Agenda of the 14th EVOmeeting, GSI, 15.02.2008
- "Mapping Separation power", talk, Klaus Götzen
- Question: Status report on the next GSI-PANDA meeting?
- next meeting
Results and Decisions
"Mapping Separation power"
(using the Fast Simulation), talk, Klaus Götzen
We had one very good presentation where Klaus showed how we can use the information from the Phase Space plots of the reaction channels that we got from Marc Pelizaeus and Bertram Kopf and the parametrization of the detector performance that you provided to find eventually gaps in the particle separation power of our PANDA PID system. He discussed how to determine the separation power from Simulation (using the "rms" of the particle distributions) and where problems are (having Landau-tails in the dE/dx for example). For the separation power deduced from two (or more) detectors covering the same momentum and theta regin he used the square root of the quadratic sum of the single separation powers.
Thus for the performace of the single detectors Barrel DIRC, Disc DIRC, RICH, MVD, STT, TOF he produced a Map and combined them to a complete Map of the PANDA PID detectors.(The TPC was not shown but will be used in the next step by performing the tracking calculations with a better dE/dx resolution of 7% instead of 20% for the STT. The EMC is more difficult to include since there is not something as separation power but probabilities determined from shower shapes and so on. Nevertheless this important information should appear in short future also)
To find where we need good separation Klaus compared the momentum and angle distribution of the Kaon from a benchmark reaction (pbar p-> phi phi -> K+K-K+K-) with the background produced with the DPM-Generator. This reaction seems to be covered well by our PID detectors and the Kaon seems to be well separated. Yet there appears a gap in the overlapp region between the Barrel DIRC and the Disc DIRC underneeth22 degrees. This might be an artefact due to the finite granularity of the map or comes from an older Disc DIRC parametrization that does not yet include all known effects to optimize the performance. Klaus Föhl will send the actual values to be included in the Fast Simulation.
- In the end Klaus discussed the limitations of his approach.
- Information based on parametrizations only
- no EMC informations so far
- distributions are not gaussian
- does our definition of sep. power make sense here
- due to finite granularity (bin width) \u2192 artifacts in the map
- is the estimate for the total nSigma,tot correct?
- Lars Schmitt added
- that we have to add the Myon detectors to this map,
- that the separation e-pion has to appear (Klaus made the separation for neighbouring particles only) since in many reactions (time lke formfactors, Drell Yan) this is the crucial information to have
- On other points to add we agreed that are
- that we have to make for different reaction channels special background simulations (beyond the DPM-Generator)
- that the binning of the map should be finer (demanding more statistics)
So the very good work of Klaus Götzen showed where we still have to work and
which information still has to go in his calculation (provided by us).
Update of the "Mapping of Separation", Klaus Götzen, 19.02.2008
- included are now:
- TPC plots
- e/pi separation
- finer Binning + higher Statistics (1M per particle species)
- p upto 6 GeV/c
- n_sigma upto 8 sigma
- not yet there
- new input von Klaus Föhl for the Endcap DIRC
- verification of the RICH parametrization; see the values:
- _detName = "Rich";
- _dthtc = 0.01; // single photo resolution = 10 mrad
- _nPhotMin = 5; // minimum photon number cutoff
- _nRefrac = 1.05; // refraction index
- _effNPhotons = 0.075; // photo-detector effective efficiency
- _dRich = 1.0; // radiator length = 1 m
- _dp = 0.01; // relativ momentum resolution
- it looks good but one of the RICH experts from Glasgow might give here a statement
Question: Status report on the next GSI-PANDA meeting?
- We agreed that we will give a status report on the next PANDA meeting at GSI to show our progress.
- Georg mentioned that we have to write a "Recommendation on the PID Detectors used in PANDA" until the end of summer (based on Fast Simulations presumably). It was agreed to start already now but not to have a preliminary report before the final report mentioned above.
- the next meeting will take place on Wednesday the 5th of March at 11:00 in C24 (container 24) (please ask local people to find this new meeting location) (this meeting will be chaired by Carsten Schwarz since I can join the PANDA meeting only at Wednesday afternoon)