Structure Update Installation

1
Structure UpdateInstallation Building
UpdateRevisions OutlinedCosts Revisited(since
given to Gina)

• Jeff Nelson
• Fermilab

2

• Reminder of current scintillator detector
  concepts and alternatives
• Whats different in liquid vs solid?
• Installation Update
• Value engineering summary
• Cost summaries

3
Basic Object

• For either scintillator use same structure
• Liquid needs 3cm gaps instead of 1.7cm gaps
• Simple adjustment in structure
• Alternating view every 8
• Strips 4cm wide

180 m
readout
14.5 m
readout
readout
14.5 m
14.5 m
885 planes
4
A detector unit

• Units made from natural OSB dimensions
• 8 x 48ft x 8ft
• 2 or 3 pieces of wood in 8 layers
• Two 30-strip modules which are 4ft x 48ft
• Encased in wood and captured with clips
• Step in z every 8ft

End view of single unit
5
Building up the detector
Next one here
6
Structural Properties of a Stack

• Bob Wands Bob Wands using an ANSYS FEA model
  tested extreme cases for some quick answers
• Horizontal with 2 or 3 supports along entire
  length

7
Deflection and Stress in an Extreme Case
Maximum deflection is 0.9 Maximum stress is 562
lbs
8
Bobs Conclusions

• There are no particular structural issues
• Stresses and loads low (Vics conclusion too)
• Bulking not an issue
• JKN Comment FEA results say unsupported out
  vertical is at less then 30 degrees is fine
• He see no need for any bookends other that to
  start construction
• Expansions is global if bottom treated with low
  enough coefficient of friction
• Only question is do you want to do something to
  control bulk expansions for maintaining alignment
  for physics
• Thought container engineering more interesting
  problem

9
Installation Update

• Model unchanged except
• Stacks make in industrial building (e.g. MINOS
  surface building)
• Gets your floor space a lot cheaper at grate
• Nathaniel stack production layout needs 5X
  surface building
• Earl scales it at 3M
• Revisited number of FTE
• Some double counting
• 34 -gt 28 (plus support staff)

10
Putting up the units

• Rigged using rotating vacuum fixture
• Just like every piece of MINOS steel
• Bob Wands verified that a stacks weight is well
  within specifications for commercially available
  devices with cups specified with for porous
  surfaces
• Bob also says hes convinced the stresses within
  are stack are safety below levels that could lead
  to structure failure in the material during a
  light

11
Main changes of WRT Estimate

• APDs modern PPD electronics ( and channel
  count go down 10m -gt 14.5m)
• U loop reduces fiber costs increases uniformity
• Step scintillator at near end (15 reduction in
  volume)
• Factory managers update shows machines and
  throughput OK machine costs
• Earl updated building scaling Staging facility
• Progression (Detector without buildings)
• 119M for M64 (L10m) no changes from MINOS
  (Brajesh)
• 98M for APD (L14m _at_ July talk)
• 93M more detailed costs remove GA double
  counting (now)
• 163/m2 including electronics

12
Solid Scintillator Cost Summary
13
Liquid Version

• Based on MINOS RD summarized in P.Border et al
  NIMA 463194,2001
• Documents a structural and mechanical design (to
  ME masters theses on these issues)
• Liquid added after erection, light, and leak
  tests
• Extensive studies of the effects of aging on
  optical and mechanical interactions of materials

14
Liquid Scintillator Cost Summary
15
Solid Scintillator Cost Summary
16
Some personal extras
17
Want more light?

• 0.8mm -gt 1.0mm fibers
• Gives 36 more light for 5M

18
RPC vs Solid Scintillator

• 20 in FOM -gt 45 in reach (using document values
  for both)
• Pulse height info vs technique ???
• Costs are indiscernible for RPC vs Solid with
  current costing uncertainties (esp. with
  contingency)
• Very low risk for the solid design
• Scaling production factor of 5 vs factor of 100
  for RPC
• Direct Fermilab experience in scintillator
  production APDs
• MINOS used many institutions for production and
  same would be true for Off-Axis
• Comparable cost, better performance, lower risk

19
Solid vs Liquid

• Active components 33 cheaper (?20M)
• Would guess that increased contingency would eat
  this difference at this time
• Would propose solid as baseline and liquid as a
  value engineering option
• Look at reoptimization of solid ? could decrease