[g13] ST timelines are online

[Yordanka Ilieva]

Hello Chuck and Daria,

I finally solved the issue with the lack of paw memory and the problems 
this was causing me with generating the figures containing the timelines 
of the ST delta_T. The web site is updated with the latest results:

http://www.jlab.org/Hall-B/secure/g13/ST_CALIB/g13b/global_monitoring/st_moni/index.htm

The thumbnails on the web site show delta_T for pi+. Below the 
thumbnails is a colored figure showing the position of the main peak of 
the delta_T distributions as a function of paddle number (***).

At the top of the page, below the title, there is a link called "list of 
monitored runs and run files" which leads to a text file showing which 
run_partition corresponds to the file number shown on the x-axis of the 
timeline histograms.

You can see the timelines for all the particles for specific run and 
paddle, by clicking on the corresponding thumbnail. That opens a web 
site in which the axes on the figures are pretty much self explanatory. 
Each row of figures is for protons (top), pi_plus (middle), and pi_minus 
(bottom). The middle histogram shows the width of the delta_T 
distribution as a function of file number. The most right histogram in 
each row shows the one-dimensional projection of the most-left histogram 
onto the y-axis. (***) It is the position of these peaks that are 
plotted on the colored figure on the first page of the web site.

1. One sees that the means of the delta_T distributions  vary from run 
to run within a range of 150 ps. For most of the paddles the variations 
are within 100 ps or better. A notable exception is the  run range Nick 
has been investigating  (54730 - 54740). On the scale of +- 1 ns all 
looks pretty flat.
2. For a given particle type, the mean values of the delta_T 
distributions vary from paddle to paddle within a range of 100 ps.
3. The mean values of the delta_T distributions vary from particle to 
particle and paddle to paddle within a range of 200 ps.
4. The variations in 3 are on top of the variations in 1, so the global 
spread of the delta_T's is probably larger than 200 ps.
5. The first few thousand files do not have any ST data due to a bug in 
my scripts.

Is it worth to improve the  calibration for some subranges for:
-sector 1 paddle 2
-sector 3 paddle 3 and paddle 4
before pass2?

Best regards, Yordanka