ATLAS Upgrade ID Barrel: Services around - PowerPoint PPT Presentation

1 / 11
About This Presentation
Title:

ATLAS Upgrade ID Barrel: Services around

Description:

According to the drawing Preparation outer cylinder volume reservation' ... combination of services from barrel, pixel discs 1-3 and end-cap discs 4-8. ... – PowerPoint PPT presentation

Number of Views:12
Avg rating:3.0/5.0
Slides: 12
Provided by: ucap5
Category:

less

Transcript and Presenter's Notes

Title: ATLAS Upgrade ID Barrel: Services around


1
ATLAS Upgrade ID Barrel Services around outer
cylinder

According to the drawing Preparation outer
cylinder volume reservation (C.Bault 1.07.08),
the inner surface of the outer cylinder is at
R1033. Given this radius, it is possible to
accommodate a continuous layer of cooling pipes
from the SCT barrel with unmanifolded alternating
exhaust and input pipes or grouped exhaust/input
pipes in a single layer IF the number of staves
is still no more than the original 28, 36, 44,
56 and 72 ie 236 (total for one side, with each
stave having one input and one output pipe.
Although cooling pipes in this layout would not
occupy more than 10mm in R, the structure/s that
support them and protect them from being bashed
during Installation (eg a SERVICES MODULE) would
have to be substantial and probably take up an
additional 20mm in R without including possible
insulation. If power cables are used (rather than
power tapes), it is possible that they could fit
into another 10mm envelope in R, if evenly spaced
and not bunched. The space envelope allowed for
on the drawing between the inner surface of the
outer cylinder is 63mm (R1033 R970). This is
insufficient for the combination of services from
barrel, pixel discs 1-3 and end-cap discs
4-8. IF SERVICES MODULES WERE USED, HOW WOULD
THEY BE INSTALLED?

