rtm and vartm design, optimization, and control …automated design • selection of initial gate...
TRANSCRIPT
UD-CCM l 2 July 20031
RTM AND VARTM DESIGN, OPTIMIZATION, AND CONTROL WITH SLIC
Kuang-Ting HsiaoUD-CCM
Report Documentation Page Form ApprovedOMB No. 0704-0188
Public reporting burden for the collection of information is estimated to average 1 hour per response, including the time for reviewing instructions, searching existing data sources, gathering andmaintaining the data needed, and completing and reviewing the collection of information. Send comments regarding this burden estimate or any other aspect of this collection of information,including suggestions for reducing this burden, to Washington Headquarters Services, Directorate for Information Operations and Reports, 1215 Jefferson Davis Highway, Suite 1204, ArlingtonVA 22202-4302. Respondents should be aware that notwithstanding any other provision of law, no person shall be subject to a penalty for failing to comply with a collection of information if itdoes not display a currently valid OMB control number.
1. REPORT DATE 26 AUG 2004
2. REPORT TYPE N/A
3. DATES COVERED -
4. TITLE AND SUBTITLE RTM And VARTM Design, Optimization, And Control With SLIC
5a. CONTRACT NUMBER
5b. GRANT NUMBER
5c. PROGRAM ELEMENT NUMBER
6. AUTHOR(S) 5d. PROJECT NUMBER
5e. TASK NUMBER
5f. WORK UNIT NUMBER
7. PERFORMING ORGANIZATION NAME(S) AND ADDRESS(ES) University of Delaware Center for Composite Materials Newark, DE 19716
8. PERFORMING ORGANIZATIONREPORT NUMBER
9. SPONSORING/MONITORING AGENCY NAME(S) AND ADDRESS(ES) 10. SPONSOR/MONITOR’S ACRONYM(S)
11. SPONSOR/MONITOR’S REPORT NUMBER(S)
12. DISTRIBUTION/AVAILABILITY STATEMENT Approved for public release, distribution unlimited
13. SUPPLEMENTARY NOTES See also ADM001700, Advanced Materials Intelligent Processing Center: Phase IV., The original documentcontains color images.
14. ABSTRACT
15. SUBJECT TERMS
16. SECURITY CLASSIFICATION OF: 17. LIMITATION OF ABSTRACT
UU
18. NUMBEROF PAGES
30
19a. NAME OFRESPONSIBLE PERSON
a. REPORT unclassified
b. ABSTRACT unclassified
c. THIS PAGE unclassified
Standard Form 298 (Rev. 8-98) Prescribed by ANSI Std Z39-18
2 July 2003© 2003 University of Delaware All rights reservedHsiao ONR Workshop - 2
The Resin Transfer Molding (RTM) Process
Preform lay-up
Mold closure
Resin injection
Cure and
Mold openingFabric cutting
Heat Flux
2 July 2003© 2003 University of Delaware All rights reservedHsiao ONR Workshop - 3
The Vacuum Assisted Resin Transfer Molding (VARTM) Process
Fiber preform
MoldCured part
Resin injectionResin impregnates
fibers and cures
Resin injection Fiber preformunder vacuum
Vacuum pump
Vacuum Bag
2 July 2003© 2003 University of Delaware All rights reservedHsiao ONR Workshop - 4
Governing Equations for RTM Flow Simulations
injPP =
∂∂
+∂∂
−=tP
KnP
KA
Q ntnnµinj
ventPP =
01
=
∂∂
+∂∂
−=tP
KnP
Ku ntnnn µ
or
j
iji x
PKu
∂∂
−=µ
Volume averaged velocity-pressure relationship for flow in porous media:
Darcy’s Law
Continuity equation for the Resin
0=∂∂
i
i
xu
0=
∂∂
∂∂
j
ij
i xPK
x µ
2 July 2003© 2003 University of Delaware All rights reservedHsiao ONR Workshop - 5
Simulation-based Liquid Injection Control: Philosophy
SLICSLIC
Artificial IntelligenceOptimized Design
For RTM/VARTM
Sensors & Recognition Algorithms
Process Monitoring
Control Algorithms
Feedback Control
Mesh
Preform PermeabilityFiber Volume Fraction
Resin ViscosityX
Y
Z
2 July 2003© 2003 University of Delaware All rights reservedHsiao ONR Workshop - 6
Features of SLIC
Layout of Flow Runners and Flow Distribution Media
Gates/Vents DesignMold Filling Monitoring and Characterization of Fiber Volume Fraction and Permeability
Online Mold Filling Control
Preform PermeabilityFiber Volume Fraction
Mesh Resin Viscosity
$$$$$$Cost Function& Constraints
SLICSLIC
2 July 2003© 2003 University of Delaware All rights reservedHsiao ONR Workshop - 7
Case 1: Optimize Gate and Vent LocationsInput
Length=1.50mWidth=1.00mHeight=0.20m-------------------------------------------------------Thickness = 0.01mKxx = Kyy = 1E-10 m^2Vf=0.5-------------------------------------------------------Resin Viscosity = 0.12 Pa-sec = 120 cps-------------------------------------------------------Injection Pressure = 3.03E+5 PaVent Pressure = 1.01E+5 Pa
Injection Gate
Vent
Optimized Mold Filling
SLICSLIC
2 July 2003© 2003 University of Delaware All rights reservedHsiao ONR Workshop - 8
Gate/Vent Candidates
Flow of Optimizing Gate(s)/Vent(s) with SLIC (Case 1)
Constraints and Cost Function f=f(Equip.Cost, Filling Time, Dry Spot)
Use Corresponding Macro(Either SLIC Default or
User Defined)
Select Desired Feature
Mesh (part.dmp)
SLICSLIC
Injection Gate
Vent
2 July 2003© 2003 University of Delaware All rights reservedHsiao ONR Workshop - 9
Case 2: A VARTM/Co-Cure Case Study
How to optimize the distribution network design?
