• I took all the lbw calibration runs jan14 to apr16.
• Measurements were used if they met the following criteria;
• za < 14 deg (so tsys vs za wouldn't affect the sefd)
• srcflux > .5 Jy..
• 25 K < tsysfit < 50k .. to get rid of lousy fits
• The dataset consisted of:

• freq MHz	number of measurements	sefd (Jy)		rmsJy	median/mean
  		median	mean
  1155	67	3.91	3.93	.22	1.0
  1175	138	3.62	3.61	.18	1.0
  1300	136	3.07	3.05	.15	1.01
  1310	71	3.10	3.13	.17	0.99
  1375	207	2.95	2.95	.17	1.00
  1415	373	2.80	2.82	.17	.99
  1550	164	2.73	2.76	.18	.99
  1610	164	2.60	2.67	.18	.97
  1666	166	2.57	2.60	.18	.99

The plots show the sefd vs frequency for lbw (.ps)   (.pdf)

Comments;

• I included the mean and median (to check for outliers..)
• the spread at each frequency was .17 to .22 Jy
• the jump at 1300 , 1310 is well within this error.
• The higher Tsys at lower frequency mainly comes from the increase in Tsys at lower frequencies..
  

processing: x101/160422/sefdvsfreq.pro

 home_~phil