ATLAS Placement and Alignments

1
ATLAS Placement and Alignments
David Lissauer / BNL
Experimental Hall is very small compare to ATLAS
Size. Ship in a Bottle Need to be aligned to
the beam line to mm and many of large Many
envelopes have less than 20mm stay clear
areasbetween moving Objects.
Every Silly mm Counts!!!
Result of many discussions over the past year
with Survey Team. LHC Machine and Civil
Engineering. Toroid and Muon Placement. Calorimet
ers LAr/Tile. Inner Detector. Beam Pipe. EC
Toroid, Shielding Physics.
2
ATLAS Placement Considerations
Items under active consideration Floor
Stability -- Monitoring on going HO/HS
Structures -- Accepted Feet Rail Placement --
Feet Accepted/Final Rail Aug 05. Barrel Toroid
Placement -- Ongoing Barrel Calorimeters/Solenoid
-- Ongoing EC Calorimeter -- Under study ID
Placement -- Preliminary agreement/Ongoing Beam
Pipe Placement -- Preliminary agreement/Ongoing Sm
all Wheel -- Under Study EC Toroid -- Under
Study Big Wheels -- Under Study EO Chambers --
Under Study JF Shielding -- Under Study
3
LHC Nominal Beam and IP.

• Nominal Beam Line
• Close orbit calculations uncertainty in the
  Nominal Beam Line (Geometrical line) is 3 mm.
  (I.e. actual beam vs. nominal beam line)
• Uncertainty might be reduced by the time Barrel
  Calorimeter is installed (05)
• Beam Adjustments
• Immediate lt 1 mm by changing the magnetic
  field in the last magnet.
• Short term 1 mm by adjusting the Jacks under
  the last triplet.
• Long term Re-align a string of magnets along
  the tunnel. Adjustments could be as much as
  10-20 m.m. (K. Potter et al.)
• Discussions with Machine, Survey (K. Potter et
  al.)
• Machine will be able to align itself to the ATLAS
  experiment to the level of 10-20 mm during a long
  shut down. The machine does not expect to be able
  to do this more than once every 3 Years.

4
Placement Strategy

• Stages I Surface assemblies.
• Placement of assemblies on the surface.
  Components need to be aligned and adjusted
  relative each other on the surface.
• E.g Barrel Cryostat Assembly EM calorimeter,
  Solenoid magnet and Cryostat.
• ID Barrel Assembly Included TRT , SCT and
  Pixel Tube.
• Stage II Cavern assemblies.
• Placement of Assemblies in the Cavern. Components
  need to be aligned and adjusted relative each
  other (Geometrical Beam line at the same time).
• E.g Barrel Cal Assembly Tile and Barrel
  Cryostat Assembly are aligned relative to each
  other and the nominal beam line.
• Stage III Final placement.
• Placement of assemblies relative to the nominal
  beam. (Usually)
• E.g Barrel Can Moves to final position and
  aligned relative to the Nominal Beam.
• Barrel Toroid.
• Due to the Floor movement we need to decide if we
  place the object at nominal position of Off
  nominal position expecting floor movement.

TC/Survey/Systems follow each of the
assemblies Working groups follow Placement
specifications, and survey needs.
5
ATLAS Exp. Hall Floor Movement

• Civil Engineering expectation
• The existing calculation were done with
  simplifying assumptions.
• Too large of a grid
• Uniform Load on the floor
• Floor Settlement -2 mm from the time the
  concrete if poured and the time ATLAS got
  possession of the experimental hall. (No
  measurements to verify this)
• ATLAS weight -5.5 mm (Total 7.5 mm) due to the
  weight of the exp. Over 6 month. Adiabatic
  as the weight is added.
• Floor upward lift due to Hydrostatic Pressure
• 1 mm / year due to hydrostatic pressure.
  Up To
• 20 mm over 20 years.

6
GRID Monitoring Floor Movement
A GRID of Survey Rivets has been installed at
strategic position. Monitoring of the floor
stability is ongoing from July/August of
03. Expect to re-measure position every 2-6
month. Accuracy 0.3 mm datum point deep
reference in the tunnel (same for machine and
experiment geometry)

7
Hydrostatic Monitoring

Hydrostatic System has been operating since June
04. (See TMB report by Hélène MAINAUD DURAND
TS/SU/MTI) Time span is small but measurements
indicate movement of the floor. Higher on USA
side than US side. (For details see presentation
by Jean-Christophe Gayde) Relative height
accuracy better than 100 Microns. Absolute
(relative to beam line 0.3 mm Sigma)
8
Hydrostatic data quality

Note Scale In Microns!!
9
Floor Stability 8/03 ? 9/04.
Precision (1 sigma) of the dZ 0.2 mm
10
Floor and Bedplate Stability 3D Model
From 25 June 04 to 08 Sept 04
Floor Movement uneven. Possible combination of
Hydrostatics lift and sage due to weight. Need
more time dependent data.
11
Floor Movement Best Estimate
Settlement Due to Cement Contraction before 1st
meas.
3 MM Settlement Due to ATLAS Weight.
0.4 MM/year lift Due to hydrostatic pressure
Summary of the pessimistic prediction for the
floor movements.
12
Feet/ Rails Installed.

Rail Have been aligned and removed for Toroid
installation. Final alignment after Toroid is
finished.
Feet Have been placed in position. All within
Envelopes.

