Author: Petra Krivkova
Supervisor: Dr. Andrew Brandt
This report summarizes the general description and updated status of the Tevatron accelerator, the D0 experiment and its Forward Proton Detector. During the summer 2001 I worked at D0 as a member of the FPD group. My activities during this period are summarized at the end of this report.
The Tevatron was designed and later upgraded to collide the protons and antiprotons, at beam energy 1TeV and luminosity 2x1032 cm-2s-1.
The accelerator system was (and is going to be more) upgraded to be able to
reach the higher luminosities and higher energy of the colliding particles. To
improve the accelerator performance: the Anti-proton Recycler is going to be
rebuild; the number of the colliding bunches is going to be increased (132ns
bunch separation).
Note: The minor changes had been made to the Tevatron layout to be able to
install the Forward Proton Detector.
After the bunches of protons and antiprotons are created and
accelerated to the sufficient energies, they are injected to the Tevatron. The
beams are not steady flow of particles, but these particles occur in bunches.
So far the maximum number of the proton and antiproton bunches is 36 and 36, with
the following structure: The Tevatron circumference is 6.28 km and the
particles move at speed of light, it means that one turn lasts 21 ms. There are three superbunches within the
turn.
Each bunch contains 12 bunches which are separated by 396 ns, it means that the
superbunches are long approximately 4.36 ms
and the gap between the individual superbunches is 2.6 ms.
The Tevatron was designed and is operated the way that the collisions can happen only at D0 and CDF. The bunches are spaced that way that at D0 (same for CDF) happens 36 bunch crossings (collisions of bunches) per the evolution (21 ms) during the 36x36 operational mode.
Operational mode 36x36 means that in the Tevatron are 36 bunches of protons and 36 bunches of antiprotons and that the collisions happen every 396 ns at D0 (also at CDF) except the time between the superbunches.
Operational mode 1x8 means that in the Tevatron is one bunch of protons and
8 bunches of antiprotons. Usually the first bunch of the first superbunch of
protons is filled and first four bunches of the second and the third superbunch
of antiprotons. So at D0 (same for CDF) only one collision of bunches can happen per the evolution. Note: During
such mode the FPD detector was timed in.
The D0 experiment was built in late eighties on Tevatron to
study the proton-antiproton interactions. Between years 1992 and 1996 the data
were taken for the first time. One of the main results was the observation of
the top quark in 1995.
In 1996 The D0 started to be upgraded to be able to run at the new accelerator
conditions. The data taking began again in March 2001.
The D0 experiment is a complicated system of several subdetectors. The closest
to the beam is Silicon Tracker, which main aim is to measure momenta of
outgoing particles, reconstruct primary and secondary vertexes. The Silicon
tracker is inside the new superconducting magnet together with the Fiber
Tracker, which measures the particle momenta. Outside the solenoid is the
calorimeter, which measures the energy and direction of jets and the individual
particles. In front of the calorimeter is the preshower, which should help in
electron identification. The furthermost from the beam is the muon detector
with its own magnetic field. Its purpose is to measure the momenta of muons.
The Forward Proton Detector was incorporated into the Tevatron lattice at
several positions around D0. FPD is going to detect and measure the momenta of
protons and antiprotons scattered at small angles after the interaction.
The amount of outgoing raw data from D0 is very huge so the sophisticated
triggers to record only the interesting events have to be implied. This is done
in three steps using first hardware and later software.
The Silicon Tracker consists of silicon disks and barrels
formed into 6 disk/barrel modules. Each barrel module consists of 4 (radial)
layers of detector ladder assemblies. Layers one and three are single sided
silicon microstrip detectors and layers two and four are double sided with 2-degree
stereo angle. Each disk module has 12 wedge shaped double-sided detectors with
a 30 degree stereo angle. There are also three sets of end disks on each side.
The special SVX chips were developed to read out the signal from the Silicon
Tracker.
The expected hit resolution is 10 mm.
The CFT consists of scintillating fibers mounted on eight concentric cylinders. The fibers are double clad and are 835 ms in diameter. The fibers are constructed in ribbons each 128 fibers wide composed of two 'singlet' layers. These singlet layers are formed into the 'doublet' layers which form the ribbon by placing the fiber centers of one of the singlet layers in the space between the fibers of the other singlet layer. The base-line Central Fiber Tracker design consists of eight doublet layers of scintillating fiber aligned along the beam axis, Axial layers. And four each of U and V stereo fibers which make about a
2-degree angle with the beam axis.
