flow-3d v.8.0.1 (retired software, used to solving ... · can be done if it is decided that flow-3d...
TRANSCRIPT
a /47
SOFTWARE RELEASE NOTICE
1. SRNNumber: 2
2. Project Title: Project No. manyGeneral usage (TEF, WFO, ENFE) cfd modeling
3. SRN Title: -FA z 3 t/4v .Or& B. I4. Originator/Requestor: Randall Fedors| Date: 4/08/03
5. Summary of Actions
o Release of new software 0 Change of access software
o Release of modified software: * Software Retirement
o Enhancements made
o Corrections made
6. Validation Status
o Validated
o Limited Validation
* Not Validated Explain: Exploratory use only at this time. Validationcan be done if it is decided that FLOW-3D will be usedfor YM license application review
7. Persons Authorized Access
Name Read Only/Read-Write Addition/Change/Delete
Randall Fedors RO AdditionSteven Green RO AdditionDavid Walter RO Addition
8. Element Manager Approval: Date . @ -
9. Remarks:
CNWRA Form TOP-6 (09/01)
a
SOFTIWARE RELEASE NOTICE
.1. SRN Number: 2. .
2. Project Title: Project No. manyGeneral usage (TEF, WFO, ENFE) cfd modeling so. C or C- .--Or. o l |
3. SRN Title: IrLto -3- 7f Hi w A, A ||
4. Originator/Requestor: Randall Fedors |Date: 10/04/02
5. Summary of Actions
* Release of new software 0 Change of access software
o Release of modified software: 0 Software Retirement
o Enhancements made
o Corrections made
6. Validation Status
o Validated
a Limited Validation
* Not Validated Explain: Exploratory use only at this time. Validationcan be done if it is decided that FLOW-3D will be usedfor YM license application review
[ 7. Persons Authorized Access
Name | Read Only/Read-Write | Addition/Change/Delete ]Randall Fedors RO AdditionSteven Green RO AdditionDavid Walter RO Addition
8. Element Manager Approval: X By <Date: 10-1 -
9. Remarks:
-AV- -- Ad - I ^ -. 1E
1JO ig)vt-
to " A runn I ur-O k09/U I)
3//
SOFr'WARE SUMMARY FORM
01. Summary Date: 10/4/02 02. Summary prepared by (Name and phone) 03. Summary Action:Randall Fedors 210-522-6818
New04. Software Date: 05. Short Title: FLOW-3DI
2001 1 j06. Software Title: FLOW-3D version 8.0.1 [ 07. Internal Software ID:
08. Software Type: 09. Processing Mode: 10. Application Area
O Automated Data System D Interactive a. General:* Scientific/Engineering a Auxiliary Analyses
* Computer Program D Batch a Total System PAO Subsystem PA E Other
[1 Subroutine/Module * Combinationb. Specific:Fluid flow and heat transfer in open cavities and porous media.
11. Submitting Organization and Address: 12. Technical Contact(s) and Phone: www.flow3d.comFlow Science, Inc.
CNWRA/SwRI 683 Harkle Road, Suite A6220 Culebra Road Sante Fe, NM 87505San Antonio, TX 78228 (505)982-0088
13. Software Application:Computational fluid dynamics code for solving complex fluid flow and heat transfer problems using the volume-of-fluidtechnique, and rectangular grids with versatility for other geometries accomplished through the use of fractional facearea and fractional volumes of each element. Three-dimensional solutions for Navier-Stokes equation, and simplificationsof said equation, for two-phase and two-component problems, including laminar and turbulent flow and incompressibleand compressible fluids.
14. Computer Platform 15. Computer Operating 16. Programming 17. Number of SourceDual AMD 1600+MP System: Language(s): Executable Program Statements:
LINUX Mandrake 8.1 compiled from Fortran & C N/A
18. Computer Memory 19. Tape Drives: 20. Disk Units: 21. Graphics:Requirements: N/A Minimum needed 150 MBytes N/Aminimum 128 MBytes
22. Other Operational Requirements
23. Software Availability: 24. Documentation Availability:E Available E Limited * In-House ONLY [ Available E Preliminary * In-House ONLY
25.
