\documentclass[11pt,a4paper]{muontdr}
\setcounter{secnumdepth}{4}
%\renewcommand{\paragraph}{\@startsection{paragraph}{4}{0mm}{-\baselineskip}%
%{0.5\baselineskip}{\normalfont\normalsize\bfseries}}
\usepackage{epsfig}
\usepackage{biblhcb}
%x\usepackage{amstext}
%\usepackage{chaplhcb}
%\psdraft
\usepackage{graphicx}
\usepackage{here}
\usepackage{rotating}
\usepackage{mathptm}
\renewcommand{\textfraction}{0.1}
\renewcommand{\floatpagefraction}{0.8}
\renewcommand{\topfraction}{0.9}
\renewcommand{\bottomfraction}{0.9}
\begin{document}
\pagenumbering{Roman}
\thispagestyle{empty}
\vspace*{-10mm}
\begin{tabular}{ll}
\rule{110mm}{0mm}&{\bf CERN/LHCC 2003-002} \\
& {\bf LHCb TDR 4 Addendum 1} \\
& {\bf 15 January 2003} \\
\end{tabular}
\vspace{65mm}
\begin{center}
{\bf \Huge LHCb}\\
\vspace{10mm}
{\bf \LARGE {Addendum to the \\ Muon System Technical Design Report} \\[25mm]}
\vspace*{1\baselineskip}
{\bf \Large{LHCb Collaboration}} \\
\vspace{15mm}
%{\bf \Huge DRAFT-0}\\
%printed \today\\
\vspace{45mm}
CERN\\
Geneva, 2003\\
%ISBN 92-9083-180-4
\end{center}
\setcounter{page}{0}
\pagenumbering{roman}
%\cleardoublepage
~
%\newpage
%\input{TDRauthors}
%\newpage
%%%\input{acknowledgments}
%\newpage
%\tableofcontents
%\cleardoublepage
%~
\newpage
\begin{figure*}[t]
\centering
\includegraphics[angle=90,width=12cm]{eps/lhcb-light-det.eps}
\caption{Side view of the LHCb spectrometer.
M1 -- M5 are the five Muon Stations.
\label{lhcb_layout}}
\end{figure*}
\cleardoublepage
%Uncomment for 2 column output
%\twocolumn
%\linewidth=\columnwidth
%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
%definition
%% global
\def\LHCb{LHCb}
%%%%%
\newcommand{\bfmath}[1]{\mbox{\boldmath$#1$\unboldmath}}
%%%%% macros for LHCb Muon TDR
%%
%% Luminosity
\newcommand{\lum}{\mathcal{L}}
\newcommand{\lumin}[2]{{#1}\;10^{#2}\;\mathrm{cm^{-2} s^{-1}}}
\newcommand{\lolumi}{\mbox{$\lum\ =\,2 \times 10^{32}$~cm$^{-2}$ s$^{-1}$~}}
\newcommand{\hilumi}{\mbox{$\lum\ =\,5 \times 10^{32}$~cm$^{-2}$ s$^{-1}$~}}
%% Units
\newcommand{\mq}{\ensuremath{\; \mathrm{m^2}}}
\newcommand{\cmq}{\ensuremath{\; \mathrm{cm^2}}}
\newcommand{\mmq}{\ensuremath{\; \mathrm{mm^2}}}
\newcommand{\cms}{\ensuremath{\mathrm cm^{-2}s^{-1}}}
\newcommand{\us}{\; \mu\:\mathrm{s}}
\newcommand{\bb}{$\rm b\overline{b}$}
\newcommand{\invcmq}{\ensuremath{\mathrm{/{cm^2}}}}
\newcommand{\mHzpercmq}{\ensuremath{\mathrm{MHz/cm^2}}}
\newcommand{\kHzpercmq}{\ensuremath{\mathrm{kHz/cm^2}}}
\newcommand{\Hzpercmq}{\ensuremath{\mathrm{Hz/cm^2}}}
\newcommand{\Cpercm}{\ensuremath{\mathrm{C/cm}}}
\newcommand{\Cpercmq}{\ensuremath{\mathrm{C/cm^2}}}
\newcommand{\ItwoC}{\ensuremath{\mathrm{I^2C}}}
\newcommand{\x}{\ensuremath{\times}}
\newcommand{\micron}{\ensuremath{\mathrm{\mu m}}}
\newcommand{\gra}{{\mathrm{^o}}}
\newcommand{\oC}{\; \mathrm{^oC}}
\newcommand{\NIM}{Nucl. Instr. and Meth.}
\newcommand{\Meg}{\ensuremath{\mathrm{M}{\Omega}}}
%\renewcommand{\today}{1 Mar 2001}
\def\GeVc{GeV$/c$}
\def\GeVcc{GeV$/c^2$}
\def\MeVcc{MeV$/c^2$}
%\def\microns{$\mu$m}
\def\PT{p_{\mathrm{T}}}
%%%
\def\Y#1S{{\Upsilon\mathrm{(#1S)}}}
%
\def\pipi{\pi^+\pi^-}
\def\ubar{\bar u}
\def\dbar{\bar d}
\def\sbar{\bar s}
\def\cbar{\bar c}
\def\bbar{\bar b}
\def\tbar{\bar t}
\def\Kbar{\rm {\kern 0.2em\overline{\kern -0.2em K}}{}}
\def\Bbar{\rm {\kern 0.18em\overline{\kern -0.18em B}}{}}
\def\Dbar{\rm {\kern 0.2em\overline{\kern -0.2em D}}{}}
\def\pbar{\rm {\kern 0.06em\overline{\kern -0.06em p}}{}}