VentGateFlow Channel
Distribution Layer
Preform
Pre-manufactured Ribs
2 July 2003© 2003 University of Delaware All rights reservedHsiao ONR Workshop - 10
Experimental Setup (for Case 2)
2 July 2003© 2003 University of Delaware All rights reservedHsiao ONR Workshop - 11
Trial and Error – 1 (for Case 2)
Fill Time ~ 19 min
2 July 2003© 2003 University of Delaware All rights reservedHsiao ONR Workshop - 12
Trial and Error – 2 (for Case 2)
Fill Time ~ 12 min
2 July 2003© 2003 University of Delaware All rights reservedHsiao ONR Workshop - 13
Trial and Error 4 - Expert Guess (for Case 2)
Fill Time ~ 11 min
2 July 2003© 2003 University of Delaware All rights reservedHsiao ONR Workshop - 14
Procedure – Optimizing Distribution Layers and Runner Channels with SLIC
1. Create Finite Element Mesh (Geometry) of the composite part.
2. Collect Permeability/Fiber Volume Fraction of the Preform.3. Characterize the Permeability of the Distribution Media by
using SLIC.4. Calculate the Effective Permeability of the Flow Runner
Channels.5. Use SLIC to optimize the placement of Distribution Media
and Flow Runner Channels.
Run an experiment to verify the design.
2 July 2003© 2003 University of Delaware All rights reservedHsiao ONR Workshop - 15
Case 2: A VARTM/Co-Cure Case Study
Pre-manufactured Rib Structure
2 July 2003© 2003 University of Delaware All rights reservedHsiao ONR Workshop - 16
Distribution Layer Permeability Measurement with SLIC
Permeability ratio of the distribution layer and the preform is assumed as 5, 10, 15,…,150.
Times for top surface
Times for bottom surfaceExperiment
Match
SLIC
12 Experiments were conducted, the permeability ratio was obtained as 20-40.
2 July 2003© 2003 University of Delaware All rights reservedHsiao ONR Workshop - 17
Flow Distribution Network Design by SLIC
Fill Time ~ 13 min
2 July 2003© 2003 University of Delaware All rights reservedHsiao ONR Workshop - 18
Intuitive (Trial-and-Error) Design vs. SLIC’s Design
Final (fourth) intuitive design SLIC’s design
2 July 2003© 2003 University of Delaware All rights reservedHsiao ONR Workshop - 19
Case 3: A VARTM Case Study
D3: h = 5×h0
D2: h = 5×h0
L3: A =3× A0
L2: A =2× A0
Design 1Design 1
L1: A = A0
GateVent
D3: h = 4×h0 D2: h = 4×h0
L1: A = A0L2: A = 2×A0
Design 2Design 2L1=L2: A = A0
D1=D2=D3: h = 4×h0
Design 3Design 3
D: Distribution Media h0: Thickness of 1-ply distribution media L: Flow Runner A0: Cross-section Area of reference Flow Runner
2 July 2003© 2003 University of Delaware All rights reservedHsiao ONR Workshop - 20
Case 4: Steps on a Boat Deck (VARTM with Flow Runners)
Design 2Design 1 Design 3
Gate Vent Flow Runner
2 July 2003© 2003 University of Delaware All rights reservedHsiao ONR Workshop - 21
Permeability Variations
•Injection Pressure/Port
•Vent Pressure/Port
•Resin Viscosity
•Fiber Volume Fraction
•Permeability of the the Preform
LIMS
Characterization Challenges !Mold wall
Preform(fiber tows)
gap
Deformed fabricDraping overa tool surface
K and Vf change
??
2 July 2003© 2003 University of Delaware All rights reservedHsiao ONR Workshop - 22
Case 4: Using SLIC to Characterize the Racetracking
0
10
20
30
40
50
60
A B C D
E (La
ser) A B C D
E (La
ser) A B C D
E (La
ser)
Ave
rag
e ra
cetr
acki
ng
str
eng
th
Edge R1 Edge R2 Edge R3 Edge R4 Edge R5
The Standard deviation in racetracking strength by each operator can be as high as 80%.