Software DevelopeT. '~A-: Date: __________
CNWRA Fon) TOP4-1 (05/98)
* 6 C/61
CENTER FOR NUCLEAR WASTE REGULATORY ANALYSESQA VERIFICATION REPORT
FOR-PACQUIRED SOFTWARE NOT TO BE MODIFIED 4
Software Title/Name: o i 3 b
Version: 3, O I
Demonstration workstation: wDue k P ) 6 Above- M
Operating System: L 0o *s L ke eIlUser: R n Fe s | S-ie C-e'-
NOTE: Acquired software may or may not meet all requirements and will be evaluated on a case-by-case basis.
Installation Testing [TOP-0 18, Section 5.6]
Has installation testing been conducted for each intended computer platform and opera ing system?tarteA Am)'-t~ Yes: No:O N/A:O
Computer Platforms: 1 o°°tAP Operating Systems: [- ' MiN k J u t A'.. I
Location of Ae Test Results: zJ wozz 0 eA q ta ZComments:.
Software Output [TOP-018, Section 5.5.4]
Is software designed so that individual runs are uniquely identified by date, time, name of software andversion?
Yes: SltNo: O N/A: fDateandTimeDisplayed:_17 I Z(O\ta Z , O ; S o.?Name/Version Displayed: - l , ', c , t, }
Comments:
NOTE: Output identification content and format is typically taken as is.
Medium Documentation [TOP-018, Section 5.5.6]
The physical labeling of software medium (tapes, disks, etc.) conjains: Program Name, Module/Name/Title,Module Revision, File type (ASCII, OBJ, EXE), Recording I;yte, and OperatirdSystes(s)?
Yes: 9' No:C N/A: 0Comments:
(04/01) Page 1 of 3
CENTER FOR NUCLEAR WASTE REGULATORY ANALYSESQA VERIFICATION REPORT
FOR-ACQUIRED SOFTWARE NOT TO BE MODIFIED 4-
User Documentation [TOP-018, Section 5.5.7]
Is there a Users' Manual for the software and is it up-to-date? Yes: No: 0 N/A: 0
User's Manual Version and Date: V gjrS o' - ,Comments: SA-C,\\ _ ~ re~^
Are there basic instructions for the installation and use of the software?A ~~~~~ ~~~~~ ~~~~~~~~Yes: 1 No: O N/A: O
Location of Instructions: See VJ Yckns.JComments: ?o- ° t 'd (U, -_
Configuration Control [TOP-01 8, Section 5.7, 5.9.3]
Is the Software Summary Form (Form TOP-4-1) completed and signed?Yes: 1l No: 0 N/A: 0
Date of Approval: °A d- l i-'
Is the list of files attached to the Software Summary Form complete and accurate?Yes: No: 0 N/A: 0
Comments:
Is the source code available or, is the executable code available in the case of (acquire commercI codes)?Yes: 1 9 No: ,N/A: E
Location of Source Code: __________ C £ ' cowComments: jot 0 -A
Have all the script/make files and executable files been submitted to the Software Custodian?
Only the executable files are being submitted.
Yes: G/ No: 0 N/A: 0
Location of executable files: 5C- 4 .G d bComments:
(04/01) Page 2 of 3
0 ~~0
CENTER FOR NUCLEAR WASTE REGULATORY ANALYSESQA VERIFICATION REPORT
FOR-#ACQUIRED SOFTWARE NOT TO BE MODIFIED 4-
Software Release [TOP-018, Section 5.9]
Upon acceptance of the software as verified above, has a Software Release Notice (SRN), Form TOP-6 beenissued and does the version number of the software match the documentation?
AD q) Yes: S/ No: O N/A: OSRN Number: 6 YComments:
Software Validation [TOP-0 18, Section 5.10]
Has a Software Validation Test Plan (SVTP) been prepared for the range of application of the software?
Yes: O No: IYV N/A: O
Version and Date of SVTP:
Date Reviewed and Approved via QAP-002:
Comments: X ,v
Has a Software Validation Test Report (SVTR) been prepared that documents the results of the validationcases, interpretation of the results, and determination if the software has been validated?