\def\Jpsi{\ensuremath{\rm J\mskip -3mu/\mskip -2mu\psi\mskip 2mu}}
\def\Ks{\rm {K^0_S}}
\def\Kl{\rm {K^0_L}}
\def\BS{$\rm B^0_s$}
%%%%%%
\def\BJMUK{$\rm B^0_d \rightarrow \Jpsi(\mu^+ \mu^-) K^0_S$}
\def\BJMUKVIS{$\rm B^0_d \rightarrow \Jpsi(\mu^+ \mu^-) K^0_S(\pi^+ \pi^-)$}
\def\BJMUKBOLD{$\bfmath{\rm B^0_d \rightarrow \Jpsi(\mu^+ \mu^-) K^0_S}$}
\def\BJMUPHI{$\rm B^0_s \rightarrow \Jpsi(\mu^+ \mu^-) \phi$}
\def\BSMUMU{$\rm B^0_s \rightarrow \mu^+ \mu^-$}
\def\BSMUMUBOLD{$\bfmath{\rm B^0_s \rightarrow \mu^+ \mu^-}$}
\def\BJPSIKS{$\rm B^0_d \rightarrow \Jpsi Ks $}
\def\BMUX{$\rm b \rightarrow \mu\ X$}
\def\pkm{$\rm \pi,K \rightarrow \mu\ \nu$}
%%
%Gas
\def\C2H2F4{$\rm C_2H_2F_4$}
%\def\isob{$\rm iso-C_4H_{10}$}
\def\isob{$\rm C_4H_{10}$}
\def\SF6{$\rm SF_6$}
\def\CF4{$\rm CF_4$}
\def\CO2{$\rm CO_2$}
\def\rpcgas{\ensuremath{\mathrm{C_2H_2F_4/C_4H_{10}/SF_6}} (95:4:1)}
\def\mwpcgas{\ensuremath{\mathrm{Ar/CO_2/CF_4}} (40:50:10)}
\let\bra\langle
\let\ket\rangle
%%
%% General
%
%
%
\setcounter{page}{0}
\pagenumbering{arabic}
%
%=======================================================================
%
\begin{sloppy}
%
%
\section{Introduction}
\label{sect:introduction}
\begin{sloppypar}
This Addendum to the Muon System
Technical Design Report TDR (TDR) \cite{bib:muontdr}
is intended to be a concise
description of the modifications required in the Muon System of LHCb
to replace the Resistive Plate Chamber (RPC) detectors.
Fig. \ref{lhcb_layout} shows schematically the LHCb
experiment, with the five stations (M1--M5) constituting
the Muon System.
The reason for this change is basically an aging effect of the
RPC, which rapidly reduces its capability to handle high rates.
This is briefly explained in Section
\ref{sect:rpc}.
The RPCs will be replaced by MWPC detectors, identical to those
used in stations M2 and M3, regions R3 and R4,
of the Muon System, except for the
dimensions. A short update of the MWPC detector is given in Section 3.
The conversion to an all-MWPC system poses obvious questions
for the project organization.
480 RPC chambers,
covering about 48\% of the total
detector area, will have to
be replaced with wire chambers. This represents a considerable
increase (more than 30\%) in the number of MWPC to build. The
issues of project organization, including the schedule and
cost, are discussed in detail in Section \ref{sect:project}.
\section{RPC Aging}
\label{sect:rpc}
Resistive Plate Chambers were proposed in the Muon TDR
\cite{bib:muontdr}
for installation in the outer regions (R3 and R4) of stations
M4 and M5 (a total of 480 double-gap chambers).
Initial tests made by us on these detectors showed that they
could stand the maximum flux density foreseen
in those locations (750\,\Hzpercmq).
An extended aging test at the CERN GIF
was started in January 2001 on two identical
detectors. Preliminary results have been reported in
\cite{bib:passaleva,bib:ganis}.
One detector (RPC A)
was exposed to high radiation flux, the other (RPC B) was used as a reference
and shielded from the radiation. HV and gas were exactly the same
for the two detectors.
After integrating a charge of 0.4\,\Cpercmq\
equivalent to
10 LHCb-years in region R4, the rate capability of RPC A was somewhat
reduced, but still adequate to meet the TDR specifications.
During the test the current through RPC A decreased steadily
and dropped in total by a factor of about 6,
as can be seen from Fig. \ref{fig:IvsTime}.