1
TJF 24.02.09 updated 12.03.09
2
SERVICES MODULES containing cooling pipes, power
and opto cables or Optofibres and DCS cables
how are they installed? Step A
REAL ID BARREL with staves
Outer cylinder
Services modules
separate ends of services with connectors
Services jig Barrel end outer cylinder section.
ID Barrel JIG in same cleanroom or adjacent,
visible area
ID (IT) barrel in cleanroom
2
TJF
3
SERVICES MODULES how are they installed? Step
B
After fitting on jig (pipe bending etc) and
securing in services module shell, remove first
module from jig.
REAL BARREL with staves
Outer cylinder
Services modules
Services jig Barrel end outer cylinder section.
3
4
SERVICES MODULES how are they installed? Step
C
then fit to outer cylinder and connect up
cables and pipes, except for cables which may
overlap those of the adjacent services module.
REAL BARREL with staves
Outer cylinder
Services modules
Services jig Barrel end outer cylinder section.
4
5
SERVICES MODULES how are they installed? Step
D
Next, fit a new services module adjacent to the
second services module to be fitted on the real
barrel..
REAL BARREL with staves
Outer cylinder
new services module
Services jig Barrel end outer cylinder section.
5
6
SERVICES MODULES how are they installed? Step
E
Make sure the services from adjacent modules
fit, then remove the module to be installed on
the real barrel and fit.
REAL BARREL with staves
Outer cylinder
new services module
Services jig Barrel end outer cylinder section.
6
7
IS THIS PROCESS USING SERVICES
MODULES FEASIBLE? Each services module contains
a few of each type of services, some rigid and
formed such as cooling pipes and some very
fragile such as readout micro cables or
fibres. The filled services module would be an
unwieldy and heavy object to move about with
metal edges a hazard to thin cables. It could
also bend and twist over its length. There would
be services sets with ends of different lengths
very difficult to handle, first when assembling
on a jig and then when installing and connecting,
even if only transporting from one part of the
cleanroom to another. Some services would have to
be embedded under others when packing the
services module so that inspection would be
difficult if such a cable or pipe failed during
testing in situ after connection. Long pipe
lengths inside the Services Module and extending
to end barrel connections would certainly place
stress on the barrel staves and would thus
require substantial bracketry on the barrel end
to take the stress off the stave structure. Power
cables inside sleeving would place even more
stress on barrel end connectors and the
stave. How strong does the stave with integral
connectors have to be at the barrel end? Very! As
we observed on the SCT barrel, a spider of 5 or
6 cooling exhaust pipes, even with bellows to
allow flexibility, was difficult to connect up to
the barrel. In the end bellows could not be used
and each exhaust pipe was connected up separately
as were the input pipes. A 32 way layout in phi
would mean that Services Modules would take up
more space if they are in the form of channels or
boxes with lids.
7
8
CONCLUSION Probably unfeasible as an all
services module given only limited
access handling and connecting a bunch of
services all of one type is difficult enough
without combining several types. Cooling pipes
with connectors would be very difficult to
control when in a mixed bunch and would obstruct
access to other connectors. The ID barrel would
have to be rotated for each SM installation.
A 32 way layout still means there would be
more than 7 staves served by one Services Module
which is a lot, probably 6 too many! POSSIBLE
SOLUTIONS Connect up each type of service
separately, finishing and testing one type
before starting another. Bunch small sets of the
same type which could be packed in a small
module or channel with others. Cooling could be
bunched but not packed and could occupy dedicated
space envelopes without intertwining with cables.
WHAT TO DO NEXT? Assume all new
services from ID out to PP2 but probably best to
also assume same or very similar position in phi
to old services ie based around 45 degs for
cooling with readout either side of this and
power cables spaced approx. evenly in quadrants
allowing space for cooling and (im)possibly big
opto boards. Assume a SINGLE baseline LAYOUT for
ID which may well change many times!
8
9
MAPPING for BARREL services based on the
following layout for ONE end ACTIVE TOTAL
COOLING COOLING POWER READOUT
DCS LAYER R STAVES OUTPUT INPUT
CABLES ORDER
R 380 28 28 28
28 x 2 28 x 2
bunch 5th
R 490 36 36 36
36 x 2 36 x2
bunch 4th
44 44 44
44 x 2 44 x 2 bunch
3rd
R 600
R 750 56 56 56
56 x 2 56 x 2
bunch 2nd
R 950 72 72 72
72 x 2 72 x 2
bunch 1st
Using the only known baseline dimensions and
stave numbers
9
10
MAPPING for BARREL services based on the
following layout for ONE end ONE QUADRANT eg
side A , QUADRANT A1 with cooling centred on 45
degs.
POWER CABLES one stave, (half length of barrel)
in one sleeved bunch. 59 sleeved bunches from one
quadrant, going to 20 power services
modules arranged in 4 sets of 5 either side of
COOLING centred on 45 degs and READOUT centred on
22.5 degs
READOUT CABLES or FIBRES one stave, (half
length of barrel) in one sleeved bunch. 59
sleeved bunches from one quadrant, going to 4
R/O services modules arranged in 2 sets of 30
either side of COOLING centres on 22.5 degs. NB
this is assuming opto board is mounted
somewhere on the stave, rather than being remote
from the detector. IF remote then distribution of
cables will need to be spaced out rather than
bunched.
OUTPUT COOLING one stave, (half length of
barrel), one pipe with no manifolding. ie 59
pipes directed to 2 layered bunch centred at 45
degs
INPUT COOLING one stave, (half length of
barrel), one pipe with no manifolding. ie 59
pipes directed to 2 layered bunch centred at 45
degs either side of output cooling
10
11
READOUT cables or fibres
spare
BARREL MAPPING
Services fan-in to main pathways from area
of octant, all layers via the shortest possible
route
COOLING output
COOLING input
POWER and DCS cables
space reserved for rails etc
spare
11
Write a Comment
User Comments (0)
About PowerShow.com