PolB compression 03jun11

    A2335 (gass) reported horizontal strips in their on/off images on 03jun11. These stripes were mainly seen in polB. Their setup is:

• lbw linear pol with 1320 hipass filter installed.
  
• interim cor 25 MHz bw 1024 channels with 1 sec integrations.
• 300 second on,off processing.

10jun11: resolution:

• the radar blanker power plug was faulty. When this was fixed, the rfi that a2335 reported went away (when the radar blanker was used).
• I'm not sure how the radar blanker could have caused the long recovery time so we may still have a problem?
  

The first image shows a dynamic spectra for the 300 second on position (scan 115400375) (.gif)

• This is polB for the 2nd sbc.
• You can see the horizontal lines reported by the observer.
  

The 2nd set of plots show the total power vs time for the first 3 onoff patterns (.ps) (.pdf)

• Page 1: total power vs time for 3 on,off pairs
• top PolA, bottom PolB
• The 3 different subbands (sbc) use black,red, green (with offsets for display)
• The gap between the on and offs is the 1 minute used to slew between the two  positions.
• Wall time is shown. The 300 1 second records for the on or the off take longer than 300 seconds since radar blanking was active.
• The negative going spikes in polB are caused by compression.
• The negative going spikes stopped during the off position of the 3rd on,off pattern.
  
• Page 2: blowup of total power vs time.
• 100 seconds of wall time is displayed.
• Horizontal dashed lines are aligned with the first negative going spike and then spaced by 12 seconds of wall time.
• Pol B is taking many seconds to get back up to the uncompressed power level
• Page 3: az, za position vs time for the first 100 seconds of observation.
• This shows the az,za area covered during the compression.
  

The total power vs time for a2335 (mar11 thru 03jun11) was also plotted so you could see if this occurred before (not obvious).

SUMMARY: (top)

• lbw polB is compressing with a 12 second period.
  
• The compression is 1 to 2% in a 1 second integration
• This is most likely a radar (since they have 12 second rotation periods).
  
• The hilltop monitoring system was not running during this observation so we can't say which radar was broadcasting  (later in the evening a2333 (auds) saw punta salinas, remy radar (1270,1290) and FAA (1330,1350) active.
• Since the radar blanker was being used, you might assume that it was not the FAA radar causing the problem.
• The long compression recoveries may make it harder to say this.
  
• The 12 second compression went away after about 30 minutes.
  
• The strength of the rfi in the receiver is a function of the az,za position of the feed relative to the radar and platform. It  is not unusual for the strength to peak up as certain az,za positions (see az dependence of alfa compression).
• PolB is taking up to 4 seconds to recover from the compression.
  
• This is not normal and needs more investigation.
  

processing: usr/a2335/polBCompress_03jun11.pro