We proved that this was due to an increase
of the bakelite resistivity. The increase continued
even when in the second part of 2001 the irradiation ceased,
while the gas was kept flushing through the detector.
The test was restarted in May 2002, this time with both detectors
irradiated by the GIF source. The resistivity referred to as
$\rho_{20}$ (the value
normalized at $20 \oC$)
was measured online with a
novel method described in \cite{bib:rpcnim1}. The
results (see Fig. \ref{fig:rhobak}) show a continuous increase
of $\rho_{20}$ for the two detectors, which reached by the end of December
2002 similar values around $2 \cdot 10^{12}\; \Omega$cm, although
RPC B had a lower resistivity at the beginning of the
test. The final value of $\rho_{20}$ is
about 100 times higher than the value at the time of
detector construction in 1999. This is most likely
due to the drying out of the bakelite when flushed
with dry gas. The irradiation plays a
secondary role: probably the current flow
through the bakelite merely speeds up the drying process.
Separate measurements of the rate capability, using the charged particle
beam of GIF, confirmed that the detectors could not stand more than
400\,\Hzpercmq. However, this value was obtained in July
at an average temperature of $26 \oC$, and must be reduced
by about a factor of two for operation at $20 \oC$ (the nominal
temperature in the experimental area), and by another
factor of two to take into account the increased bakelite resistivity
observed at the end of 2002 (Fig. \ref{fig:rhobak}). Thus the
rate capability will be at most 100\,\Hzpercmq, a
value completely inadequate for LHCb.
The trend of Fig. \ref{fig:rhobak} could suggest
a possible saturation of the resistivity, however this would still
be too high for the Muon System. It has been proposed
that a possible reversal of the effect could occur
by adding water to the gas
mixture. However, this could be dangerous in our
high-radiation conditions, where water could give rise to
HF acid formation, and was not considered as an option.
Our conclusion was that RPC detectors could be built
to meet the TDR specifications, but that their
performances would quickly degrade because of the resistivity increase.
This fact, and a series of problems recently encountered in the
RPC industrial manufacturing by other experiments
led us to the decision of
replacing them with MWPC detectors.
Aging of MWPCs is of no concern even at rates a factor 50 above the
expected rates in regions R3 and R4 of stations M4 and M5
\cite{bib:seva}.
\begin{figure}[htb]
\begin{center}
\epsfig{file=eps/fig10.eps,width=0.45\textwidth}
\caption{Current for RPC A corrected for temperature
plotted versus time during the 2001 aging test.
Deviations from the exponential decrease
are due to changes in the HV and gas mixture, and to insertion
or removal of other detectors in front of the source.}
\label{fig:IvsTime}
\end{center}
\end{figure}
\begin{figure}[hbt]
\begin{center}
\epsfig{file=eps/ra_vs_time_2002.eps,width=0.45\textwidth}
\epsfig{file=eps/rb_vs_time_2002.eps,width=0.45\textwidth}
\caption{Bakelite resistivity (normalized at $20 \oC$)
vs. time for two
similar RPC exposed to photons from the GIF during the
2002 aging test. The time origin is the beginning of the test
(1 May 2002).}
\label{fig:rhobak}
\end{center}
%\vspace*{-0.3cm}
\end{figure}
\cleardoublepage
\section{MWPC Detectors}
\label{sect:mwpc}
The MWPC detector has retained its basic structure, described in the Muon TDR
\cite{bib:muontdr} and in the references therein, except for some design
modifications dictated by the results of the various R\&D studies.
The most important are the increase of wire pitch from 1.5 to 2 mm and
the use of two gaps, instead of four, in station M1.
These modifications are described below.
The detailed design of the detectors will be discussed in
the EDR/PRR scheduled for April 2003.
\subsection{Wire Pitch}
The basic geometry of the MWPC as described in the TDR~\cite{bib:muontdr}
leads to an electric field of 8 kV/cm on the cathodes at the operating
point.
As a consequence, the tolerances for detector construction are very tight,
and the large electric field on the cathode might cause additional problems
in the long term operation.
It is well known that the cathode field can be reduced by increasing the
wire pitch, which leads on the other hand to a reduced time resolution and
in turn to a reduced efficiency within a 20\,ns time window.
Simulation
studies showed that the time resolution has an intrinsic limit and cannot be
improved in reducing the wire pitch below 1.5\,mm. This value has
therefore been assumed optimal and used for the prototype studies at the
time of the TDR, accepting the drawbacks caused by the large cathode
field.
In a recent beam test a detailed performance comparison of double-gap
chambers with 1.5 mm and 2 mm wire pitch has been carried out
\cite{bib:M3R1_note_in_preparation}. An important result has been that
a time resolution of about 4 ns at the operating point can also be
obtained with 2 mm wire spacing, leading to 99\% double-gap efficiency
within a 20 ns time window, fully satisfying our requirements.
Fig. \ref{fig:pitches} compares the results for both wire and cathode
readout obtained with the two different wire pitches. It was
therefore decided to adopt the larger wire pitch for all
chambers.
