Time for cal to rise last 5% of maximum


     If the cband cal was left off for many minutes and then turned on, it would take up to 5 seconds for the cal to go from 95% of the maximun value up to the maximum. To check all diodes on all the receivers the following test was done:
  1. Setup the receiver. Use the high uncorrelated cal. This sends diode 1 to pol A and diode 2 to pol B.
  2. Square law detect a 5 Mhz bandwidth with a 20 millisecond time constant. Sample both polarizations at 100 hz.
  3. Leave the diode off for a few minutes. Then toggle the diode off 5 secs, on 10 secs, off 5 secs, on 10 secs, off 10 secs.
For each polarization a 2nd order polynomial was fit to the cal off values and subtracted from the sampled data.  The top plot has the sample data  in a/d counts with the baseline overplotted in red. The bottom plot is the cal deflection as a fraction of the full scale cal. The red lines show where the cal was turned on and the maximum cal value. For each cal the ratio of cal to system temperature is also plotted (the dome was at 8.8 degrees).
The plots are: Only cband has this problem of slowly approaching the maximum. The second time that the cal was fired (5 seconds later) the problem went away. There must be some type of gain variation with temperature that is going on here.
processing: x101/010607/doit.pro