The light from the fibers is converted into electrical pulses by visible light photon counters, VLPCs. These are small silicon devices which have an array of eight photo sensitive areas, each 1 mm in diameter on their surface. When operated at temperatures from 6 to 15 degrees Kelvin they have a quantum efficiency value of over 80% and a gain of 20,000 to 50,000.
The momenta of charged particles will be determined from their curvature in the 2 T magnetic field provided by a 2.7 m long solenoid magnet with the radius 60cm.. The superconducting (SC) solenoid provides the uniform magnetic field inside the tracking volume. This uniformity is achieved in the the absence of a field-shaping iron return yoke by using two grades of conductor with higher current density near the ends of the coil. From the value of the field integral and the space point precision provided by the silicon and fiber tracking system: Dp / p2 = 0.002. The SC coil plus cryostat is about 0.9 radiation lengths thick. The solenoid is the first thin solenoid for a particle physics detector which operates at 2 T.
The D0 Run II Luminosity Monitor consists of two arrays of plastic scintillation counters located on the inside faces of the end-cap calorimeters, 135 cm from the center of the D0 detector along the z direction (beam axis), and arranged symmetrically about the beam pipe. The detector covers a region in the pseudorapidity of 2.7 < |h| < 4.4.
Each of the Luminosity Monitor arrays consists of 24 identical 5/8'' thick scintillator wedges, with Hamamatsu 1'' diameter fine-mesh photomultiplier tubes mounted directly on the faces. The counters will be located in a region where the magnetic field maps indicate a nearly axial magnetic field of appr. 1 T.
The central and forward preshower are made of layers of
scintillators and lead. Scintillators are connected to the WLS (wave length
shifting) fibers, which are read out by VLPCs.
The central preshower has the cylindrical shape which covers region |h|<1.2 and is placed between the solenoid
and the central calorimeter cryostat. The forward preshowers have the shape of
plates, consisting of the triangular scintillators, positioned in front of the
end-cap calorimeters.
The D0 has the highly stable with good energy scale accuracy and resolution, liquid argon calorimeter since RunI. It consists of three units, the Center Calorimeter (CC), and the two End-cap Calorimeters (EC).
Each calorimeter cell consists of a liquid argon cap between an absorber plate and a G10 board. The G10 board has a high-resistivity coating to which a potential is applied with respect to the absorber plate to create the drift electric field. Particles transversing the gap produce and ionized trail of electrons and ions. In the electric field the electrons drift toward the G10 coating, producing a current. The current induces and image charge on a copper pad etched on the G10 board under the resistive coat. A readout cell is formed from many pads grouped together.
The signal from each cell is brought to a feed through port on a resistive coaxial cable. Then the signal is carried from the feed through port to the preamplifier inputs on twist and flat cables. The preamps, mounted on a motherboards, The preamps integrate the charge produced by the calorimeter cells to produce voltages. The pulses of voltage are carried again by twist and flat cables to the shaper and baseline subtractor (BLS). The BLS process removes slowly varying offsets in the input voltage. The outputs are then read out and digitized by the Analog to Digital Converters (ADC).
The components needed to be replaced for the upgrade were the cables between the feed-through ports and the preamps, the preamps and their motherboards, the shapers, BLS's and sample and holds, the power supplies for both these systems, and a new charge injection pulser system for calibration and monitoring stability. The timing, trigger and readout control also needed to be modified to handle the faster timing, higher trigger rates and analog buffering expected later in RunII.
Muon system consists of three parts: one Central detectors,
which covers the region |h|<1 in
pseudurapidity, and two Forward Detectors, which cover the region 1<|h|<2.
Both central and forward detector consist of three layers of scintillators and
drift chambers. The
toroid magnet is installed between the first and the second layer.
The information from the scintillators is used together with the CFT signals in
L1 trigger to achieve the unprescaled trigger of the interesting events. The
information from the drift chambers is combined to calculate the muon momenta.
The trigger system at D0 has three levels.
Level 1: pipelined hardware stage, with tracking and calorimetry;
Level 2: second hardware stage, which combines and refines Level 1 information
with preprocessors and a global processor;
Level 3: software stage, in which a Level 3 processor farm partially
reconstructs events.
Level 1
The Level 1 Trigger System is a hardware trigger system filtering the 7.59 MHz Beam Crossing rate with minimal Dead Time (less than 5% in normal desired running conditions). The Level 1 Trigger System consists of the Level 1 Trigger Framework and the Level 1 Trigger Subsystems.