In addition 100 kCHF in wire cost can be saved.
\begin{figure}[htb]
\begin{center}
\epsfig{file=eps/eff15.eps,width=0.45\textwidth}
\epsfig{file=eps/eff20.eps,width=0.45\textwidth}
\caption{Double-gap MWPC efficiency for wire and cathode readout
in a 20 ns window. 1.5 mm pitch (left) and 2.0 mm pitch (right).}
\label{fig:pitches}
\end{center}
\end{figure}
\subsection{Layout of M1 chambers}
The chamber design is unchanged in stations M2 -- M5,
with four gaps, read independently in two
pairs.
In order to minimize the material in front of the electromagnetic
calorimeter and preshower,
M1 chambers will have only two gaps instead of four.
The two gaps will be read out by independent preamplifiers in
order to retain adequate redundancy. Simulations have shown that the
trigger efficiency is practically unaffected, even for single-gap
efficiencies as low as 80\% \cite{bib:note041}, leading to
a chamber efficiency of 96\%.
In station M1 the panel core will be Nomex honeycomb \cite{bib:muontdr},
in contrast to
polyurethane foam (Esadur 120) foreseen as core material for the panels
in the other stations. Esadur panels can be built industrially
for a rather low price (Fig. \ref{fig:mold}). On the other hand,
Nomex honeycomb has the advantage of
a radiation length considerably larger than Esadur.
With this design the average thickness of M1 decreases from 0.33\,$X_0$
to 0.15\,$X_0$.
We have a wide experience with Nomex honeycomb, since
it was used in most of the prototypes already built.
\begin{figure}[htb]
\begin{center}
\epsfig{file=eps/panel_mold.eps,width=0.4\textwidth}
\caption{Precision mold for industrial production of the panels
by injection of polyurethane foam. The foam is injected betweeen
two copper-clad FR4 laminates.}
\label{fig:mold}
\end{center}
\end{figure}
\subsection{MWPC production}
In the original planning four centers to produce 864 MWPCs
were forseen: one in
St.\,Petersburg's Nuclear Physics Institute (PNPI), two in Italy
(Ferrara and Laboratori Nazionali di Frascati, LNF), and one at CERN.
These centers will be equipped with similar tooling,
which is automated to a
large extent to speed up the construction, and allows to obtain the required
precision and tolerances. The main equipment consists of:
\begin{enumerate}
\item a table for gluing the wire frames to the panels
(Fig. \ref{fig:fe_wire})
\item a wiring machine (Fig. \ref{fig:fe_wire})
\item an automated station for laser soldering of the wires
(Fig. \ref{fig:lnf_solder})
\item an automated station to check the wire tension and pitch
(Fig. \ref{fig:lnf_solder})
\end{enumerate}
\begin{figure}[htb]
\begin{center}
\epsfig{file=eps/PNPI_gluing.eps,width=0.45\textwidth}
\hfill
\epsfig{file=eps/Ferrara_machine_2.eps,width=0.45\textwidth}
\caption{Gluing table for preparation of the panels (left);
automated wiring machine (right). This machine
can also be used for wire gluing.}
\label{fig:fe_wire}
\end{center}
\end{figure}
\begin{figure}[htb]
\begin{center}
\epsfig{file=eps/soldering.eps,width=0.45\textwidth}
\hfill
\epsfig{file=eps/CERN-pitch.eps,width=0.45\textwidth}
\caption{Laser soldering station (left);
station for wire pitch and tension check (right).}
\label{fig:lnf_solder}
\end{center}
\end{figure}
The implementation of this tooling minimizes human intervention,
in particular for the wiring and soldering of the planes.
The construction time for a chamber is now dominated by the
various gluing processes. Most of the epoxy-based glues used need about
8 hours before sufficient polymerization is reached.
The construction capacity in
the various centers is furthermore limited by the number of chamber components
which can be prepared in parallel and the available space.
Other time consuming phases are the final assembly and testing,
which require significant human intervention.
\subsection{Chambers for the inner part of M1}
The two inner regions (R1, R2) of station M1 are exposed to particle
rates above 100\,\kHzpercmq, and require good aging properties
for the detectors \cite{bib:muontdr}.
We have tested different technologies for this region: one is
triple-GEM, the others are
modifications of our standard MWPC design.
A full-size triple-GEM
prototype for R1 ($20 \times 24 \cmq$)
has been tested succesfully on beam \cite{bib:gem},
using standard GEM foils. This technology
seems particularly promising because of its aging properties. Our X-ray tests
have shown that triple-GEM performances do not suffer after more than 10-year
of LHCb operation.
The idea behind the MWPC modification is to reduce the accumulated charge
by halving the gas gain, without any change to the average signal charge
on the cathodes. In one design the cathode pads on both sides of the wire
plane are read out, in the other the wires are placed asymmetrically
in the gap at 1.25\,mm from the cathode pads read out, which allows also
to reduce somewhat the cluster size. A full size prototype for region R1
with both configurations was successfully tested on the beam~\cite{bib:amwpc}.
