KLM summary - PowerPoint PPT Presentation

About This Presentation
Title:

KLM summary

Description:

... the groove to increase the light collection by 1.4-1.7: equally efficient for ... A p.e. ~ 2 units (1.2 mV) Test of the SiPM read out by evaluation board ... – PowerPoint PPT presentation

Number of Views:94
Avg rating:3.0/5.0
Slides: 17
Provided by: Pakh
Category:

less

Transcript and Presenter's Notes

Title: KLM summary


1
KLM summary
  • P. Pakhlov (ITEP)

Status of Geant4 simulation of scintillator
KLMby Gallina PAKHLOVA (ITEP) Optimization of
the endcap acceptanceby Roman MIZUK (ITEP)
Read out electronics testsby Dmitry
LIVENTSEV(ITEP) Mechanical structure for endcap
KLMby Pasha PAKHLOV (ITEP)
Parallel-A KLM
2
Comparison with T2K
  • Use and imrpove T2K experience. Why we have to
    differ? Belle KLM strips are at least 2 times
    wider to reduce cost (light 1/sqrtwidth) and
    (some of them) 1.5 times longer (light
    exp(-length)). We need to be economical and try
    to collect as much light as possible to keep high
    efficiency at far end!
  • T2K 1 mm Kuraray Y11 fiber. We are going to use
    1.2 mm fiber, taking advantage of 1.3 X 1.3 mm2
    SiPMs (both Hamamatsu and CPTA produce this
    size).
  • T2K fibers have dry connection to the strips.
    We glue fiber to the groove to increase the light
    collection by 1.4-1.7 equally efficient for both
    central groove (Fermilab strips) and sawed groove
    (Kharkov strips).
  • T2K aluminized the polished end and got 70
    mirroring efficiency. We glue the 3M mirror to
    the far end this double the light yield from the
    far end.

10
BelleII
T2K FGD
3
Strips mass production
  • Operations, required to produce strip segment
  • Quality control of fibers (automatized with
    fiber rewind)
  • Cuting fibers and polishing fiber ends (to be
    done simultaneoulsy for hundreds fibers)
  • Glueing the mirror to the far end (experience of
    T2K group from INR, Moscow is useful)
  • Glueing fiber inside the strip groove
    (experience from OPERA)
  • Glueing the SiPM housing to the strip
  • Quality control of SiPMs (and database for 16k
    SiPMs)
  • Glueing the SiPM to the housing and soldering to
    preamplifier
  • Glueing 15 strips into one segment (good for
    transportation, match with 16 channels
    electronics, 5 segments fit the existing iron gap
    size) Unlike T2K we do not use substrate because
    of different thermal expancion of polysterol and
    good materials for skin (G10, Al etc)

To be elaborated during full size prototype
production (fall 2009)
  • Estimated time and manpower
  • Strip production (12000 kg) 4 6 months.
  • Fiber production (30 km) ?? (to be checked with
    Kuraray)
  • SiPM production (16 k) ?? (to be checked with
    Hamamatsu and CPTA)
  • Glueing 20 30 min/strip 16 k (3 5)
    man-years
  • Mounting segments to frame 2 hours 4
    man/sector 104 0.3 man-year

preliminary
4
Mechanical structure
Al sector frame is required for transportation at
B4, mounting/dismounting in iron gaps. The
existing ones are obviously fit the gap well (and
are not cheap) try to reuse them
The weight of one sector with RPC Al frame
20 kg two Al covers
2 30 kg RPC
120 kg Plastic support str 40 kg
Cables/strip masks etc 20 kg Total
250 kg
The weight of one sector with scintillator
strips Al frame 20
kg Scintillator 120 kg
Support str 20 kg Cables
etc 10 kg Total
170 kg
5
Acceptance and support structure
  • Acceptance is similar to the present one with
    RPC (the inner part can be milled to circle,
    provided the mirrored fiber is slightly shorter)
  • Rigidity of the frame is provided by the net of
    Al profiles in x and y directions welded or
    screwed to the frame.
  • Segments are screwed to the Al profiles in few
    points, as polysterol thermal expansion is much
    bigger than that of Al.

6
Mounting segements
  • Using I-beams The x-y profiles are welded
    together segments are inserted from one side
    then the assembled x-y net is screwed to the
    frame.
  • Using T-beams the x-y profiles are welded
    together and to the frame segments are inserted
    from the top and screwed to the profiles

7
Sector installation, repair
With the present configuration the installation
is only possible when KEKB optics near IR is
dismounted. Ineterference with time schedule for
KEKB installation? No chance to repair? Can this
be improved?
8
Electronics (by D.Liventsev)
Test of the SiPM read out by evaluation board
provided by Gary
1 kOm 1 GHz
SiPM signal with LED (5 p.e.) A p.e. 2 units
(1.2 mV)
Noise spectrum besides random noise some
structures are seen (x-talk of digital ? analog
part?)
9
Scheme with preamplifier
Working with simple preamplifier, gain 15. The
noise is not increased, The signal is 30 units/
p.e. To use the borad dynamical range (-500
500 units) the gain should not be larger than
15.
10
With preamplifier P.e. peaks are seen even
without LED (SiPM noise) In LED spectrum (mean
4.2 p.e.) the p.e. peaks are nicely
resolved. Time resolution (even with 4 p.e. only)
is better than 1 ns, this is very important for
fake KL rejection provide time of flight
measurement for slow KL
11
Conclusion on electronics
  • The universal read out board is appropriate for
    our tasks, However some improvements are still
    required either reduce the internal noise of the
    board (differential input should help a lot?)
    or(and) to amplifiy the SiPM signal
  • X-talks between channels are measured and
    consistent with Garys note 2-5 (not a big
    problem, but it is better to reduce them)
  • HV ajustment is required. Can it be implemented
    in the read out board?
  • Can we place read out board to the iron gaps near
    SiPM
  • reliability, ageing?
  • heating?
  • optimization of geom. size?

12
Optimization of the Endcap Acceptance
Roman Mizuk (ITEP, Moscow)
no signal, only bg?
Nakano-san
13
backward
forward
1
2
3
4
5
6
7
K L M L A Y E R S
8
9
10
11
12
13
14
R (hit position), cm
R (hit position), cm
forward full acceptance of all layers required
backward layers 1-5 full acceptance
(Rinner130cm) 6-8 Rinner150cm
9-12 180cm
more detailed optimization should be done with
Belle II MC
14
Status for G4 simulation of scintillator KLM
  • Galina Pakhlova, ITEP

Many thanks to Leo and Hara san
15
Geometry structure
  • klm_end_cap
  • 4 klm_end_cap_sector
  • 14 klm_end_cap_gap
  • klm_end_cap_scint (module)
  • klm_end_cap_air
  • klm_end_cap_strip
  • 75 vertical
  • 75 horizontal

16
Plans by TDR
  • Elaborate the mass production operations.
  • Design the necessary machinery.
  • Fix the mechanical structure! Or at least
    estimate reasonability of few possibilities. In
    the moment all sizes are flexible. If there are
    constraints on geometry from mechanics, it should
    be understood soon.
  • Produce full size prototype using Fermilab
    Kharkov strips and Hamamtsu CPTA SiPMs
    (September-October).
  • Deliver the prototype to KEK and check the
    compatibility with mechanical structure
    (January10).
  • Choose the producers for strips, SiPMs.
  • Optimize the geometry using Geant4 simulation.
  • Decision on electronics design.
  • KLM trigger (Nakano san, Yusa san).
Write a Comment
User Comments (0)
About PowerShow.com