Each Level 1 Trigger Subsystem processes detector-specific information and produces, for every Beam Crossing, Input Terms to the Level 1 Trigger Framework. These And-Or Network Input Terms convey to the Framework a summary of the activity seen by the Level 1 Trigger Subsystem in its associated Detector component.
Using these Framework Input Terms and other information about the readiness
of the DAQ system to begin another acquisition cycle the Level 1 Trigger
Framework determines for each Beam Crossing whether the resulting event should
be rejected, or captured for further analysis in the Level 2 Trigger System.
The desired rate for events selected by the Level 1 Trigger System is 10 kHz.
The requirements for the L2 Global processor are to handle a 10 kHz input rate, provide the rejection factor of approximately 10 with high signal efficiency and provide information to the L3 trigger to assist the software filter decision.
The L2 consists of the hardware framework, the preprocessors from the
individual subdetectors and the global L2 processor, which combines the
information from L1 and the L2 preprocessors and performes the final decision.
The Level 3 is the software trigger stage, which uses the
high performance commercial processors to run event filter algorithms, with
each filtering node handling complete events.
The data input to the L3 is from the buffer memory modules of the digitizing
crates. The output event rate from L3 should be about 10-20 Hz. The outgoing
data from L3 are stored on the tapes.
The FPD is one of the new subdetector at D0. The FPD is the
fiber tracker detector located at several positions around the D0 collision
point inside the beam pipe in the Tevatron tunnel. The two qaudrupole
spectrometers (A1 at +23 m and A2 at +31 m from the D0 collision point)
and two dipole spectrometers (D1 at +57 m and D2 at 59 m from the D0
collision point) detect the scattered antiprotons and two quadrupole
spetrometers (P1 at –23 m and P2 at –31 m from the D0 collision point) detect
the scattered protons. Each quadrupole has four detectors: 'U' - up from the
beam, 'D' - down from the beam, 'O' - outside of the accelerator ring and 'I' -
inside of the accelerator ring. Each dipole has one detector which is in the
position outside from the ring. All eighteen positions are labeled as P1U, P1D,
P1O, P1I, P2U, P2D, P2O, P2I, A1U, A1D, A1O, A1I, A2U, A2D, A2O, A2I, D1 and
D2.
At the Tevatron the proton and antiproton are collided. They interact between each other mainly via the strong interaction. The theory which describes these interactions is called the quantum chromodynamics (QCD). The QCD is very succesful in the describing the interactions during which the high momenta are tranfered - 'hard' processes, but almost 40% percent of the strong interactions at Tevatron are 'soft', caused by the elastic (the scattering particles remain unchanged after the interaction only their direction is different) and diffractive (proton or antiproton or both survive the interaction, while they loose only small amount of their initial energy; during this process the secondary particles are produced only in the certain areas of the detector) scattering. In this regime the QCD is not succesful yet and there exists no other fundamental theory which would desribe these 'soft' interactions.
During RunI the hard diffractive scattering (HDS) was observed. In this
process again the proton or antiproton (single HDS) or both (double HDS)
survive and in the final state the jets or W,Z boson appear ('hard' process
happened) while some areas of the detector are completely without outgoing
particles.
At the first time such processes were observed at UA8 on the Spp~S
collider in CERN, later also at ZEUS and H1 on HERA collider in DESY.
As the part of the D0 upgrade the FPD will have the capability to detect the
scattered protons and antiprotons at small angles and using the Tevatron
magnetic field measure their momenta and energies. Then using the central D0
detector the whole final state will be analyzed and the (hard) diffractive and
elastic scattering studied.
The FPD purpose is to measure the protons and antiprotons
scattered at small angles. To do that the detector's active area has to be
inside the beam pipe. As it was said before the FPD is a part of D0 upgrade, so
the changes to the Tevatron lattice had to be made. The place was found for six
castles. Each castle is the vessel from the stainless steel, connected at the
opposite ends to the Tevatron pipe. Inside has to be maintained the high vacuum
as it is in the whole Tevatron. The castles at dipole position have one Roman
pot each, the castles at antiproton and proton quardupole side have four Roman
pots each. These Roman pot are the parts of the castles, which is possible to
move using the remotely controlled motors. The pots are designed to be able to
insert the cartridges housing the detectors into them, so the detectors
themselves would not be in the vacuum. The active area of the detectors is at
very bottom of the pots. The special 150 mm
steel window, which is the bottom part of the roman pot, was designed to
separate the active area of the detector from the vacuum; so when the scattered
particle passes through the detector it first has to go through the thin steel
window, but because the window is very thin, there is a small probability that
the particles are going to interact inside it.