The accumulated charge on the wires
for the modified MWPCs would be about 1.4\,\Cpercm\
in 10 years of LHC operation in region R1, and 0.6\,\Cpercm\ in region R2.
A decision on the technology will be taken in Summer 2003, after
an extensive aging test has been performed at the ENEA-Casaccia facility
near Rome \cite{bib:casaccia} with a $^{60}$Co source.
A test run has already taken place.
\end{sloppypar}
\cleardoublepage
\section{Project Organisation}
\label{sect:project}
The decision to abandon the RPC technology poses a number of
organization issues to the Muon Group and to the
Collaboration.
Excluding the two innermost regions of M1, the number of MWPC detectors
passes from 864 to 1344, and in terms of surface area
this represents almost a 100\% increase.
Unlike RPCs,
MWPC detectors cannot be produced industrially and therefore
%more expensive than RPC and, in principle,
demand more manpower for the construction. On the other
hand, recent problems in the industrial
construction of RPC detectors have required a substantial increase
of institute personnel for quality control and tests, making
the overall effort rather similar for both detectors.
Some synergies are also possible for an all-MWPC
solution. In this section we will address these problems and present
our construction plan and schedule. We also update,
whenever applicable,
the information given in the TDR.
\subsection{Production centers}
According to the original plan with four
production centers, PNPI had the responsibility of the 384
chambers of R4 in stations M2,M3 (stations with wire readout
only). CERN was assigned the 72 chambers with combined
cathode-pad and wire readout of stations M2--M3, R1--R2.
The remaining 408 chambers (cathode-pad readout) were assigned
to Ferrara and LNF.
The production
of 480 extra chambers of large
size will require either a substantial boosting
of the capacity in the existing
four centers or the addition of new ones.
Boosting the capacity cannot be solved simply by adding extra
manpower, since also extra tooling and space is required. Moreover,
it would be complicated and expensive to move technicians and
physicists from one institute (e.g. one of those originally
involved in the RPC production) to another.
It was therefore considered more efficient to add
another center in Italy (INFN 3) and a second center in PNPI
(PNPI 2). For INFN 3 Firenze is the candidate
\footnote{The matter is subject to approval from INFN. Discussions
are in an advanced phase.}, since they could
easily enlarge the clean room already planned for CMS.
PNPI 2 will be the center presently used to assemble the CSC chambers
for the Endcap-Muon-System (EMU) of CMS, which will end its activity at the
end of 2003.
PNPI 2 will take in charge 192 chambers (M4R4)
but could increase this number in case of need.
The remaining chambers will be shared among the
Italian centers, which will in total be responsible for 696 chambers.
No changes for the CERN center are foreseen. The total
production capacity has enough margin to accomodate unforeseen
delays.
Table \ref{tab:centers} summarizes some basic parameters related
to the production. It is worth recalling
that the number of chambers is not a direct measure of
the construction effort.
The nominal production rates try to take into
account these factors and are
also corrected for the holiday periods.
In order to estimate the production curve, we have assumed that
the nominal rates will be reached after some
training period. Its duration was estimated to be
between four and seven months for the different centers,
during which the production rate will progress up to the
nominal value. The longer learning periods apply to centers
where extensive training of personnel is needed. Centers starting later
have a steeper learning curve, since they could
somewhat profit from the experience of those which began
their production earlier.
\begin{table*}
%\vspace*{-0.4cm}
\caption{Nominal capacity of the production centers and
approximate dates for start of
production (see text).}
\vspace*{0.5\baselineskip}
\label{tab:centers}
%\vspace{4mm}
\centering
\renewcommand{\arraystretch}{1.5}
\begin{tabular}{|l|c|c|c|c|c|c|}
\hline
Center & PNPI 1 & PNPI 2 & LNF & Ferrara & INFN 3 & CERN \\
\hline
No. Ch. & \multicolumn{2}{|c|}{576} & \multicolumn{3}{|c|}{696} & 72 \\
\hline
Rate & 20/mo. & 16/mo. & 10/mo. & 10/mo. & 10/mo. & 4/mo. \\
(est.) & & & & & & \\
\hline
Start date & Jul & Jun & Jul & Oct & Jan & Jul \\
& 2003 & 2004 & 2003 & 2003 & 2004 & 2003 \\
\hline
\end{tabular}
\end{table*}
Fig.~\ref{fig:prorate} shows graphically the production of
chambers. The production will
will be completed by beginning 2006.
Even considering the possibility of
redistributing part of the chamber production among centers in
case of unexpected difficulties,
the safe assumption is that the construction will end
in March 2006. The overall work program and schedule
are summarized in Fig.~\ref{fig:schedule1}.
\begin{figure}[hbt]
\begin{center}
\epsfig{file=eps/total.eps,width=0.7\textwidth}
\caption{Planned total chamber production in all centers.
For the input data see Table \ref{tab:centers}.}
\label{fig:prorate}
\end{center}
%\vspace*{-0.3cm}
\end{figure}
\subsection{Installation and commissioning}
Installation and commissioning are shown in
Fig. \ref{fig:schedule1}.
