Meeting No.8: BigBite Optics #1

I have started working on the BigBite optics. I have started by investigating three different approaches. Two analytical approaches written by Jin Huang and myself and a standard transport matrix approach written by Jin Huang. When I looked closed to a BigBite library code a realized that so far we have been using the analytical approach. The matrix approach has been coded but not used.

When I changed my analysis script so that it uses the matrix code, I got strange errors, when trying to read the "DB_BB.optics.dat" file. I found out that the problem was that a method, that is used to read the database was defined in the header file but not implemented. Since the THaOpticsHRS.cxx file is an exact copy of the regular code used for the optics of the HRS spectrometers, I copied the missing code from the THaVDC.C file. Things work fine now.

I have started the determination of the BigBite transport matrix with the following idea: I want to determine the first order terms of the BigBite matrix. Since BigBite consists of the single dipole magnet I assumed that dispersive variables are not coupled with the non-dispersive ones. I have tried with the following parametrization:

    delta = (delta|x)*x + (delta|th)*th
    th    = (theta|x)*x + (theta|th)*th
    y     = (y|y)*y     + (y|ph)*ph
    ph    = (ph|y)*y    + (ph|ph)*ph

In order to determine the matrix elements I decided to use an elastic run #2814. Since in this run q-vector corresponds to momentum of the protons in the bigbite I was able to use it to determine the correlation between the target coordinates and the measured focal plane variables, assuming that HRS-L works perfectly. In order to determine the matrix elements I took 1000 random events and used chi-square minimization to determine the matrix elements. Unfortunately the minimization did not return results that I was hoping for. Therefore I decided to make a step back and first look at some "general" plots that should give me a rough idea on how target variables depend on focal plane variables. For these plots I decided to use hydrogen sive-slit runs #3487 and #3488.

I have also tried to get some reasonable results with my analytical approach but these results are also not very promising at the moment. I will have to dedicate a bit more time to determine the correct distances between the various parts of the spectrometer that are required by my code. With my current numbers thing is not working very well.

Here are some of my plots:

The cuts that I have used in my analysis are:

    if (fabs(L_gold_dp)>0.045  )  continue;
    if (fabs(L_gold_th)>0.06  )  continue;    
    if (fabs(L_gold_ph)>0.03  )  continue; 
    if (((int)DL_evtypebits&32)!=32) continue; 
    if (((int)BB_tr_n)!=1) continue;

1.)

2.)

3.)

4.)

5.)

6.)

7.)

8.)

9.)

10.)

Plots with different cuts:

The cuts that I have used in my analysis are:

    if (fabs(L_gold_dp)>0.045  )  continue;
    if (fabs(L_gold_th)>0.06  )  continue;    
    if (fabs(L_gold_ph)>0.03  )  continue; 
    if (((int)DL_evtypebits&32)!=32) continue; 
    if (((int)BB_tr_n)!=1) continue;
    if (fabs(L_gold_y)>0.02  )  continue;

10.)

11.)

12.)

13.)

14.)

15.)

16.)

17.)

18.)

19.)

Last Modified: 11/01/09