Transcript ScECAL_Jun

Short report on
MPPC gain monitoring system
S. Uozumi
Jun-27 2008 ScECAL meeting
At the FNAL BT, having the gain monitoring system for all the channel is crucial for :
• keep measuring the MPPC gain
… Correct drift of the MPPC gain occurred by some reasons.
Also indispensable for saturation correction.
A few photoelectrons are necessary in each channel.
• Electronics inter-calibration
… need to measure gains of electronics in low and high gain mode.
Also necessary for the saturation correction.
Large amount of light (~100p.e., at least >10p.e.) is necessary to be distributed.
So far, not so good…
(Ikeda-san`s talk on Jun-13)
●:acrylic plate
●:side emitting fiber 1
●:side emitting fiber 2
●:optical fiber with
notches
Dark noise level (0.02 p.e.)
Revision of the system
• Look over whole system and correct any problematic points.
–
–
–
–
Optical connections
status of fiber
Fixture of fiber
And so on …
• Change LED to very bright one.
• Use CMB to issue LED driving pulses.
Some remarks on light distribution
• Connection between different fiber/material causes a lot of light loss.
• We should just use clear fibers, rather than acrylic bars or side-emitting
fibers with clear fiber bundles.
• After realizing this, only the choice is clear fiber with notches.
We will adopt it for the FNAL BT.
Acrylic bar
Clear fiber
Almost of lights are lost here
Clear or side-emitting fiber
Clear fiber with v-shape notches
Revised bench-test system
CMB
Cable (~5cm)
LED
Fiber bundle
(20 fibers ganged)
f1mm clear fiber
V-shape notches
Scintillator-layers
(Kuraray fiber mega-strip
used at 2007 DESY BT)
CMB
•
•
•
•
Originally developed for AHCAL calibration system by Prague guys.
Can drive up to 12 LEDs.
Provide 8-50 ns pulse into LED.
Currently bench test is ongoing with borrowed CMB, but 2 CMBs will be
delivered soon from Prague.
• Control through CANbus is recently succeeded by Daniel.
12 LEDs
CANbus
line
Power
TCALIB (timing signal)
VCALIB (amplitude signal)
Adopt Nichia NSPW500GS-K1 W-rank white LED
One of the most efficient LED currently available.
Brightness : 31000 - 44000 mcd @ IF=20mA (~x3 blighter than LED previous used)
Forward voltage / current : 3.2-3.5V / 30mA (100mA possible with pulsing mode)
Emission spectrum more-or-less matches to MPPC PDE spectrum.
Blightness (arbitrary unit)
Blightness (arbitrary unit)
•
•
•
•
•
LED
Wavelength l (mm)
PDE of the MPPC
Emission angle
Clear fiber bundle
• Bundle of 20 clear fibers, made for 2000 HCAL beam test @ FNAL.
• Diameter of the fiber is 1mm.
(One concern : does it fit to 1 mm gap in ScECAL?)
• Fiber length ~ 75cm.
• White disk is attached for fixture purpose.
Notches on the clear fiber
• Making many uniform notches is important.
• To achieve that, melting method by heating tool would be better
than cutting or gliding.
• f=0.8mm wire is used as a space to control the depth of notches.
LED
Notches
• With the notched fiber, stronger light is emitted to opposite
side of the notches.
Fiber arrangement on the mega-strips
• 3mm-wide slits are made on reflector and black sheet.
(2mm hole will be enough for actual system at FNAL)
• Notched fiber is put on the slit (notches come upside)
and fixed by black tape.
Mega-strips (Kuraray fiber readout, used at DESY BT)
Holes for MPPCs
Notched
fiber
To bundle
Result of quick test
• CMB DAC count can be set to 0-255 counts.
(it is proportional to amplitude of the LED driving pulse.)
• LED starts flashing around DAC~100 counts.
• With DAC~120, we saw >10 photons at most far-side strip.
• With higher DAC setting (>150 DAC count), all the strips get
too much light which causes ADC overflow.
• Plots and numbers will follow soon, but notched fiber seems
to be distributing large amount of lights more than enough!
Summary & Plans
• Revised system with CMB, new LED and notched fibers works
well for the MPPC gain monitoring.
• At the first quick test, we saw large amount of light with this
system.
• Plots, numbers and some more tests will follow soon.
• At the same time, we will design the actual layout of the
system at the FNAL beamline.