The first part of the muon system to be installed are the muon filters.
This work will be carried out during the year 2004.
\begin{figure}[hbt]
\begin{center}
\epsfig{file=eps/projplan.eps,width=0.95\textwidth}
\caption{Schedule of the LHCb Muon System, showing production and
installation of the detectors, electronics, and the infrastructure.}
\label{fig:schedule1}
\end{center}
%\vspace*{-0.3cm}
\end{figure}
The muon chambers will undergo installation and commissioning starting in
the second half of 2005, after the installation of the support structures
together with the required infrastructure (gas pipes, electronics racks etc.)
has been terminated. All chambers should be installed by the end of July
2006, with a short interruption for the LHC
injection test, scheduled in April.
Commissioning with other LHCb sub-detectors, using common DAQ will begin in
October 2006. Six months of operation in this mode are foreseen to ensure
the muon detectors will be ready to take data at nominal LHCb luminosity
in April 2007.
\subsection{Milestones}
The major
milestones for the Muon System are summarized in
Table~\ref{tab:milestones}.
A more detailed schedule, showing the details for the various
production centers,
will be presented in the EDR.
\begin{table*}
%\vspace*{-0.4cm}
\caption{Muon Project Milestones}
\label{tab:milestones}
%\vspace{4mm}
\centering
\small{
\begin{tabular}{|l|l|}
\hline
Milestone & Date \\
\hline
{\bf MWPC} & \\
\qquad Engineering design completed & 04.2003 \\
\qquad Begin chamber production & 07.2003 \\
\qquad 10\% chamber production & 03.2004 \\
\qquad 50\% chamber production & 02.2005 \\
\qquad Chamber production and test completed & 03.2006 \\
\hline
{\bf Chambers for the inner part of M1} & \\
\qquad Technology choice & 06.2003 \\
\qquad Chamber construction completed & 12.2005 \\
\hline
{\bf Electronics} & \\
\qquad Full chain electronics test completed & 06.2003 \\
%%\cline{1-1}
{\bf \em Integrated Circuits} & \\
%%\cline{1-1}
\qquad CARIOCA review and decision on FE chip & 09.2003 \\
\qquad DIALOG design and test completed & 09.2003 \\
\qquad SYNC design and test completed & 09.2003 \\
\qquad IC Engineering Run (CARIOCA,DIALOG,SYNC) & 12.2003 \\
%%\cline{1-1}
{\bf \em FE Boards} & \\
%%\cline{1-1}
\qquad Begin board production & 04.2004 \\
\qquad 10\% board production & 08.2004 \\
\qquad 50\% board production & 03.2005 \\
\qquad Board production and test completed & 10.2005 \\
%%\cline{1-1}
{\bf \em IB,ODE,SB Boards} & \\
%%\cline{1-1}
\qquad Begin board production & 04.2004 \\
\qquad 10\% board production & 12.2004 \\
\qquad 50\% board production & 06.2005 \\
\qquad Board production and test completed & 12.2005 \\
\hline
{\bf Muon filter and support structures} & \\
\qquad Iron filter installation completed & 12.2004 \\
\qquad Chamber support structures installed & 06.2005 \\
\hline
{\bf Commissioning} & \\
\qquad Muon System commissioning completed & 09.2006 \\
\qquad Muon System ready for beam & 04.2007 \\
\hline
\end{tabular}
}
\end{table*}
\subsection{Costs}
The total cost for the Muon System has been carefully revised.
Quite precise cost estimates are now available for most
items, including
the electronics, where all the components have been designed
and final prototypes are in preparation.
With respect to the
TDR we have an increase of the detector costs and a reduction
in the electronics cost. Part of the cost increase for the detectors
is due to the replacement of the RPCs.
Table~\ref{tab:cost} shows the cost estimate split according to the
system components. For the chambers and electronics about 10\%
for spares and contingency have been included.
The cost now includes the M1 inner part,
under the
assumption of a mixed GEM-MWPC solution, with triple-GEM
equipping region R1. The extra expenses
due to the modifications of the gas system have been taken into
account.
The overall cost of the system
(subtracting 4000 kCHF of the iron filter, which was already
made available from the CERN reserve) is slightly higher
with respect to the 6830\,kCHF of the TDR.
Note that the cost of about 200\,kCHF for the engineering run of the three
ASICs (CARIOCA, DIALOG and SYNC) is not included, as it is considered part of
the development phase.
The costs for the assembly in the PNPI production centers,
the shipping and part of the
tooling are reported in Table~\ref{tab:cost}
under the ``Miscellaneous'' entry.
The CARIOCA chip is the baseline option for the front-end;
the cost would increase by about
700\,kCHF in case the adapted ASDQ chip should be required.