The cartridges are made of aluminium. Inside each cartridge is one fiber
detector together with 7 MAPMTs and 1 L0PMT. Through the lid of the cartridge
the signal and high-voltage cables go.
To control the pot motion remotely, the special electronics had to be installed
in the tunnel and in the control room and the software to control it was
written.
FPD's final assembly will consist of 18 fiber detectors. Each detector has 6 layers of optical fibers and one optical block (called also the 'trigger'). As the particle after the interaction goes through the detector, it passes the detector in the following order: u,u' layers -> optical block -> x,x' layers -> v,v' layers.
The fibers in the x,x' layers are parallel to the optical block, the fibers in u,u' layers are rotated by +45 degrees, the fibers in the v,v' layers by –45 degrees.
The u,u',v,v' layers have 20 channels each. The x,x' layers have 16 channels
each. One channel is 0.8 mm thick, the separation is 0.27 mm between channels.
Each two layers with the same orientation of the fibers have the channels
shifted with respect to each other, so when the particle passes for example
through the u,u' layers it has to go through the optical fiber of at least one
of these layers.
Each channel have 4 square (diameter appr. 0.8mm) fibers. In the frames the
fibers of the individual channels are positioned above each other (1x4 fibers)
to increase the amount of light produced by one channel. As the fibers face the
photomultiplier they are rearranged to 2x2 formation to cover efficiently the
photocathode.
The active area of each detector (where the light can be produced) is 17mm x 17mm perpendicular to the beam. This was reached by splicing of the scintillating fibers with the clear fibers and by the gluing of the scintillating block to the clear block. The environment of the scintillator in each detector is placed at the position closest to the beam; during the data taking the particles go through and produce the light. The light then transmitted via the clear fibers and clear block to the photomultipliers.
The fully assembled FPD detector is going to have 2016 channels of the optical fibers and 18 channels of the optical blocks.
18 channels of the optical blocks are going to be read out by 18 Philips
photomultipliers (L0PMTs), which operates at high voltage appr. –1800 V.
Signal from all these photomultipliers is sent without amplifying via long high
quality coaxial cables to the trigger electronics, positioned in the D0 movable
counting house. After combining the trigger signals the decision is made if the
event will be passed to the further data processing or not. This is done at
Level 1 trigger.
2016 channels of the optical fibers are read out by 126 Hamamatsu
multianod-photomultipliers (MAPMTs), each of them has 16 channels and operates
at high voltage appr. –960 V. So for each detector we have 7 of these
photomultipliers (6 of them read out channels 1-16 of the layers
u,u',x,x',v,v'; 1 photomultipliers reads out the channels 17-20 of the layers
u,u',v,v').
From MAPMTs the signal goes via the short (6 m or 12 m) 16 channel coaxial
cables to the amplifiers, which are still in the Tevatron tunnel. From all
amplifiers the signal goes via the long (appr. 100 m) 16-channel coaxial cables
to one crate, where the analog signal is converted to the digital one, and it's
passed to other electronics and online computers to find the tracks in the
detectors. Depending on the results of the tracking, the event is read out or
not - this is done at Level 3 trigger stage.
FPD uses 5 Light Mixing Boxes (LMBs) to calibrate their
photomultipliers and check the whole read-out system. One LMB is going to be at
each castle position (dipole, A1, A2, P1, P2).
The LMB is the light-tight box. Inside there is a small circuit with light
emitting diodes (LEDs), a clear optical block to mix the light produced by
LEDs. The end of the optical block is faced to the black coated optical fibers,
which distribute the light from LMB to the photomultipliers.
Inside the LMB there is also the PIN diode which checks the amount of light
produced.
LMB has two input and on output cables:
On March, 1 2001 the Tevatron started to collide protons and
antiprotons and the D0 and CDF began to collect the data after the
several year shutdown.
The phase between the March and October 2001 is called the commissioning run.
The checking of the detector and DAQ system is taking place and the
installation of the remaining parts is still going on during the short
shutdowns, which are partly caused by the accelerators break downs and
necessary repairs after the long shutdown.
The accelerator performance
Between March, 1 and April, 2 only the proton studies took place. 1x8 stores were performed at the beginning of April. Since the middle of the April the Tevatron is mostly running in 36x36 mode and still improving its performance - increasing the luminosity.
