26dec05 A1312 no power polB for 1 file

        On 26dec05 project a1312 took two files of OH data with lbw, circular polarization, and the 1550-1820 Mhz filter in.  The first file had no power in the correlator polB data. The second file was ok.
    The plot shows the status of the data during the first file  (pdf) (corfile.26dec05.a1312.1):

• Top: Lag 0 total power versus time. Black is polA, red is polB. The vertical scale is linear in power. The value 1 has the measured power equal to the optimum power. PolB (red) has no power.
• 2nd:  The correlator attenuator values versus time. The polB attenuators where at the minimum value (0) since the input power was low.
• 3rd: The power measured at the input to the upstairs fiber optic transmitters. PolA and polB are similar. The value should be close to -40 dbm (which it was).
• 4th: The power at the top of the downstairs iflo (after the fiber optic receiver and the variable amps. This value was only read once (at the very start of the run). Both values were reasonable. We don't know whether this changed during the run.

Conclusions:

• The dewar/first lo is probably ok since the upstairs power meter read ok for both pols. It is possible that the incorrect filter was inserted in polB. The if1 power meter reads about 500 Mhz bandwidth so it is possible that it read ok and then the correlator 3 Mhz bandpass was out of band. This is doubtful since the upstairs attenuation for polA and polB were the same.
• The correlator power counters (50 MHz bandwidth) also had no power. The power counters are analog and they are before the correlator digitizers. So this is not a problem with the half band filters and probably not the correlator.
• The downstairs iflo power meter was only read once at the very start of the run (when the command adjpwrif2 was executed). At the beginning the input power to the downstairs if/lo was ok.  This could have changed later in the observation.
• It was probably not a problem with the downstairs synthesizers since each synth supplies power to both pol A and polB simultaneously.
• My guess is that one of the downstairs switches on the polB side did not setup correctly causing no power to pass to the correlator. When the later observation was done, the iflo was resetup and things worked correctly (although this is only a guess).

 home_~phil