[G8b_run] Polarization modification

[Michael Dugger]

Volker,

Yes.

Using your criteria, I think that 4% would be a good estimate for the 
systematic uncertainty associated with the consistency of the 
polarization.

Take care,
Michael

On Fri, 27 Jan 2012, Volker Crede wrote:

> Dear Michael,
>
> Thanks for the update. I have already implemented your new numbers and submitted a few jobs on our cluster.
>
> Neglecting the 2.1 GeV dataset and comparing the numbers in your table (Slide 3), would you agree that the data can still be made self-consistent to within 4%? This is the largest number I see in your table (for the 1.7 (auto) - 1.9 overlap region, PARA). My feeling is that your new results are even somewhat better than the previous ones. Can we conclude that 4% is a good estimate for the systematic error associated with the degree of polarization provided that:
>
> 1) The latest polarization tables have been used.
>
> 2) The polarization corrections (based on the overlap studies) have been applied.
>
> 3) The event-based energy cut has been applied:  event edge - 200 MeV < E(gamma) < event edge.
>
> Best wishes and thanks again
>
>      Volker
>
> PS: I am sending this from my gmail account, so it may not go to the g8b run group. FSU has a huge email problem at the moment and JLab regularly rejects emails from @fsu.edu.
>
>
> On Jan 27, 2012, at 2:24 PM, Michael Dugger wrote:
>
>>
>> Hi,
>>
>> I have made some slight changes to the polarization modification. Volker
>> and Ken helped me determine that there was an inconsistency in how I was
>> determining which polarization look-up table to use for each event. The
>> inconsistency led to the coherent edge being wrong by up to 4 MeV (~ half
>> an eCounter).
>>
>> The results are very similar to what I had shown previously. A pdf showing
>> a comparison of the new results to the old ones can be found at
>> http://www.jlab.org/Hall-B/secure/g8b/ASU/polCorrectionV3.pdf
>> where the new results are in blue.
>>
>> Note: There is one difference in results worth mentioning. I no longer see
>> the need to remove eCounters > 299 (within the overlap region) for the 1.9
>> GeV data set :)
>>
>> To implement the new modification, one only needs to replace the ldPar
>> values with the those given at the bottom of this email (also listed on
>> the pdf).
>>
>> I apologize for any inconvenience this may cause.
>>
>> Sincerely,
>> Michael
>>
>> New ldPar values:
>>
>> 1.3 manual
>> ldPar(1) = 1.007560
>> ldPar(2) =-0.5872520E-02
>> ldPar(4) = 1.012520
>> ldPar(5) =-0.6416970E-02
>>
>> 1.5 manual
>> ldPar(1) = 1.012790
>> ldPar(2) =-0.4507690E-02
>> ldPar(4) = 1.002800
>> ldPar(5) =-0.1880110E-02
>>
>> 1.7 manual
>> ldPar(1) = 0.9893110
>> ldPar(2) =-0.1880110E-02
>> ldPar(4) = 0.9893210
>> ldPar(5) =-0.2127790E-02
>>
>> 1.7 auto
>> ldPar(1) = 0.9962880
>> ldPar(2) =-0.1159220E-02
>> ldPar(4) = 1.030480
>> ldPar(5) =-0.1110110E-01
>>
>> 1.9 auto
>> ldPar(1) = 1.012930
>> ldPar(2) =-0.2772200E-02
>> ldPar(4) = 0.9998200
>> ldPar(5) =-0.1616160E-02
>>
>> 2.1 manual
>> ldPar(1) = 0.9968230
>> ldPar(2) =-0.2214870E-01
>> ldPar(4) = 1.007930
>> ldPar(5) =-0.7647950E-02
>>
>>
>> _______________________________________________
>> G8b_run mailing list
>> G8b_run at jlab.org
>> https://mailman.jlab.org/mailman/listinfo/g8b_run
>>
>
>
>