[G8b_run] polarization table question

[Ken Livingston]

Hi Brian,
The directories are:
Beam_1.3_Manual_Pass_1
Beam_1.5_Manual_Pass_1
Beam_1.7_Auto_Pass_1
Beam_1.7_Manual_Pass_1
Beam_1.9_Auto_Pass_1
Beam_2.1_Manual_Pass_1

These reflect the data sets we're using in the final analysis. The 
initial plan was to flip the polarization every ~20min in an automated 
way, as had been done in labs like Mainz. Hence, we started out like 
that. We eventually abandoned it because the position and repeatability 
of the peak was not reliable. We're not using the smaller, unreliable 
amounts of autoflip data from 1.5 and 1.3 settings, hence there are no 
pol tables for these.

The shape and range of positions of the coherent peak is, in general, 
different for PARA and PERP - irresepctive of whether it was changed 
manually or automatically.  Hence there are separate PARA and PERP 
tables for each setting.

http://nuclear.gla.ac.uk/~kl/g8b/pol_progress2.html

As you said, in the TAGR bank, and probably in some ntuples too, the 
numbering goes from 1-767.
Here's something pasted from an old email on that.

#There are 384 Ecounters, which overlap to give 767 E_Ids in the TAGR bank
#If we have the E_Id energies (ie bin centres) the best way to get the
#Ecounter bin centres
#is just to take the odd numbered E_Id values.
#
#Ecounters (Ec=384)
#   1   2   3   4   5   6   7
#   |   |   |   |   |    ------
#   |   |   |   |   |------
#   |   |   |   |------
#   |   |   |------ |
#   |   |------ |   |
#   |------ |   |   |
#------ |   |   |   |
#   1 2 3 4 5 6 7 8 9 .....
#E_Id (= 2Ec-1 = 767)


I use this to map 1-767 to 1-384

E_id = (int)((TAGR[bestPhoton].E_id+1.0)/2.0)

Cheers,
Ken




On 12/16/2011 04:52 AM, Brian Vernarsky wrote:
> Hi all,
>
> Some of you know by now that I have finally worked past my issues with
> the Monte Carlo generation, which means that I am finally able to move
> forward with the analysis of the data.  So I have tried to start using
> the polarization tables to get the polarization for each event.  I
> have downloaded them from:
>
> http://nuclear.gla.ac.uk/~kl/g8b/newpoltables/g8poltablesv2.tgz
>
> and have modified the code given at:
>
> http://nuclear.gla.ac.uk/~kl/g8b/polTableCode.C
>
> so that I can use it in a c++ environment.  All that has gone well,
> but I have two issues.  The first is that the tables that I extracted
> from that .tgz file did not include the autoflip information for the
> 1.3 or 1.5 settings, all the others were there.  Also I'm a bit
> confused by the titles of the files in the autoflip directories, as
> they still have PARA and PERP in their titles despite the fact that
> the plane settings in the edge files for those runs flip from 0 to 1
> and so on in any given run.  Are the two states just separated out
> into these separate files?  If so then that's fine, otherwise could
> you tell me what the names refer to in this situation.
>
> The second issue is that the Tagger Paddle ID used in the polarization
> tables only runs from 1-385, whereas the convention that I have for
> the tagger paddles is from 1-767.  I realized that the 767 number
> takes into account the overlap between adjacent paddles and that 385
> is roughly half of 767, I just want to make sure I have the convention
> for converting between the two numbers correct.  Is it safe to just do
> my eid divided by 2, such that 1 and 2 in my convention would both go
> to 1 in the other convention, and say 700 and 701 would map to 350 and
> 351 respectively?
>
> Thanks for the help,
>
> Brian
> _______________________________________________
> G8b_run mailing list
> G8b_run at jlab.org
> https://mailman.jlab.org/mailman/listinfo/g8b_run