1. The detector arrangement 2. The lifetime system 3. Digital Doppler measurement 4. AMOC RK R Martin-Luther-Universität Halle The detector system of the EPOS system The detector system of the EPOS system
1. The detector arrangement
2. The lifetime system
3. Digital Doppler measurement
4. AMOC
RKR
Martin-Luther-Universität Halle
The detector system of the EPOS systemThe detector system of the EPOS system
• 3 experiments: lifetime spectroscopy (16 BaF2 detectors); Doppler coincidence (2 Gedetectors), and AMOC (1 Ge and 1 BaF2 detector)
Detector systemDetector system
• digital detection system:
- lifetime: almost nothing to adjust; time scale exactly the same for all detectors; easy realization of coincidence
- Doppler: better energy resolution and pile-up rejection expected; easy coincidence setup
0 10000 20000 30000 40000 50000 60000 70000
100
1000
10000
100000
MC-simulated spectrum:lifetimes: 0.15 ns 2 ns 140 nsIntensity: 5 % 10 % 85 %
Cou
nts
Time (ps)
Repetition time: C E G I
Simulation parametersStatistics: 107
FWHM: 0,2 nsBackground: 0,04%Channel width: 0,1 ns
Beam repetition time for lifetime spectroscopy Beam repetition time for lifetime spectroscopy
Martin-Luther-Universität Halle
• trep=77 ns repetition time is standard operation mode at ELBE
• is used for FEL’s
• is perfect for lifetimes τ < 7 ns, i.e. most materials
• for nano-porous materials: trep too short
• slow mode with trep > 500 ns necessary
• real advantage of electron LINAC
• primary time structure can be adopted with low loss of overall intensity • thumb rule: τmax = 0.1 trep
Lifetime systemLifetime system
Martin-Luther-Universität Halle
• lifetime will be measured with direct digitizing of anode pulses
• external coincidence system is required to avoid useless digitizing (not shown)
one of eight parallel lifetime channels
Lifetime systemLifetime system
Martin-Luther-Universität Halle
• problem with mixing into one channel: delay cable must be elongated for slow mode
• several choices for the detector tubes: XP2020, HH3378-50, R7400U-096, and also the MCP-PMT’s(previous talk)
Selection of PMT
Philips
Type XP2020QHead-on
H3378-50Head-on
R7400U-096Metal package
R3809U-57MCP-PMT
photocath.diameter (mm)
BA51.0
BA51.0
Cs-Te11.0
Cs-Te11.0
voltage (V)gain
rise time (ns)transit time (ns)TTS (ps)
1.4 0.7 0.78 0.1528 16 5.4 0.55~250 370 ~100 25
windowrange (nm)peak λ (nm)quant. eff.
fused silica fused silica fused silica160-650 160-650 160-320420 420 2400.25 0.24 0.113000 3000 800 -30003×107 2.5×106 5×104 2×105
cost (EUR)
MgF2115-320
2300.11
1000 3650 700 15000
EPOS-02 (J. Cizek)
XP 2020XP 2020
Martin-Luther-Universität Halle
Sweep of two anode signals
Digital lifetime spectrometer
• XP 2020 too slow for a positron pulse of σt< 100 ps (TTS ≈ 250 ps)
recorded with 2 GS/s
• successfully used e.g. at Tokyo University by H. Saito
• obtained resolution of 110 ps in coincidence setup (only ≈ 50 counts/s)
• not very stable in long-term use
• two tubes available in Halle
• in May 2004 we’ll get 2 Photek MCP-PMTs for testing -> we will compare all detectors for final decision
Martin-Luther-Universität Halle
Hamamatsu H3378Hamamatsu H3378--5050
photo taken at Tokyo University
Martin-Luther-Universität Halle
Anode spectrum of Hamamatsu R7400UAnode spectrum of Hamamatsu R7400U--0909
• a faster digitizer is required (>= 4 GS/s)
• however: very small window (only 10 mm opening)
• Anode pulse of ultra fast Hamamatsu R4700U-09 as measured with 2 GS/s digitizer
• spectral sensitivity fits best for BaF2: sensitivity for slow component (310 nm) reduced by 0.10 compared to fast one (220 nm)
recorded with 2 GS/s
R 7400U-09
Cs-Te + silica photocathode
EPOS-02 (J. Cizek)
slowfast
• geometrical problem when arranging 16 detectors in a ring
Martin-Luther-Universität Halle
PhotekPhotek PMT 325PMT 325
• expected pulse height of single-stage PMT´s: ≈1 mV
• however: expected lifetime > 10 a of continuous operation
• amplification necessary (40…60 dB)
• easily done for fT > 5 GHz
• amplifiers also necessary to decouple the anode pulses when mixed together to one coincidence channel
• otherwise the anode pulses are intermixed and the coincidence circuit cannot work
Martin-Luther-Universität Halle
Amplifiers necessaryAmplifiers necessary
• sample rate can be smaller 50…100 MHz
• resolution should be 14 bit (16384 channels)
• when 511 keV is at 80% of maximum energy -> 39 eV/ch
• only one such digitizer available: Compuscope 14100 (GaGe)
• is PCI-Card with dual input (single input: 100 MS/s and dual input: 50 MS/s)
• one card with additional memory is available in Halle (27 k€)
Martin-Luther-Universität Halle
Digital Doppler measurementDigital Doppler measurement
• single channel Doppler and coincidence mode easy to realize by software
• hope: better energy resolution, higher throughput
• better time resolution expected
• more accurate detection of pile-up pulses -> lower background at E > 511 keV
• which is only reason for high momentum background in a 511 keV system
Martin-Luther-Universität Halle
Doppler coincidence easy to realizeDoppler coincidence easy to realize
-2000 0 2000-20
0
20
40
60
80
100
120
140 Channel A Channel B
Ampl
itude
(mV
)
Time (µs)
recorded with 50 MS/s and 12 bit resolution
-200 -100 0 100 200-25
0
25
50
75
100
125
150
Channel B Channel A
Ampl
itude
(mV)
Time (ns)
Martin-Luther-Universität Halle
• when additional disturbance of signal -> pulse will be ignored
• example additional RF of about 1 MHz overlayed
• only pulses with certain shape can be selected
Martin-Luther-Universität Halle
LineLine--shape discrimination possibleshape discrimination possible
-2000 -1000 0 1000 2000
-20
-10
0
10
20
30
40
50
60
70
Channel B Channel A
Ampl
itude
(mV)
Time (ns)
recorded with 50 MS/s and 12 bit resolution
• combination of lifetime and Doppler spectroscopy
• test needed -> can both cards operated in the same PC
• external coincidence required
Martin-Luther-Universität Halle
AMOCAMOC
• Digital detector for the whole EPOS possible
• Main advantages:
- hardly anything to adjust
- extreme stability
- easy remote control
- time scale for all detectors exactly known and equal within 10-5
- line-shape discrimination of pulses possible
• drawback: pulse rate of a single detector limited by online software processing of pulses to about 5x104 s-1
Martin-Luther-Universität Halle
ConclusionsConclusions
Thank you for your attention!
This presentation can be found as pdf-file on our Websites:http://positron.physik.uni-halle.de
http://positronannihilation.net
contact: [email protected]