters from AK)      for the 8192-channel AK in continuum mode

a similar command is already necessary for the selection of the desired line backend:

SET   BACKEND_L   <backend_id>

• <backend_id> =    A (k-1024)                for the 1024-channel AK
• <backend_id> =    N (ew 8192-AK)       for the 8192-channel AK

Second, the integration time for one data point must be specified.

Here we have one new parameter:

SET   NPREINTEG  < t1 >    <t2>

•  <t1>      =        dump_time for line observations ( in millisec )
•  <t2>      =        integration per data point for continuum observations ( in millisec )

>>>>>     Do not use dump times too short, otherwise the 8192-AK will loose dumps or hang up !     <<<<<

EXAMPLE   for command input:

        OBSE>    set backend_c   f

        OBSE>    set backend_l    newak

        OBSE>    set npreinteg   60000   640

The effect of these inputs in OBS_E are:

Take a dump time of 60 sec for line procedures like PSWITCH, observing with the 8192-AK. For continuum procedures like POINTING take an integration time per data point of 640 millisec, also observing with the 8192-AK. By Switching from line- to continuum-procedures the system always observes with the 8192-AK.

In order to display the different channels ONLINE, one can use the usual command for the pen-driver:

SET    PEN    <channel_nr>

The observing mode like T (otal Power) , necessary to construct the pen-formula for the ONLINE-display, is taken as defined in the relevant continuum procedure (for example: POINTING Total ).

Some Notes About OBS_INP:

The backend number of this new backend is 4, but the parameters are taken from backend 3 (which is the 8192-AK in line mode). The name of the backend which also appears in the reduction programs is AK8KCONT. The 80-channel continuum backend is backend 1 and its name is MOBA80CH, the 8192-channel AK in line mode is backend 3 with the name AKN8192K and the 1024-channel AK is backend 2 with the name AKS1024K.

Using OBSINP the decision , in which mode (line or continuum) the 8192-AK shall observe, is made by the splitmode of the 8192-AK. The allowed splitmode + 1000 is interpreted as continuum mode, less than 1000 as line mode.

Effelsberg Home