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).