For the filterbanks (FBS) to work, you first have to set the computer up.
This will connect FBS to Frontend 1 (whatever you defined with the receiver command. See section6 for details.
For the computer actually do these changes, it will have to be given a command like ``go'' or ``pswitch''.
For more details on the IF interface, see section 5.5.1.
Unlike the AOSs, there is no easy way to look at the filterbank display when not actually taking data. If you are taking data, then ONLINE CHEF presents the subscan results as the scan progresses.
If you go into the computer room (i.e. the AOS room), then on the Filterbank rack (beside the door to the left), you can find the total power meter of the filterbanks. If, when a hot load is inserted, the total power level is between 500 and 750 millivolts, then all is well. If below 500, then you will have decrease the attenuation, and if above 750, you will have to increase the attenuation.
The current attenuation is displayed beside the filterbank level - usually a number between 3 and 7 (except for the SIS490 or HEBOL where it will be much higher due to the very high system tempatures!)
If you need to change the attenuation:
where number is the new filterbank attenuation.
You can also look at the number of counts (which are displayed in the CAL scans. If the number of counts for the hot load approaches 800,000 or more then you will have to increase the filterbank attenuation. If it gets below 100,000 or so for the hot load, then you will want to decrease the attenuation.
The individual channels of the filterbanks sometimes need adjustment. If you notice a channel (or more than one) is bad, do a cold calibration scan (so that the channels will be displayed by ONLINE CHEF). Then use ``mbc ha/pl'' in OBST to print out the calibration scan.
Give a copy to the FoT so that he can inform the appropiate staff member for action. If the channel is in an very critical spot in your spectra, we will try and fix pronto. Otherwise, it will be as soon as possible. Note that in most cases, you can shift your spectra changing the source velocity by a couple of km/sec and move the line away from any bad channels.