[BDXlist] BDXmini energy calibration

[Andrea Celentano]

Dear all,
I loaded a first version of the energy calibration constants for BDXmini crystals. I determined the calibration constants by taking the charge distribution for data corresponding to events with cosmic rays passing through a given crystal. Selection was performed by using the information from the TOP or BOTTOM veto caps, and determining which INNER VETO SiPM had the largest charge, event by event. In this way, I did a sort of “tracking” for cosmic rays. For each crystal, I tried to determine the selection resulting in the corresponding charge distribution having a well defined minimum-ionizing peak.

(see the slides that I showed last week at our meeting: https://gcc01.safelinks.protection.outlook.com/?url=https%3A%2F%2Fdocs.google.com%2Fpresentation%2Fd%2F1MLOFdwaQ50VfW7A1lKUgi7b4k7XDBkEGhE_oTnmeGig%2Fedit%3Fusp%3Dsharing&data=02%7C01%7Cbdxlist%40jlab.org%7C18e94e97f7274f73974808d6c98b1d0c%7Cb4d7ee1f4fb34f0690372b5b522042ab%7C1%7C0%7C636917997185411526&sdata=YxyPEyxh8vyrwWrl9xOZMSZ2ZKjxuz3uLXyz2y3iYx4%3D&reserved=0 <https://gcc01.safelinks.protection.outlook.com/?url=https%3A%2F%2Fdocs.google.com%2Fpresentation%2Fd%2F1MLOFdwaQ50VfW7A1lKUgi7b4k7XDBkEGhE_oTnmeGig%2Fedit%3Fusp%3Dsharing&data=02%7C01%7Cbdxlist%40jlab.org%7C18e94e97f7274f73974808d6c98b1d0c%7Cb4d7ee1f4fb34f0690372b5b522042ab%7C1%7C0%7C636917997185411526&sdata=YxyPEyxh8vyrwWrl9xOZMSZ2ZKjxuz3uLXyz2y3iYx4%3D&reserved=0>)

After Marco produced the MC files for cosmics, I analyzed them in the same way. MC data is already reported in absolute units (MeV). Therefore, the calibration task is to determine the scale factor for the charge of each crystal, E = Q / scale.

I did so by fitting the MC shape to the data (I used RooFit do to so), allowing the scale parameter to be a free parameter (the MC is also normalized to the data on the Y scale - RooFit handles this automatically).

Preliminarily, I determined the starting value of the scale parameter by doing the fit “by eye”, adjusting the scale until a good agreement was found. I did a manual test, changing by +- 10% the cal. constants, and I saw that with this modification the MC shape is no longer in agreement with data. Hence, I expect the quality of cal. constants to be of this order of magnitude.

In the attached PDF, you find the fit results for run 1071, the 3-days long run we took last weekend. Results from run 1070 (2 days-long) are similar.
I loaded these calibration constants on our CCDB, so that the reconstruction is now getting them automatically for the objects of type “CalorimeterHit”. 

Next step will be to put this information in the RootSpy gui - so that by eye one can see if the system is stable or not - and eventually have the calibration procedure done automatically at the run end. We can use a CCDB variation that is not loaded by the reconstruction, to have cal. constants saved - but not used by default.

Bests,
Andrea



-------------- next part --------------
An HTML attachment was scrubbed...
URL: <https://mailman.jlab.org/pipermail/bdxlist/attachments/20190425/8ef4afb5/attachment-0002.html>
-------------- next part --------------
A non-text attachment was scrubbed...
Name: data.1071.root.pdf
Type: application/pdf
Size: 1318405 bytes
Desc: not available
URL: <https://mailman.jlab.org/pipermail/bdxlist/attachments/20190425/8ef4afb5/attachment-0001.pdf>
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <https://mailman.jlab.org/pipermail/bdxlist/attachments/20190425/8ef4afb5/attachment-0003.html>