R1
R2
R 3
R 4
R 5
Line injection
S1S2
S3
S4
S5
Five different operators A, B, C, D and E run 10 experiments each. A, B, C and D cut the fabrics by hands, E used a laser cutter.
A B C D
E (La
ser) A B C D
E (La
ser)
2 July 2003© 2003 University of Delaware All rights reservedHsiao ONR Workshop - 23
Streamlined Flow Monitoring & Control - From Design To Automation
Preform PermeabilityFiber Volume Fraction
Mesh Resin Viscosity
$$$$$$Cost Function& Constraints
Automated Mold Filling Monitoring & Control
Desired Performance & Process Constraints
Flow Detection/Monitoring Database
Flow Control Database
Equipment Installation Instructions
SLICSLIC
2 July 2003© 2003 University of Delaware All rights reservedHsiao ONR Workshop - 24
Developing Flow Sensing/Control System with SLIC
Hardware InstallationCode (HIC)
Vent
Pump
CAC 4
CAC 3
CAC 2
CAC 1
Control Action Code(CAC) for Each Disturbance Mode Controlled Results
Pum
p1 x 7 x
6 x
Number of Pumps, Lines,and Sensors to be used Potential Locations for
Gates/Vents/SensorsMesh and RacetrackingInformation
SLIC
2 July 2003© 2003 University of Delaware All rights reservedHsiao ONR Workshop - 25
Flow of Automation
SLIC
Computer
Simulation-based Liquid Injection Control (Intelligent Design Software)
Process Design•Hardware Installation Code•Dimensionless Time Vectors•Control Action Codes
Hardware
Setup
Instruction Operator (or Robot)
Vent
Installation(Hardware Setup)
Pum
p
LabView
2 July 2003© 2003 University of Delaware All rights reservedHsiao ONR Workshop - 26
Case 5: Online Flow Sensing/Control with SLIC
Injection gate with flow runner
Vent
…32 Disturbance Modes
Dry Spot Dry Spot
Dry Spot
2 July 2003© 2003 University of Delaware All rights reservedHsiao ONR Workshop - 27
Experiment Preparation (for Case 5)
Tekscan is laid on the bottom plate
Tekscan sensor
Fabrics and insert are placed on the top of Tekscan
FabricsInsert Sensor location
VENT
AG_1
AG_2IIL_1
IIL_2
AG: Auxiliary gate IIL: Initial injection line
Plexiglas Mold
2 July 2003© 2003 University of Delaware All rights reservedHsiao ONR Workshop - 28
Case 6: Online Flow Monitoring & Control with SLIC
Control action trigger sensor (CS)
Initial injection gate (IG) with flow runner
Auxiliary gate (AG)
Fixed vent
Disturbance detection sensor (DS)
Experimental resin arrival times
t0, t1, t2, t3, t4 are all collected
Disturbance Mode 29 is selected from the Database
Implement the customized control action for Mode 29
TekscanTM Sensor Area(Pressure Grid Film)
Control action Mode 29 is taking place. • CS1 >>> Close IG2• CS2 >>> Open AG1• CS3 >>> Close IG1• Vent Sensor >>> Close All Gates.
AG1
AG2
CS2IG1 IG2
CS1
CS3
Successful injection
2 July 2003© 2003 University of Delaware All rights reservedHsiao ONR Workshop - 29
Summary
SSLLIICCAUTOMATED DESIGNAUTOMATED DESIGN
• Selection of Initial Gate and Vent Locations• Optimization of the Flow Distribution
Network• Online Flow Sensing/Permeability
Characterization System Design• Creation of Online Flow Control Solution
Advantages of developing RTM/VARTM with SLICAdvantages of developing RTM/VARTM with SLIC
•• Rapid design for RTM/VARTM.Rapid design for RTM/VARTM.
•• Less cost for process development.Less cost for process development.
•• Reliable and comprehensive mold filling solution.Reliable and comprehensive mold filling solution.
•• Advanced flow monitoring/control technology Advanced flow monitoring/control technology provides the opportunity to elevate the part provides the opportunity to elevate the part quality and reduce the cycle time.quality and reduce the cycle time.
PREFORM
RACETRACKING(DISTRUBANCE)
RESIN
00.20.40.60.8
1
0 0.02 0.04 0.06
µ
α (Τ)
2 July 2003© 2003 University of Delaware All rights reservedHsiao ONR Workshop - 30
Acknowledgements
Ø Professor Suresh G. AdvaniØ Ms. Delphine CoatlevenØ Mr. Mathieu DevillardØ Ms. Susanna LaurenziØ Mr. Dhiren ModiØ Mr. Yeshwanth Rao .K. NaveenØ Dr. Pavel Simacek
Ø Office of Naval Research (ONR)