Other pages related to fiber ripple:

14,19oct11 ripple

• Average each file of 10 minutes. The 7 Mhz band was used.
  
• compute (fft(spcAvg)^2 for each averaged spectra (giving the acf^2). The round trip delay in the fiber is about 6 usecs.
• Overplot each beams acfs.
  
• plot with offsets the spectra for  1 beam
• Compute the strength of the reflected signal at the fiber delay and then plot this vs time to see how much the ripple changes in amplitude.
  

• 14oct11: galfa data plotted
• 33 files plotted
  
• 19oct1l : daytime, upstairs, downstairs fibers cleaned.
• 19oct11: evening : galfa data plotted.
• 34 files plotted.
  

• Page 1: ripple amplitude (Tsys) vs delay (usecs). 6 usecs is the delay from the fiber cable.
• Each frame is a different alfa beam. The ripple amplitude at each beam is overplotted
  
• black is pola, red is polB
• The power at short delays  is the bandpass shape
• The fiber ripple is showing up at  6use.  (1/6usec)= 167 KHz period for the ripple
• Page 2: The average spectra for beam1a overplotted (with offsets).
• 14oct11: bm1a : has strong ripple
• 19oct11: bm1a: no ripple seen in 1a display.
• Page 3: The amplitude of the fiber ripple versus hour of day (ast)
• The ripple peak near 6 usecs (channel 40) was found for each file, beam, pol
• Top: polA ripple peak vs hour of day
• Bottom: polB ripple amplitude vs hour of day

Summary:

• Table of peak ripple strength:

• Peak fiber ripple amplitude (% Tsys) by date

  date	bm0		bm1		bm2		bm3		bm4		bm5
  	polA	polB	polA	polB	polA	polB	polA	polB	polA	polB	polA	polB
  14oct11	.05	.03	.2	.07	.12	.03	.12	.14	.2	.02	.03	.02
  19oct11	.02	.03	.02	.05	.1	.02	.13	.15	.2	.02	.02	.02

  
  
• bm1a got better after the cleaning
• bm2a,3a,3b,4a were unchanged.
  
• bm4a variation with time for 14 and 19oct11 look very similar. This may be a temperature related problem.
  
• back on 23apr11bm1a and bm4a were also the largest ripples.
• during 14oct11the azimuth remained fixed while the dome changed in za from 2 to 19 every 12 minutes.
• The constant ripple in beam1a,4a from 20.5 to 21.5 hrs occurred when the dome was moving up and down (19 to 2 degza)  two complete times so. So the ripple is not being caused by the fiber cables changing in the wrap, cable drap.
  

23apr11:

• Compute the average band pass for each file (spcAvg)
  
• compute (fft(spcAvg)^2 for each spectra (or acf^2) to isolate the ripples in the bandpass.

• Each frame is a different alfa beam
• black is pola, red is polB
• The vertical scale is Tsys.
• The power at short delays  is the bandpass shape
• The fiber ripple is showing up at  6usec (1/6usec)= 167 KHz period for the ripple
• Pixel 4a is the strongest

• Page 1: Beam4A overplot the average spectra for the 34 files.
• Each spectra is averaged over 1 file (10 minutes)
• The bottom is the first file, to top is the last file
• An offset has been added for display purposes.
• The vertical scale is Tsys.
• Page 2: The amplitude of the fiber ripple versus hour of day (ast)
• The ripple peak near 6 usecs (channel 40) was found for each file, beam, pol
• top: polA ripple peak vs hour of day
• Bottom: polB ripple amplitude vs hour of day
• beam4a ripple starts around .003 Tsys. In 3 hours it has decreased to .005 Tsys.

Summary:

• The ripple from the fiber is present again.
• strongest in beam4a (as before).
• The peak is now in channel 40 (166KHz period). Back in 2005 it was closer to chan 38 (176 KHz period).
• There is a strong time variation of the ripple.
• for beam4a in goes from .0028 Tsy to  .0005 Tsys in about 3 hours.