2. Thermal FEA

cypriotcamelΠολεοδομικά Έργα

29 Νοε 2013 (πριν από 3 χρόνια και 6 μήνες)

71 εμφανίσεις

Recent Mechanical Barrel Stave Development

BNL; S.
Duffin
, A.
Gordeev
,
D. Lynn
, G. Mahler: LBNL; C. Haber, M.
Gilchriese
: Yale; W. Emmett, A. Martin, P. Tipton

1.
Current Mechanical Prototypes

2.
Thermal FEA

3.
End Insertion

4.
Measurements of LBNL Stave

BNL/Yale/LBNL Update on Mechanical Barrel Stave Prototypes

Prototyping Plan/Schedule



Prototype 1. 107 mm x 350 mm composite without tubing


Completed



Prototype 2. 107 mm x 350 mm composite with tubing


Week of Oct13
-
17








Weeks of Oct24
-
Nov5



Prototype 3. 107mm x 1350 mm composite with tubing


Week of Oct 27
-
31








Week of Dec 1
-
6




Strike
-
throughs show schedule as presented at MIWG meeting early October.


Start on proto
-
type 2 delayed due to delay in getting pipe
-
foam assembly.


Decided to push back prototype 3 so we can first do some thermal tests on prototype 2.


Prototype 1 Completed

35 cm x 10.7 cm , 3mm thick honeycomb

3mm honeycomb, rms(thickness)=19 um, min
-
max=77um

Original Honeycomb Glued BN filled Hysol 9396

Magnified view of grinded honeycomb

Demonstrated our ability to grind and glue the CF honeycomb

Assembly Fixture Prototypes II and III

Bracket locations correspond to possible
support locations on barrel. Precision
located to +/
-

25 um.

Rare earth magnet
to attract steel rod

3 mm OD carbon
fiber tube for end
insertion

Steel bracket to channel magnetic
flux and hold CF tube

Steel Rod

Prototype II


We are in the process of assembling prototype II.


Tube
-
foam assembly uses stainless steel 2.77 mm OD, 255 um wall
stainless pipe


We use
poco

foam over k
-
foam as
poco

foam is more uniform in
cell size


Did some brief comparisons between thermal gels for interface,
and chose CGL do to its softness. This will be an area of future R&D
however


Will apply
Hysol

9396 epoxy for bonding between facing and
honeycomb/foam. For honeycomb, only apply at points of contact






Layup of Prototype II

Pocofoam

(top) compared to K
-
foam (bottom)

CGL Foam Interface

Prototype III Components

90/0/90 CF Facings




K13C2U, 1 meter long


Hysol 9396 with 30% BN



Facing
-
honeycomb,



CGL epoxy





Facing
-
poco foam




2 pcf carbon honeycomb



Will grind to 3
-
5 mm


Side tubes for mounting



Have 3 or 5 mm OD CF tubes


SS Tubing





2.77 mm OD, 2.26 mm ID (255 um walls)


Bus Cable





Samples provided by LBNL from






prototypes for 6 cm stave….will patch





several pieces together


Dummy Detectors




97 mm x 97 mm


Dummy Hybrids




Resistors on 99 mm x 25 mm AlN





substrate


Prototype III


Proto 3 serves will be a 1 meter long stave closer to the present baseline stave.


It is shorter than current basesline (135 cm) because of the limited length of our existing
carbon fiber.


We will mount with dummy modules for full thermal and mechanical testing and
comparison to simulation. Thermal FEA by Yale, Mechanical at BNL


It may be either a 3 mm or 5 mm thick inner core (foam, honeycomb) stave, depending
upon tests and FEA of prototype II. It is difficult to encapsulate 2.77 mm tube with poco
foam unless one goes to a thicker foam. We need to understand tradeoff.

Prototype IV



Envision making another stave ~ summer 09. Would have full length (~ 1220 mm), custom
K13D2U or other facing with low areal density, and with end closeouts (on 117 mm edges).


Before making this we will do R&D into the foam
-
pipe interface, correct CF facing layout to
maximize thermal performance

Thermal FEA

End Insertion

IDEA


Hold stave at those tube locations that were held to an accuracy of +/
-

25 um during assembly

Use 4
-
5 brackets per stave for support (corresponds to 20
-
25 cm bracket spacing)


Two options for alignment of brackets:


A.
If accuracy of mounting holes on barrel is on the order of +/
-

50 um, alignment unnecessary.



B.
Otherwise, end brackets first mounted and adjusted into position for stave to be parallel to Z
axis. Universal stave template mounts on end brackets, and is used to align middle brackets.

End Insertion
--

Bracket


Bracket arms are elastic. Hold stave through
compression.


Right arm is fiducial side and is stiffer (flexes
~ 100 um). Left arm flexes ~ 0.5
-
1mm.


Flexibility absorbs tolerances in bracket
mounting.

Carbon fiber preliminary design

Compression = 5 N, Mass = 7 grams

R.L. for 5 brackets/stave ~ 0.05%

Modified design for
stereolithographic

bracket.

Modified design for
stereolithographic

bracket.

