Title: AGENDA
1Features and Mysteries of BOL chambers
- AGENDA
- Xray tomograph -BOL0 results (Martin)
- BOL0 and BOL2 Features (Graziano)
- Discussion (All)
2BOL0 - Hypothesis
- OBSERVATIONS BOL0
- Y-pitch between inner wire-layers 20um too large
- Y-Pitch between outer wire-layers only slightly
too large - Y-Pitch between multilayers 75um too small
- Hypothesis
- Glue shrimp between tubes and spacer?
- (2x500um glue gap)
- --gt measured lt5um effect for 1mm gap
- Xplate deforms during gluing. In particular the
flange - --gt presented today
- Additionally
- Re-analysis of BOL0 construction RASNIK readings
cumulative effects lead to better agreement!
s
3Floppy Xplate wrt tubes
Rotation of unknown relative size
_at_longbeam
Longbeam constraints body Xplate locally
Tube-length bends body
Tube direction bends flanges
Away from longbeam
s
4Sag Adjustment
Rotation of unknown relative size
Longbeam constraints body Xplate locally
Tube-length constraints flanges
Tube direction bends flanges
Away from longbeam flange bends differently!
BOL0 is scanned with sag-adjustment (partially)
on! (Longbeam sag is assymetric in BOL0)
When we are optimistic deformation is symmetric
for up/down ML.
s
5Sag Compensation
Rotation of unknown relative size
Longbeam constraints body Xplate locally
Tube-length constraints flanges
Tube direction bends flanges (temporary)
s
6XP deformation
During BOL2 construction, the flange of the HV
xplate was measured with micro-gauges LBF
Longbeam- Flange BF Body- Flange
For now, BOL2 is still equipped with gauges,
located on the inner side of the HV-xplate
s
7Gluing BOL-0
- Stability Xplates during gluing
Layer 1
Yrasnik2 x Sag
Layer 1 dSag7um
Layer 2
Start glue
Layer 3
Layer 2 dSag10um
- The sag of the xplates in layer 1 and 2 change in
time. - Layer 3,4,5,6 Sag stable
s
8Sag Compensation
BOL2
Layer 1
Start -15/2 um
- During gluing of
- layer 1 and 2, significant changes in xplate sag
were observed - In-stable reduction valves?
- Effect of glue viscosity?
End -38/2 um
Layer 2
Start 30/2 um End 50/2 um
1 deg
s
9Conclusions
- OBSERVATIONS BOL0
- Y-pitch between inner wire-layers 20um to large
- Y-Pitch between outer wire-layers only slightly
to large - Effects largely explained by RASNIK measurements
during construction. - Flange deformation of order 10um could be
baked in. - Y-Pitch between multilayers 75um to small
- Flange deformations of this order are certainly
observed. However these deformations appear
symmetric and elastic. - Needs more study -gt effects of kinematic mounts
- direct -in situ- measurement dY of Frascati comb
and towers
s
10Test _at_CERN inTomograph (BOL0)
Analysis of Martin Woudstra
Wire positions RMS 16mm
s
11Parallelogram-ness
For stability bare spacer
Checks distance wrt Comb _at_ RO side
BOL
RO
rail
s
12Landing-Tool
- Problems with lowering gluing tube layers on tube
layers due to glue viscosity - Landing tool pre-aligns chamber and ensures flat
landing in glue.
BOL
Tower
Effective Easy
s
13Cosmic Ray Stand
s