Radiator Procurement

1
Main Detector Status (July 06)
D.J. Mack (TJNAF) Qweak Collaboration
Meeting at JLab July 26, 2006
2

• Overview
• Quartz bar QA
• Glue Studies
• Summary

3
Overview
Have spent 85 of our 468.5K budget. All custom
PMTs and magnetic shields are at JLab. All
TRIUMF preamplifiers for the main detector are at
JLab. (Gains for the
Lumi/beamline PAs need to be determined
soon.) Delivery of prototype TRIUMF 500 KHz
sampling ADC and JLab 250 MHz flash ADC are
expected in the coming quarter. Lots of testing
needed before we go into production (by TRIUMF
and C. Capuano WM) Quartz bar shipments are
nearing completion.
QA on first 10 finds only
1-2 questionable bars. Simulations are guiding
us in what is (not) acceptable. Lightguides are
out for bid. Collaborators answered our plea for
people-power (thanks!). Significant progress on
detector this summer with the help of 3
undergraduates (one DOE, one NSERC, and one 12
GeV) Quartz Bar QA, Bar mechanical design, Glue
studies, Scintillation and Luminescence tests,
Low Gain voltage divider tests.
4
Expenditures to Date
(From Gregs synposis in a 27June06 email) WBS1
has spent 85 or 396.4K of our 468.5K budget,
leaving 72.2K for everything else
5
Estimated Remaining Expenditures
With the remaining 72.2K we must in principle
buy
I recently realized that WBS1 is being charged
for all the beamline electronics which were
forgotten in the original Qweak budget. If that
unfairness is corrected, WBS1 wont have to build
a support structure out of 2x4s.
6
Bar Quality Assurance(work by Elliott and Team
Johnson)
7
Inspection

• Vendor QA sheets were collected and typed into an
  Excel spreadsheet.
• The 10 available bars were visually inspected
  with penlight, black background, and Elliot
  Johnsons young eyes.
• Scratches, chips, and scuff marks were
  identified.
• (These usually agree well with manufacturers
  list, but our eyes were younger and motivated
  motivated, so we found a few more defects.)

Some bars had bevels which were clearly wider
than specification. These 4 bars were flagged for
detailed measurement.
8
Vendor Length Measurements

• The length appears to meet specification.
• But as long as beam envelope stays well inside
  2000mm, length isnt critical.
• The bars now exist make sure the
  target/collimator system is built to match!

9
Vendor Width Measurements

• The width appears to meet specification.
• Bar location in the dispersive direction will be
  set by the outer edge which sits in the radiative
  tail. (Puttting the extra -1 mm of acceptance in
  the superelastic region should minimize
  corrections.)

10
Vendor Thickness Measurements

• The thickness appears to meet specifications.
• Variations in photoelectron number from bar to
  bar of order 10 can be expected. (Negligible
  impact on excess noise.)

11
Dimensional Measurements at JLab

• St. Gobain subsidiary in England provided simple
  measurements of length, width, and thickness for
  each bar, which is sufficient only to
  characterize an ideal rectangular solid.
• We wanted
• An independent check.
• Enough redundancy to bound skew/curvature.
• Surface flatness measurements.
• We were completely stumped how to do this. (Its
  not something you can do with a small-jawed set
  of calipers. )
• Greg Smith suggested the JLab CMM.

Mitutuyo CMM
Ruby probe
12
Mitutuyo CMM Setup
Position readback is real-time, so its easy to
measure points on a pre-planned grid. Some
programming was needed to label the output data
in a readable format.
For this initial study of a single bar, we
measured 3 points on the 4 smallest faces, 9
points on one of the large faces. It takes less
than one hour per bar.
Bar resting on granite table. Table is ground
flat to 0.1 mil (2.5 microns). Aluminized mylar
prevents scratching.
13
Dimensional Measurements Cross-check
The average of JLab CMM measurements agrees well
with St. Gobain for Length/Width/Thickness, but
the JLab measurements are consistently about 250
microns smaller. The surface flatness on the
single large face we examined appears to meet our
specification of - 200 microns. (Admittedly this
one-sided technique is a little ambiguous. Needs
more thought.) Jim Dahlberg and Chris Gould in
the Survey Alignment group have graciously
agreed to do CMM measurements on all our bars,
but it doesnt look like theyll do more before
the summer is over.
Thickness contours
14
Precision Bevel Measurements with WM Wire
Scanner
We needed an optical technique to check the width
of bevels that appeared to exceed specification.
The Qweak bars are too large for most travelling
microscopes. Fortunately, our WM collaborators
are experts on this. With the assistance of Team
Grimm, bevels could be accurately measured.
15
Bevel Measurement Findings
Indeed, in the worst bars, up to 25 of the total
bevel area can average 1.5 mm width. M. Gericke
simulated the impact on pe collection, and it
appears to cause only a modest reduction. We
will probably accept 6 of 6 clean bars 3 of 4
wider bevel bars And return 1 of the wider
bevel bars to have an edge polishing defect
corrected.
quartz bevel air
Lighting adjusted to enhance bevel. (plot
from cool WM software)
16
Gluing Studies(work by Patrick and Team McCarter)
17
Qweak Glue Boundary Conditions

• Cerenkov radiation spectral density is
  proportional to 1/?2 . All our pe calculations
  assume a cutoff wavelength of 250 nm which is
  well in the UV.
• We cant use convenient UV curing adhesives
  since these by definition absorb UV. (So we will
  have to keep wetted joints from moving while the
  glue is curing.)
• The middle glue joint will take the same dose as
  the quartz, about 100 kRad. (But we should allow
  for up to 1 MRad dose expected if a pre-radiator
  is used.)
• Good adherence to smooth quartz surfaces and
  reasonable strength.