End Insertion


Demonstration Prototype

“Installation Tool” is spare two brackets mounted on jig plate

After Installation

Partial Installation on Two brackets

Four brackets are mounted on original assembly fixture

Missing picture here

Prototype End
-
Insertion Stave Flatness/Deformation Measurement

Profile of stave is measured on original assembly fixture where it is held by fixture “V”
brackets (similar to those shown in prototyping photos). Stave is built to have low
deformation

Rotated Data

Original Data

End Insertion Stave on Stereo
-
lithographic Brackets


Measurements taken first with stave supported with 4 brackets spaced ~ 24 cm
apart, then with just two end brackets ~ 96 cm apart


With each support, each edge (~5 mm from side tube rails) of stave’s deformation
was measured (technique will be shown in section on LBNL stave).


Deformation with 4 brackets not as good as on assembly fixture, but very good
for first attempt.


2 point measurement demonstrates that stave is “floppy”, but end insertion
works fine with 4 brackets.

-0.800
-0.700
-0.600
-0.500
-0.400
-0.300
-0.200
-0.100
0.000
0.100
0.200
0.000
200.000
400.000
600.000
800.000
1000.000
1200.000
Deformation [mm]

Length Along Stave [mm]

Aluminum Stave Deformation Measurements

Edge 1, 4 Point Support
Edge 2, 4 Point Support
Edge 1, 2 Point Support
Edge 2, 2 Point Support

This method of insertion seems simple and viable as our initial prototype has
demonstrated. It is a low mass solution.



Second iteration will improve stiffness and have method to align brackets.
This will also be fabricated in stereo lithographic thermoplastic



We will aim to have an end insertion demonstration with proper flatness for
stave prototype III in aluminum and/or carbon fiber





End Insertion, Conclusions and Plan

CF Rod slightly beveled (sanded) to
permit low force installation

LBNL Stave V Measurements



We performed deformation measurements on the last LBNL barrel stave prototype

Stave properties



Facings are K13D2U, 3
-
ply



Carbon honeycomb



Stave mounted with dummy bus, dummy detectors, and
kapton

heaters (to simulate

hybrid power load)



Length approximately 35 cm



Technique was to measure stave surface profile nine times.

1.
Room temperature

2.
With coolant at ~
-
37 deg
-
C

3.
With power on heaters equivalent to ¼ Watt per readout chip, strip length ~ 3cm

4.
Bring back to room temperature, repeat steps 1
-
3, bring back to room temp., repeat 1
-
3

Each profile measurements
consists of five scans 11
mm apart. Each scan is in 1
mm steps

USB
Nitrogen filled cold
box, glass top
Lydall Chiller
T > -40 C
2 meter
x stage
Dual Keyence
displament measuring
lasers
Granite Table
Air bearings
Keyence Controller
N
2
BNL Measuring Station



Station can measure height of stave to accuracy of less than 10 um with a pair of
displacement measuring layers.



lasers move in stave direction in increments as small as 10 um, but we typically
use 1
-
10 mm for a stave


Chiller provides coolant down to close to
-
40 deg

Fixed Support,
no rotation
Rolling Support,
no rotation
Stave Support for Measurements

Support intended to provide cantilevered support at each end and allow expansion in Z


LBNL Stave Prototype

3
-
D Plot of Stave Using Measured Values

CF Facing

Silicon

Heater

Coolant ~
-
37 deg
-
C, no power load

Height [mm]

Height [mm]

Position Along Stave[mm]

270 mm

270 mm

-0.020
-0.010
0.000
0.010
0.020
0.030
0.040
1
91
181
Series1
Series2
Series3
Series4
Series5
-0.020
-0.010
0.000
0.010
0.020
0.030
0.040
1
91
181
Series1
Series2
Series3
Series4
Series5
-0.020
-0.010
0.000
0.010
0.020
0.030
0.040
1
91
181
Series1
Series2
Series3
Series4
Series5
-0.020
-0.010
0.000
0.010
0.020
0.030
0.040
1
91
181
Series1
Series2
Series3
Series4
Series5
Difference Plots

Room Temp 1
-

Chilled 1 Profile

Room Temp 1
-

Chilled w/power 1 Profile

Room Temp 1


Room Temp 2

Room Temp 1


Room Temp 3

Difference [mm]

Distance Along Stave [mm]

Difference Plots

Chilled 1
-

Chilled 2

Difference [mm]

Distance Along Stave [mm]

-0.020
-0.010
0.000
0.010
0.020
0.030
0.040
1
91
181
271
Series1
Series2
Series3
Series4
Series5
-0.020
-0.010
0.000
0.010
0.020
0.030
0.040
0
90
180
270
Series1
Series2
Series3
Series4
Series5
-0.02
-0.01
0
0.01
0.02
0.03
0.04
1
91
181
271
Series1
Series2
Series3
Series4
Series5
-0.020
-0.010
0.000
0.010
0.020
0.030
0.040
1
91
181
Series1
Series2
Series3
Series4
Series5
Chilled 1
-

Chilled 3

Chilled 1
-

Chilled 2 (with power)

Chilled 1
-

Chilled 3 (with power)

LBNL Stave Measurements


The change in stave shape is small as one chills with coolant ~
-
37 C, and then adds power
to mimic hybrid heat load.


The repeatability of these measurements is good over the short number of cycles (3) we
did.


Results will depend upon support. Ad
-
hoc method of support was intended to mimic
support with brackets spaced ~ 25 cm apart. Will later repeat these measurements with
prototype III when improved brackets are available first quarter next year.


We will be adding programmability to chiller to perform many temperature cycles to be
performed after which difference plots can be made.