The optimal cosmic ray detector for High-Schools. By Floris Keizer

Transcription

1 The optimal cosmic ray detector for High-Schools By Floris Keizer

2 An air shower Highly energetic cosmic rays Collision product: Pi-meson or pion Pions decay to muons and electrons A shower of Minimum Ionizing Particles (MIPs) reaches Earth

3 The HiSPARC detection station

4 How to get more accurate data Timing: time Voltage Make sure that the distance to the photosensitive area is short and equal (to reduce multipath dispersion) Interval of the pulse Efficiency: fraction of counts due to air showers Influenced by the timing and light yield on the photosensitive area

5 Research at the Cavendish Laboratory, The University of Cambridge A novel detector using the Silicon photomultiplier (SiPM) and Wavelength-shifting fibre (WLS-fibre)

6 The SiPM + Cheap + Low bias Voltage (±18V) Small light sensitive area (WLS-fibre) High noise rate

7 Noise analysis of the SiPM

8 Noise analysis of the SiPM

9 Average Max. Voltage (mv) Noise analysis of the SiPM Therefore, the threshold was set on the green line in the following plot: -15 Average signal for different noise peaks First elektron Second electron 2.5 Electron estimate Third electron estimate 18 18, , ,5 Bias Voltage (V)

10 Frequency (khz) Efficiency (%) Noise analysis of the SiPM The SiPM, connected to a scintillator tile, with the determined threshold performed as follows: (The measurements are at room temperature) Noise at 50 mv Threshold Efficiency at 50 mv Threshold , , ,5 20 Bias voltage (V) , , ,5 20 Bias voltage (V)

11 Research at the Cavendish Laboratory, The University of Cambridge

12 The splice Two WLS-fibre ends bundled and cut off Easily fabricated by polishing and glueing

13 The splice 26% more efficiency than the mirrored end Similar light yields, so improvement in timing 80 Efficiency and light yield Efficiency (%) Abs(average MIP signal) (mv) Splice (diameter 1.0 mm) Splice (diameter 1.5 mm) WLS-fibre mirrored end WLS-fibre reference

14 Adding more scintillating material

15 Adding more scintillating material 1.5x the amount of scintillating material means less performance! Significant difference between detector 2 and 3 Both effects must be caused by the different orientation/length of WLS-fibre in the three detectors Detector 3. Efficiency and light yield 37 Efficiency (%) 60 Double Detector tile Abs(average MIP signal) (mv) Double Detector tile 2. Single tile Detector 1

16 Adding more WLS-fibre

17 Adding more WLS-fibre Conclusion: rather than generating light, transporting light to the SiPM is crucial for the light yield Approximately linear relationship for Detector 5; absorption of light in the WLS-fibre is neglectable The coil-like orientation of detector 1 and 4 proves best 120 Efficiency and light yield Efficiency (%) Abs(average MIP signal) (mv) , Detector 5, 40 cm fibre Detector 5, 80 cm fibre Detector 5, 120 cm fibre Detector 1, glued Detector 4, glued

18 Potential new data: the flux A plot by Jochem D. Haverhoek in April 2006 A better resolution of the detector: new information on the energy and the coordinates of the core/centre of the shower Note: the HiSPARC scintillator tile has large dimensions to detect showers at a long radial distance

19 Outlining the optimal detector The SiPM rather than the PMT (costs; safety) Its noise can be suppressed, a small fan should maintain room temperature A splice rather than a mirrored end (timing) WLS-fibre in a coil-like orientation (light yield) The WLS-fibre glued (63% more light yield) A thin scintillator tile with a large area: to avoid multipath dispersion (timing); this is irrelevant for the light yield Aluminium foil wrapped around the scintillator tile (reflection of light; 41% more light yield) No corners in the scintillator tile (resolution) Try to make the shortest distance for light to travel to the WLS-fibre as constant as possible (resolution and timing) I tried to combine all these aspects in the following design:

20

21