D0 detector installation
The central detector components are almost completely in place. What is
still missing, is to install
the read-out electronics for the Fiber Tracker, Preshowers,.. and some trigger
electronics.
D0 DAQ
The main effort at D0 is now to understand and improve the trigger performance of the trigger network and individual subdetectors.
D0 data analysis
The first data analysis were done. They were reconstructed muons based on the Muon Detector response, jets distribution using the calorimeter, tracks of the individual particles by the silicon and fiber trackers.
The FPD has istalled two fully assembled ('real') detectors and eight pseudodetectors (roman pot with only the scintillator trigger and L0PMT inside) in the tunnel. The real detectors are at P1U and P2U positions. The pseudodetectors are at P1D, P2D, A1U, A1D, A2U, A2D, D1 and D2. The drivers and controller to move the pots are all installed, except at D1 and D2 positions. The amplifiers and LMBs are at P1U and P2U positions to readout and calibrate the 'real' detectors and their MAPMTs. All cables have been laid out between the tunnel and the D0 detector platform and also between the D0 platform and the Small Control Room.
During the commissioning run the FPD is running in the standalone DAQ mode,
because the read-out electronics for the trigger (L0PMTs) and for the signal
from fiber detector (MAPMTs) was not delivered yet. To commission our detectors
within the commissioning run we had to assembled our own DAQ. This DAQ system
consists of the NIM logic and CAMAC and VME modules. This system is placed in
the room next to the main D0 control room - it's called the Small Control Room
(SCR). To this room we had to bring all our signal cables and in addition we
brought also the cables with the Luminosity Monitor Detector signal and D0
clock.
In SCR there is the online PC on which we run the 'pot motion' program, high
voltage program and DAQ program. The pot motion control goes via the same VME
crate as DAQ, which allows us to send the information from both to the ACNET.
Trigger and DAQ
The first step which was done was to add the additional short cables to some
L0PMT signal cables to have the same cable-delay time for all L0PMT signals.
These signal cables are then plugged to the discriminators which convert the
analog signal to the digital one (it works the way that when the amplitude of
the analog signal is bigger then the threshold the digital signal is created;
this digital signal has the adjustable width - we used 50 ns to be able to
maintain the coincidences between the protons or antiprotons quadrupoles). From
the discriminators the signals go to the logic AND-OR units where are created
the proper coincidences. The trigger output from the NIM logic goes to the
CAMAC. When the CAMAC receives the trigger, it writes the data onto the PC
harddisk via VME create. This data consists of the rate information from the
L0PMTs, the time of the arrival of the L0PMT signal and the signals from the
fibers. To read out the analog signal from the fibers we use the CAMAC FERAs
(analog-to-digital convers, which integrate the negative signal over the
time window, called the ADC gate. Due to the given lenght of our cables we had
to ensure ourselves that the trigger decision is made soon enough to be able to
create the ADC gate before the signals from the fibers arrive to the
FERAs).
The fact that our data are read via the VME crate allowes us to send the
information about the L0PMT rates to the ACNET. This allowes together with the
information about the position of our pots to the operators in the main
accelerator control room to study the position of the beam.
I stayed at Fermilab and worked at the D0 experiment from
May, 16 till August, 31 2001.
During this period I worked full-time on the FPD project. I took a part in the
FPD group discusions, attended their meetings. I worked mainly on the hardware
projects, because the FPD is now in the commissioning stage.
The standalone DAQ and timing in
Temporarily the FPD has to use the standalone DAQ, which consists of the NIM
logic modules and CAMAC and VME crates. The trigger decision is made in the NIM
modules based on the signal from the L0PMTs, the D0 clock and the signals from
the Luminosity Monitor Detector.
If the trigger is satisfied, the event (L0PMT rates, timing information, signal
from MAPMTs) is collected and written to the disk; the single L0PMT rates and
the position of the pots are also sent to the ACNET.
I was helping with the setting up the trigger logic and took a part in
discussions about timing in our signals.
We used the special 1x8 store to time in our detector, when the collisions
happen at D0 only once per evolution. By comparing the structure of the L0PMT
signals we found when the collisions are happening and timed these signals to
the D0clock. The same was done for the Luminosity Monitor signals.
