[evlatests] L-Band sensitivity, and Tsys

[Rick Perley]

    An obvious question concerning our apparent strong loss of EVLA 
L-band sensitivity with increasing frequency is whether this can be 
attributed to an increasing system temperature.  To shed light on this, 
I have recovered the backend Tsys values from the test run taken 
yesterday (25 July).  These values must be adjusted, as the VLA Tsys 
machinery is not aware of the changing values of Tcal as a function of 
frequency -- it uses the values at the default frequencies (1465 and 
1385 MHz, for IFs AC and  BD, respectively) for all frequencies.  The 
corrections necessary can be significant -- as large as 30%, but more 
usually near 10%. 

    Attached (and/or below) is yesterday's table of AIPS weights, 
followed by a table of adjusted backend Tsys values. 
Antenna 18 did not have its Tsys machinery turned on yesterday (an 
oversight -- it does now!), so there are no values for that antenna. 
    If the loss of sensitivity is due solely to increase in Tsys, then a 
drop in AIPS weight by a factor of 4 will be accompanied by an increase 
in Tsys by a factor of 2.  No such relationship is seen.  For the most 
part (and especially for antennas 14 and 16), there is almost no change 
in Tsys between 1275 and 1750 MHz -- the range over which the antenna 
show a decrease in AIPS weight by a factor of about two. 

    Antenna 13 is a bit different -- there is no trend in Tsys with 
frequency to match the dramatic loss of apparent sensitivity, but I note 
that the Tsys values for the BD channels is quite a lot higher than for 
the AC channels at the lower frequencies.  This should not happen as A 
and B are at adjacent frequencies, passing through the same amplifier, 
as are C and D. 


-------------- next part --------------
An embedded and charset-unspecified text was scrubbed...
Name: BGCFix.dat
URL: <http://listmgr.nrao.edu/pipermail/evlatests/attachments/20060726/5a4ae429/attachment.ksh>