%\
%\{\bf Note that the cost of about 200~kCHF for the engineering run of the three
%\ASICs CARIOCA, DIALOG and SYNC is not included, as it is considered part of
%\the development phase. (?)}
\begin{table*}[p]
%\vspace*{-0.4cm}
\caption{Muon project cost in 2000 prices (kCHF)}
\label{tab:cost}
%\vspace{4mm}
\centering
\begin{tabular}{|l|l|r|r|}
\hline
Item & Unit & Number & sub-total \\
& & of units& (kCHF) \\
\hline
{\bf MWPC detector:} & & & 2200 \\
\qquad Panels & piece & 7000 & \\ % 680 \\
\qquad Special cathodes & piece & 2000 & \\ % 265 \\
\qquad Wire & km & 2750 & \\ % 450 \\
\qquad Wire fixation bars & m & 6500 & \\ % 380 \\
\qquad Frames & piece & 21500 & \\ % 160 \\
\qquad HV boards & board & 5000 & \\ % 200 \\
\qquad Various connectors & piece & 30000 & \\ % 60 \\
\qquad Spacers & piece & 40000 & \\ % 5 \\
\hline
{\bf Miscellaneous:} & & & 680 \\
\qquad Tooling & & & \\ % 150 \\
\qquad Assembling & chamber & 576 & \\ % 300+230 \\
\qquad Shipping & chamber & 576 & \\ % 300+230 \\
\hline
{\bf Inner part station M1:} & & & 100 \\
\qquad Detectors & chamber & 36 & \\
\hline
{\bf Electronics:} & & & 3000 \\
\qquad CARIOCA chip & piece & 16000 & \\ % 50 \\
\qquad DIALOG chip & piece & 8000 & \\ % 50 \\
\qquad FE boards & board & 8000 & \\ % 165 \\
\qquad Spark Protection boards & board & 8000 & \\ % 225 \\
\qquad LVDS links & link & 8000 & \\ % 310 \\
\qquad IM boards & board & 165 & \\ % 290 \\
\qquad SYNC chip & piece & 4000 & \\ % 50 \\
\qquad Off-Detector-Elec. boards& board& 160 & \\ % 525 \\
\qquad Service boards (ECS) & board & 160 & \\ % 290 \\
\qquad L1 boards & board & 12 & \\ % 50 \\
\qquad Optical links to L0 trigger& link& 1250 & \\ % 410 \\
\qquad Crates & crate & 42 & \\ % 310 \\
\qquad LV power supplies/cables& system& & \\ % 125 \\
\hline
{\bf Services:} & & & 1000 \\
\qquad MWPC Gas System & system & 1 & \\ % 240 \\
\qquad MWPC HV System & system & 1 & \\ % 410 \\
\qquad Support Structures & module & 10 & \\ % 350 \\
\hline
{\bf Muon filter:} & & & 4000 \\
\hline
{\bf Muon System TOTAL} & & & {\bf 10980} \\
\hline
\end{tabular}
\end{table*}
\subsection{Division of responsibilities}
Institutes currently working on the LHCb Muon project are:
Centro Brasileiro de Pesquisas Fisicas CBPF,
Rio de Janeiro (Brazil), Universities and INFN of Cagliari,
Ferrara, Firenze, Roma ``La Sapienza'' (Roma I), Potenza,
Roma ``Tor Vergata'' (Roma II),
Laboratori Nazionali di Frascati LNF (Italy), Petersburg Nuclear Physics
Institute PNPI, Gatchina (Russia) and CERN. Work on the Level 0 Muon trigger
is carried out by CPPM Marseille in close collaboration with the Muon group.
The sharing of responsibilities for the main Muon Project tasks is
listed in Table~\ref{tab:responsibilities}. It is not exhaustive,
nor exclusive. Details of the responsibilities for the various system
components % chamber construction and FE-board production and tests
will be finalized by the time of the engineering design reviews.
\begin{table*}[p]
%\vspace*{-0.4cm}
\caption{Muon project: sharing of responsibilities. For the
MWPC construction the sharing is still preliminary.}
\label{tab:responsibilities}
%\vspace{4mm}
\centering
\begin{tabular}{|l|r|}
\hline
Task & Institutes \\
\hline
{\bf MWPC} & \\
\cline{1-1}
{\bf \em Construction} & \\
%\cline{1-1}
\qquad Station M1, R3 -- R4 & Ferrara, LNF \\
\qquad Station M2 -- M4, R4 & PNPI \\
\qquad Station M5, R4 & INFN 3 \\
\qquad Stations M2 -- M3, R1 -- R2 & CERN \\
\qquad Station M2 -- M5, R3 & LNF, Ferrara \\
\qquad Station M4 -- M5, R1 -- R2 & LNF, Ferrara \\
\cline{1-1}
{\bf \em Testing} & \\
\qquad All stations & CERN, LNF, PNPI, Roma I, Roma II \\
\hline
{\bf Inner part of station M1:} & \\
\qquad Construction and testing & To be decided \\
\hline
{\bf Readout electronics:} & \\
\qquad CARIOCA chip design, production and testing & CERN \\
\qquad DIALOG chip design, production and testing & Cagliari \\
\qquad SYNC chip design, production and testing & Cagliari \\
\qquad FE-boards (production and testing)& CBPF, Roma I, Potenza \\
\qquad IB boards, design, production and testing & LNF \\
\qquad SB boards, design, production and testing & Roma I \\
\qquad ODE boards, design, production and testing & Cagliari, LNF \\
\hline
{\bf Services:} & \\
\qquad Gas systems design and construction & CERN, INFN \\
\qquad Monitoring, Control (ECS) & CBPF, Roma I \\
\hline
{\bf Experimental area infrastructure:} & \\
\qquad Chamber support structures & CERN, LNF \\
\qquad Muon filter support structures & CERN \\
\qquad Muon filter installation & CERN \\
\hline
\end{tabular}
\end{table*}
\cleardoublepage
%
%
\begin{thebibliography}{99}
\bibitem{bib:muontdr} LHCb Collaboration,
{\em LHCb Muon System Technical Design Report},
CERN/LHCC 2001-010, 2001.