Pot motion and the L0PMT rate studies
During summer we were inserting the Roman pots near the beam. First we moved
the pots at very small steps to study if the pots are influencing the beam or
not. (In case we would influence the beam we would expect the higher rates of
the proton and antiproton halo at D0 and CDF, because as the beam would
interact with our pots, some particles would be sent at the peripheral orbits
and later be lost from the beam.)
We moved all 'up' and 'down' pots at quadrupole positions and set the maximum
position limits at which the beam is not influenced.
While moving the pots the single rates of L0PMTs were also studied. We observed the increased rates of those L0PMTs, which pots were moved closer to the beam. Later we set up the elastic trigger and looked at its rate. Next step is going to be to take the data while pots are inserted under the elastic trigger condition and look at the response of our detectors - if we are able to reconstruct the tracks.
Testing the LMB system
During the summer the two pots (P1U and P2U) are assembled with the
detectors. Because of the standalone DAQ the pulse generator and the low-voltage
power supply are used to produce the SLP signal in the small control room, from
where the signal goes via the long signal cable to the tunnel, where is
splitted to two LMBs. In SCR we can check the PIN output signal to verify that
the light is produced inside the LMBs.
I took a part on the LMB installation and testing in the tunnel as well as the
testing in the SCR. During RunII (after FPD will integrate fully into D0 DAQ
system) the SLP pulse is going to be produced by the SLP boards which are
located in the movable counting house (together with the majority of the D0 DAQ
electronics). The SLP pulse produced by these boards can be controlled by the
software.
I was testing the FPD channels of the SLP board and the response of the LMB,
when these boards are used as the source of the SLP signal.
Testing the MAPMTs using the LMB
After the detectors were installed, it was needed to test the MAPMTs response and their connections. To do that we used the LMB system. With the HV on the MAPMTs we sent the SLP pulse to the LMBs and looked channel by channel at the response in long coaxial cables carrying the amplified signal from MAPMTs.
Cutting and Qualifying fibers
In august the part of the FPD scintillating and clear fibers was delivered
to Fermilab to be cut and polished. These fibers are going to be used to build
the rest of the FPD detectors.
The fibers delivered were 1m long, so I had to cut them. In total I got 5000
10cm long scintillating fibers and 3000 50cm long clear fibers. After the
cutting the fibers were polished in one of Fermilab's lab using the special
method, called the ice-polishing.
After the polishing we were inspecting the quality of the polished surfaces.
During the polishing it can happen that the cladding is chipped off from the
core of the fibers. Sometimes also the black spots can remain on the polished
surface. Such defects would lead to the light losses so we were rejecting such
fibers for further polishing.
We were also sorting the fibers by their diameter to have better efficiency
during the splicing procedure, when the scintillating fiber is melted to the
clear one.
Testing the amplifiers
FPD is going to use 63 amplifier boards to improve the outgoing signal from
the MAPMTs. The FPD amplifiers were used already during RunI by the other D0
subdetector, so they have to be retested before the further use.
The amplifier boards have 32 channels, so they can handle the signal from 2
MAPMTs each. They are housed in the special crates powered by the low voltage
power supplies.
The simple tests were done to check these boards. They were tested channel by
channel using the oscilloscope and the pulse generator, if the signal passes
through.
Making the optical piece and L0 LMB fibers
To Fermilab the more MAPMTs should arrive in September. These
photomultipliers have to be tested before we install them in the tunnel. To do
this test the LMB is going to be used as the source of the light. Because each
MAPMT has 4 by 4 photocathods, I made the special optical piece to do this
test.
I used one of the cookies, 16 25 cm clear fibers (which I inserted into the
transparent plastic tubing to protect them) and 16 special plastic connectors
to be able to connect the optical piece with the LMB fibers. To assemble
everything together I used the optical glue. The plastic connectors and the
cookie were then polished in one of the Fermilab labs.
To be able to calibrate and checked the response also of the L0PMTs there was recently proposed to use the LMB system also for L0PMTs. To do that the optical fiber has to go also to L0PMT inside the cartridge. I glued the 20 connectors to clear fibers and let them polish. These connectors will be later connected to the LMB fibers.
Presentation
In july I took a part in the Fermilab graduate students presentation. I made a poster about the construction of the FPD detectors. During the poster session I was then explaining to people how the FPD detector are suppose to work and how they are assembled. I was awarded the third place.
In this report the D0 experiment, its FPD subdetector and the Tevatron accelerator are decribed. The updated status of all of these is given. At the end my activities during the period from 16 July till 31 August 2001 are given. I spent in Fermilab three and half months as a member of D0 working in FPD group.