Bedplates Have been placed in position.
13
Toroid Placement
Center of Toroid should co-inside with the
Nominal Beam Line.
Assemble as Ellipse. 20mm move due to Toroid
weight. 10 mm due to services and Muon chambers.
Position of First two Coils Fixes the center of
the Toroid.
14
Positioning of first 2 Coils

First Coils will fix the center of the Toroid.
Physics Center of Toroid should be on the
nominal beam line. But can be as much as 20-30
mm off without affecting Muon trigger/Physics
significantly. Envelopes- require that the
Toroid be within 10 mm envelope of nominal
position.

We are more sensitive to down ward fluctuation
than to Upward fluctuation.
15
Summary Conclusions

• Toroid Assembly Oval ? Circle 0. /- 3.5 mm
• Five Years of Floor movements 3. /-
  3.0 mm
• Floor Sag due to weight -3.
  /- 2.0 mm
• Placement Accuracy 0. /- 2.0 mm
• Total 0. /- 10.5 mm

Proposal place the Toroid Center at 5mm
from Nominal Beam Line. Final Decision Week of
October 18th after latest input from Survey.
16
Barrel Calorimeter/Solenoid

• EM Calorimeter/Cryostat
• The EM Cal was placed so that it is below the
• Cryostat IWV by 4 mm.
• Solenoid /Cryostat
• The Solenoid was placed so that it is below the
• Cryostat IWV by 2 mm.
• Relative Placement
• of the EM calorimeter, Solenoid and the Cryostat
  IWV
• are now fixed and can not be changed.

17
Barrel Calorimeter/Solenoid

• Tile Calorimeter and Cryostat assembly is being
  done at Z13.
• After LAr if placed on the Tile on the Truck the
  relative position of the Tile/Cryostat
  IWV/Solenoid and EM position are fixed.
• The Tile Calorimeter/Cryostat assembly will than
  be moved to Z0.
• The Barrel Calorimeter can be adjusted as a unit
  to the Nominal Beam line.

18
Barrel Calorimeter/Solenoid
We will align the Solenoid Axis with relation to
the nominal Beam Axis. Resulting in the above
relative loactions.
19
ID - Installation

• TRT/SCT
• Are assembled on the Surface as a unit.
• The TRT and SCT Axis have to be co-linear. There
  is a small gap between the TRT and SCT but it can
  not be reduced as it is part of the thermal
  shield and the construction tolerance.
• Pixel Tube Installation
• Pixel Tube is installed first in the SCT/TRT
  module. On the Surface.
• Pixel Tube can be installed off center by as
  much as 2-3 mm.

SCT /TRT Axis is Collinear. No Possibility for
adjustments of SCT relative to TRT.
Pixel Tube can be inserted off axis by as much as
2-3 mm. Once inserted they can not be adjusted.
20
ID - Installation

• SCT/TRT/Pixel Tube
• The SCT/TRT/Pixel Tube are installed as a unit.
• No relative adjustments of the individual
  components in possible in Situ.
• The Module is support are adjusted on the surface
  so that the SCT/TRT axis will co inside with
  Barrel Cal IWV Axis.
• At this stage there is some adjustment in Situ
  possible (Max /- 4 mm) but, ID aim will be to
  bring the Barrel IWV- TRT/SCT axis to co inside.
• This requirement that the TRT/SCT axis coincide
  comes from the very tight space and the needed
  symmetry for the TRT/SCT Services.

The ID requires that the TRT/SCT axis co inside
with the IWV axis dues to the ID Services needs.
ATLAS thus agrees that the TRT/SCT Axis will be
placed 2 mm off the nominal Beam Axis and off the
Solenoid Axis.
21
ID - Installation

• Pixel Insertion
• The Pixel axis will co inside with the Pixel Tube
  Axis.
• The Pixel Tube when placed (above ground) can be
  places so that its axis and the TRT/SCT Axis are
  off by 2-3 mm.

• There are thus two options for the Pixel
  installation.
• Place Tube Axis with SCT/TRT.
• Place Tube Axis off by 2 mm to co-inside with
  Nominal beam.

The Present baseline is to install the Pixel to
Co-inside with the Nominal Beam Line
22
Summary of the Placement
23
Barrel Muon System

• The Muon Support structure
• Rail System. First adjustments above ground
  (Local Coordinate system)
• Final Adjustments in Situ before chambers are
  mounted.
• Each Tower points to the Intersection region.
• Individual Alignment/Placements for each chamber
  assembly needs are being defined.
• MDTs RPCs subassembly done on the Surface.
• Once placed on the Rails no adjustments is
  possible.
• Chamber Assembly Placement.
• The specification on chamber placement needs to
  be defined. Survey Target position etc.
• Alignment System.

24
Barrel Muon System
y
y
x
Reference plates for the alignment system
Reference points on the Toroid structure
x
Local reference frame of the small sector
Chamber rails
Local reference frame of the large sector
25
Conclusions

• Floor Monitoring systems are in place.
• ATLAS Placement has started. ATLAS has installed
  Feet Rails, HS and HO structures. (Rails can
  still be readjusted )
• This already has resulted in small change to
  detector envelopes and final placements. (e.g. HO
  chambers moved by 20 mm in Z.)
• Some Subassemblies on the Surface already
  completer Barrel EM/Solenoid/IWV completed,
  EC-C.
• Agreement on Procedure and hierarchy for Barrel
  Calorimeter/Solenoid/ID placement.
• Toroid Placement will start later this month.

Need to follow and re-optimize placement of
detectors continuously.