18
Shin-Etsu Silicones 403 and 406(Optically Clear
Silicone Elastomers)
SES 403 was identified in BaBar DIRC tests as
having excellent UV transmission and high
resistance to radiation damage. We got samples
of this as well as SES 406 from Ben Ray of
Shin-Etsu USA. (n 1.405 versus the quartz
value of n 1.47-1.48) These two-component
materials are sold in large volumes as potting
materials for electronics. Thus their ability to
glue together pieces of ultra-smooth quartz
could not be taken for granted.
Patricks glue equipment and 11K test bed (an
old prototype quartz bar).
19
Observations and Potential Applications
Both materials are stunningly transparent.
(Forget RTV. Think mineral water). After finding
a reliable cleaning solvent (100 isopropyl
alcohol), we made the following observations in
glue-on-quartz tests
SES 403 Due to poor strength and weak
adherence, SES 403 is a good candidate for a
reversible optical connection for our PMTs to
the lightguides. SES 406 SES 406 is a good
candidate for joining the quartz bars, as well as
the bars to the lightguides.
20
Upcoming Transparency Studies

• We still need to determine whether these glues
  meet our UV transparency requirements, especially
  after 100kRad-1MRad dose in the case of SES 406.
• The quartz slides are finally cut and being
  shipped thanks to Brian Kross (JLab).
• Carl Zorn (JLab) has reassembled his
  spectro-photometer setup so we can measure
  transmission in the UV down to 230 nm.
• We confirmed that Scott Lassell at NC State still
  runs a 60Co irradiation facility (about 10
  kRad/hour) which accepts small mailed samples.
  There are modest fees for time and dosimetry.

Old Qweak transmission measurements on 12.5cm
quartz with few percent uncertainties.
21
Related Main Detector Talks

• TRIUMF low noise electronics Des Ramsay
• Low Gain PMT measurements Michael Gericke

22
Recent WBS1 Related Reports

• Final Lightguide Geometry, M. Gericke Qweak
  570-v1
• Final Lightguide Specifications, D. Mack Qweak
  537-v2
• The Qweak Experiment, D. Ramsay, Qweak 571-v1
• The summer students will write up all their
  work on bar QA, low gain base prototyping, glue
  studies, etc, as technical reports
• or their travel home wont be paid.

23
Upcoming Tasks

• Testing/Fabrication
• Continue Bar QA and close out the bar order
• Cosmic tests on prototype 06-07
• Rad damage testing of glue, voltage dividers,
  etc.
• Manufacture parts for summer 07 assembly.
• Complete all 81 modules by end of summer 07.
• Test and Fabricate all 500 KHz sampling ADCs.
• Design main detector support structure.
• Background-related Measurements
• Field test of JLab 250 MHz transient digitizers
  for unbiased bkg studies
• Improve scintillation and luminescence tests
• Procure small NaI detector to measure photon
  backgrounds
• Simulations
• Q2 bias and evolution wrt rad damage,

24
Summer Student Participation

• Our U. Manitoba student, Mitchell Andersen, built
  and tested several generations of prototype bases
  this summer, worked with UV LEDs and zener
  diodes, accurately measured PMT gains with a
  current mode technique, helped measure
  nonlinearity to lt10-4, and learned how to
  body-surf at Virginia Beach.
• Our U. North Dakota student, Elliott Johnson,
  designed a prototype support rail which was built
  in the JLab machine shop, helped order a custom
  composite panel to support the glued quartz bars,
  worked with JLab and WM Qweak collaborators on
  quartz bar QA, made CAD drawings, and toured DC,
  Philadelphia, and NY city on weekends.
• Our NC AT student, Patrick McCarter, was working
  on a 12 GeV quartz hodoscope project where he
  ordered PMTs for cosmic tests, and used Monte
  Carlo ntuples to define the size of this future
  SHMS detector. But he also learned things helpful
  to Qweak wrt silicone based adhesives and how to
  measure scintillation in the quartz bars, and in
  his spare time played tennis and identified sea
  nettles the hard way at Buckroe Beach.
• Our sink-or-swim mentoring style seems to work!

25
Summary

• Main detector budget looks reasonable, although
  the lightguide cost could still surprise us.
• Thanks to student participation, JLab support
  groups, and our WM collaborators, theres been
  good progress recently on bar QA and detector
  design.
• We understand our glue candidates better now.
  Ready to do some serious glue transparency
  measurements, including rad damage and
  accelerated aging.
• Will talk about scintillation/luminescence tests
  and latest mechanical design next time.
• On track to finish all detector modules by end of
  summer 07.

26
Extras

27
Expected Performance (updated 7/18/06)
28
General Design
29
Assembly Detail