Karen O'Neil, Frank Ghigo

Set-up

• L-band
• 12.5MHz Bandwidth
  
• 32768 lags
  
• 2 Quad
  
• 4 samplers
  
• 3-level
  
• Linear Polarization
• Band centers offset by 5 MHZ (IF2=13075,1380)
• Observing using position-switching technique
• DCR was on simultaneously during observations (on accident)

Summary

• Pointing correction was less than 1'.
• No baseline ripples were seen in either bandpass
• The bandpass shapes and rms levels for the two IFs were virtually identical and equal to the theoretical prediction.
• The determined velocities and fluxes for UGC 5572 agreed, within errors, with those found by Rick Fisher in his Galaxy HI survey
• No significant RFI was seen.
• Ringing was seen in the first 200 channels of most scans.
• DCR did not appear to affect observations
• This mode is ready for general use.

Observations

• Data taken on 19 Feb, 2003
  
• Data file is T_LBAND_19FEB03
  

  Scan	Source	ON time	Comments
  119,120	UGC 5572	300s	RA offset=-87'
  121,122	UGC 5572	300s	RA offset=-87'

Analysis

I. Calibrator

No calibrator source was observed. A gain of 1.72 & 1.82 for pol. X & Y, respectively, was used (taken from 3C295 observations taken the same night in different modes).

II. Galaxy - UGC 5572

Due to spectrometer problems, only two 600s positioned-switched observations were made of U5572. The measured flux, peak, and velocity widths of UGC 5572 falls within 5% of the values published by Rick Fisher for both IFs and both polarizations. Example spectra are below:

Fig 1. IF1, average of all scans

Fig 2. U5572, IF2, all scans averaged, blown up on galaxy spectra

Fig 3. U5572, IF1, scan 120, blown up on galaxy spectra

Fig 4. U5572, IF2, scan 120

III. RMS

The rms noise for polarizations X and Y ranged from 0.068 - 0.083 for the scans. No discernable difference was seen between the two IFs. Assuming a Tsys of 21.7 K for both polarizations (taken from the average values) gives a theoretical value of 0.079 K, in good agreement with the measured values. The RMS of the averaged scans also followed the thoeretical 1/(sqrt(t) behavior as the scans were averaged.

A slight ripple was seen in the baseline of one scan.