Continuum Backend

    The 80-channel continuum backend is the standard backend for all continuum observations. It accepts TTL linedriver inputs ( up to 20 MHz ). The communication with the control computer happens via IEC-Bus, the data are delivered over a 16-bit parallel DMA connection (the so called ``Effelsberg Data Bus''). Internally the backend is able to integrate data up to 32-bit resolution. Analog outputs programmable from the control computer are available. These are used to display data on on-line pendrivers. Most of the 80 channels are connected to multi-channnel receivers in the secondary focus in a fixed manner. The first 8 channels normally deliver the signals of the active primary focus receiver or the receivers of the secondary focus, which are not connected to the backend in the standard mode.

Analog Chart Recorder Output

   The following commands serve to define and adjust the outputs of the analog outputs (i.e. the on-line pendrivers). At the moment two different outputs defined by the control computer are possible. These two definitions are also used by the on-line display task, combined with the TOOLBOX programs, to calculate the signal- and calibration- values of its graphical output. This output is generally used for the gaussfits correcting the pointing of the telescope. So, if one intends to use that option, the first output-channel has to be defined as signal and the second one as calibration of that signal. There are also some commands to scale the analog outputs to the range of the pendrivers.

AS1 < string > #

 

AS2 < string > #

With these parameters (Analog Signal 1 and 2) one defines the formula for the chart recorder outputs. Normally AS1 should be the signal and AS2 the calibration. It is defined by a sum or difference of phases according to the phase-cycling of the frontend. One can mix one or more channels. The length of < string > must be less then 80 characters. A phase-value is characterized by its channel- and its phase-number in the form:

signX(channel-nr.,phase-nr.)

sign defines the sign of the value, X is the array-name. EXAMPLE:
Assumig that we have 4 phases and the calibration is switched on in phase number 3 and 4.
AS1= X(1,1)+X(1,2)+X(1,3)+X(1,4)
AS2= -X(1,1)-X(1,2)+X(1,3)+X(1,4)

AS1 represents the signal of channel 1 and AS2 the calibration of channel 1.

In the following commands i stands for i=1 oder 2, assigning the number of the pendriver for that command.

VSi < v > #

  One sets the gain for the output of the analog channel with this command. That is the factor with which the output value is multiplied. < v > can be a positive or negative floating point number.

OSi < o > #

  This parameter sets an absolute additive offset for the output value of the analog channel. < o > can be a positive or negative floating point number. It can be used to adjust the output to the scaled window of the pendriver.

ASI < cycles > #

  This parameter sets the number of cycles < cycles > to be integrated, before a new analog value will be calculated and updated. For that time the analog output value remains the same. This command applies to all analog outputs.

Thus the update-time for the analog output value is:

t_update = cycles*n_phases*t_phase

with the following definitions (also used in the text below):
		  nphases  number of defined phases ( GNPi )
		 

 nchannels  number of defined channels ( by BE80 )
		  tphase  duration of one phase ( GPTi )
		  tblank  duration of blanking ( GBTi )    

Data Output

 

The following commands determine the data rate, the number of channels and the rate of the DAP's (i.e. Data Associated Parameters like telescope position or time information) in connection with the delivered data points. Thus the data, which are archived to disk, are directly influenced by these parameters.One should therefore be very careful when changing the default values.

INT1 < cycles > #

  This parameter sets the number of phase-cycles < cycles > , which are internally integrated in the backend for one complete dataset to be delivered to the data-taking computer. The integration time for one dataset is (with the definitions used above):
t_int = cycles*n_phases*(t_phase - t_blank)
and for a single data point:
t_point = cycles*(t_phase - t_blank)

DPD1 < ndaps > #

  This parameter (Data Per Daps) determines , how many data points < ndaps > (with the integration-resolution defined by INT1) are packed together with the DAP-information. By this parameter also the transfer characteristics (time and number of bytes) from the backend to the data-taking computer are defined.

The transfer time (i.e. the time between two consecutive transfers of data from the backend to the computer) is:
t_trans = ndaps*cycles*n_phases*t_phase

The data points transferred in such a data-transfer are:
n_data = ndaps*n_phases*n_channels
and in bytes:
n_bytes = n_data*4 + bytes_daps
bytes_daps are the number of bytes used for the DAPs. They are in the order of 80 - 100 bytes.

EXAMPLE:
Assuming a typical continuum observation with a secondary-focus receiver of 16 channels, a phase time of 16 msecs, a blanking time of 0.4 msecs and a four-phase cycle one obtains with the following settings:

		 INT1  2 #
		 DPD1  5 #
		 Integration-time/data point: 		 

 tpoint = 31.2  [ms]
		 Transfer-time: 		  

 ttrans = 640  [ms]
		 data/transfer: 		  

 nbytes = $\sim$ 1.36  [KByte]
		 data rate: 		  

 bdata = $\sim$ 2.125  [KByte/s] 

IMPORTANT:
The transfer-time is also the time resolution of the data associated parameters (DAPs) transferred with the data.

BE80 < string > #

  This parameter serves to define the channels, which shall be written to disk and archived. As above mentioned, the different receivers are connected in a standard way to the backend, some are permanently connected others only if they are asked for by the observer.

< string > can take the following values (n,m are numbers in the range 1 ... 80):

		 CLEAR 		 no channel selected
		 ALL 		 all channels selected
		 n 		 select channel n
		 -n 		 de-select channel n, if it has been selected before
		 n...m 		 select all channels from n to m
		 -n...m 		  de-select all channels from n to m
		 		 if they have been selected before
		 n/m 		 same as n...m
		 -n/m 		 same as -n...m

Several channel-commands may be written on one line. They must be separated by comma (,), space ( ) or a semicolon (;). Leading spaces and separators are ignored.

EXAMPLE:
BE80 ALL, -15...36; -8 -11,,,18/24
in total one has selected with this command:
channels:        1-7,9,10,12,13,14,18-24,37-80

NOTE: To be sure selecting the right channels independent of inputs done before, it is recommended first to enter the command ``BE80 CLEAR'' and then to enter the channels to be written to disk.

IMPORTANT:
The above selected channels are written to tape in ascending channel-nr. independent in which order they have been selected by the input command BE80. That means: The smallest channel-number selected is always the first one in the archived data and the highest channel-number selected always the last one. The channels in between are stored in the same way in ascending order.