\bibitem{bib:passaleva} G. Passaleva et al., Proceedings of the
``International Workshop on Aging Phenomena in Gaseous Detectors'', DESY,
October 2001.
\bibitem{bib:ganis} G. Ganis et al., Proceedings of the
``VI Workshop on RPC and Related Detectors'', Coimbra,
November 2001. % {\tt physics/0210045}
\bibitem{bib:rpcnim1} G. Carboni et al., ``A model for RPC
detectors operating at high rate'', LHCb-Muon 2002-069 (to be published
in \NIM A).
\bibitem{bib:seva} V. Souvorov et al., Proceedings of the
``International Workshop on Aging Phenomena in Gaseous Detectors'', DESY,
October 2001.
\bibitem{bib:note041} E. Aslanides et al., ``Performance of the
muon trigger with a realistic simulation'', LHCb 2002-041.
\bibitem{bib:M3R1_note_in_preparation} B. Schmidt et al.,
``Results from a 2 mm pitch MWPC prototype for the
LHCb Muon System'', LHCb-Muon 2003-002.
\bibitem{bib:gem} G. Bencivenni et al., \NIM~A 494 (2002) 156, and
references therein.
For a complete list see
{\tt http://www.lnf.infn.it/esperimenti/lhcb/gem}.
\bibitem{bib:amwpc} C. Lippmann et al., ``Results from a MWPC prototype
for M1R1 of the LHCb Muon System'',
LHCb-Muon 2003-001
\bibitem{bib:casaccia} S. Baccaro, A. Festinesi, B. Borgia, ``Gamma and
neutron irradiation facilities at ENEA-Casaccia Center (Rome)'',
Report CERN-CMS/TN, 95-192 (RADH), 1995.
\end{thebibliography}
%
\end{sloppy}
%
\end{document}
A tentative distribution of production responsibilities
is shown in Table \ref{tab:sharing}. The chamber for the two
inner regions of M1 are not yet assigned. It has to be remarked
that the number of chambers is not a direct measure of
the construction effort, since chambers of
large size are somewhat more demanding than the small ones.
Also, the chambers of M1 having only two layers,
they are faster to produce than the other four-layer chambers.
\begin{table*}[hb]
%\vspace*{-0.4cm}
\caption{Preliminary sharing of MWPC construction
among the production centers. The quantity of each chamber
type is indicated. (*) denotes 2-layer chambers, with Nomex honeycomb
panels. The remaining ones are all 4-layers with Esadur
120 panels (see text).}
\vspace*{0.5\baselineskip}
\label{tab:sharing}
%\vspace{4mm}
\centering
\renewcommand{\arraystretch}{1.5}
\begin{tabular}{|l|c|c|c|c|c|}
\hline
R4 & 192 (*) & 192 & 192 & 192 & 192 \\
& Ferrara & PNPI 1 & PNPI 1 & PNPI 2 & INFN 3 \\
\hline
R3 & 48 (*) & 48 & 48 & 48 & 48 \\
& Ferrara & LNF & LNF & LNF & LNF \\
\hline
R2 & -- & CERN & CERN & LNF & LNF \\
& 24 & 24 & 24 & 24 & 24 \\
\hline
R1 & -- & CERN & CERN & LNF & LNF \\
& 12 & 12 & 12 & 12 & 12 \\
\hline
& M1 & M2 & M3 & M4 & M5 \\
\hline
\end{tabular}
\end{table*}
According to this plan, PNPI had the responsibility of the 384
chambers of R4 in stations M2,M3 (stations with wire readout
only). CERN was assigned the 72 chambers with mixed (cathode-pad and wire)
readout of M2--M3,R1--R2. The remaining 408 chambers were assigned
to Ferrara and LNF.
\begin{figure}[hbt]
\begin{center}
\epsfig{file=eps/rpcA_eff_rate.eps,width=0.45\textwidth}
\epsfig{file=eps/rpcB_eff_rate.eps,width=0.45\textwidth}
\caption{RPC efficiency at several voltages as a function
of the particle flux density $\Phi$. Measured in July 2002,
at an average temperature of $26 \oC$.}
\label{fig:ratecap}
\end{center}
%\vspace*{-0.3cm}
\end{figure}