Yes: O No: ID N/A: O
Version and Date of SVTR:
Date Reviewed and Approved via QAP-002:
Comments.:
Additio mments:
Software Evaluator/User/Date Software Custodian/Date
(04/01) Page 3 of 3
Patches in Version 8.0.1April 2002
Unde__l__ text means the problem existed in versions older than .0.
-7/6,1
Solver
1. The new scour model introduced in Version 8.0 has beenmodified to resolve a "divide-by-zero" problem.
2. The new multi-block model in Version 8.0 has been modified tocorrectly calculate void areas. Among other things, this problemresulted in divide-by-zero problems in the general bubble model.
3. In Version 8.0, MUI was used as the inlet boundary viscosity inthe thixotropic viscous model. Since this may not be correct intwo-fluid problems, a correction has been introduced.
4. A bug in the topo-data reader for single-block calculations hasbeen corrected.
5. The new implicit viscous model has been corrected so that thevelocities in the emptied cells are reset to zero.
6. In Version 8.0, global values for fill or solidified fraction, usuallycomputed for all mesh blocks, were being reset to the value for anindividual mesh block when a pressure iteration or convectivefailure occurred. This has been resolved.
7. In Version 8.0, during restarts with grid overlay, large errors couldbe introduced during data interpolation between the old and newmeshes. This has been resolved.
8. Changes have been made to the force calculations for ForceWindows, baffles and obstacles to improve accuracy.
9. A bug in the mass source model, which cause inaccuracy incalculation of the effective draa force, has been corrected.
10. At restarts, when void data was redefined in the input file Dreoinr,using the IHTRST=3 option. the resulting void data was,,otent.i.yv inconsintent %;ii;; the fluid and void data in the f!ser file
20. A problem where mass sink/source terms for density incompressible flow was not right, sometimes resulting in negativedensities and temperatures, has been resolved.
21. A problem with particles moving out of left, front and bottom meshboundaries when using the multi-block model, has been resolved.
Pre-Processor
22. A check has been added to the pre-processor to make sure thepermittivity ELPERM is positive. This corrects a problem where, ifa user sets the permittivity ELPERM to zero, a divide-by-zeroproblem was created in the solver. Similar checks were made forthe dielectric constants, DIELF1, DIELF2, ODIEL and DIEPLR.
23. The pre-processor will now plot composite mesh plots for themulti-block model when each block has only two cells in a givendirection.
24. A problem in which the pre-processor would not detect when thetitle line in the prepin is missing (resulting in the omission XPUTnamelist) has been resolved.
25. A problem in which the long print spatial limits, defined in thenamelist LIMITS, were not stored separately for each mesh block(resulting in incomplete print output in the solver summary file)has been corrected.
26. In Version 8.0, where there were multiple baffles, pre-processorbaffle plots were not correct and could result in a crash. This hasbeen corrected.
27. The defaults for the variable.txt file have been corrected forDIELF1 and DIELF2.
28. In the STL file reader, some of the constants were computed in aLavhat made the results dependent on the units svstem. Thishas been corrected in Version 8.0.1.
29. Obstacle areas are computed and reported by the pre-processoronly if certain models are turned on (e.o.. heat transfer or mass
from which all other data was being read. This has been correctedin Version 8.0.1.
11. An inconsistency in the calculation of the mass source/sink termin the density equation has been resolved.
12. A sign error in electric forces on charged fluids has been fixed.
13. Previously, when IHONLY=1 at restart, all turbulence modeloptions, including turbulent diffusion were turned off, even if theywere activated by the user. Problems arose if there was turbulentdiffusion of thermal energy. RMRHOE > 0.0. because the thermalmodel at restart was not the same as in the first run due to thebug. This problem has been addressed in Version 8.0.1.
14. A problem with cylindrical coordinates (obstacles not being drawncorrectly in the x-y plane) is corrected.
15. Fluid/void initialization with pointers. Void pressure may not be setcorrectly when using pointers in FL namelist. This has beenresolved.
16. The user-customizable routines, QSADD and FORCAL, havebeen updated for the multi-block model. This mainly involves
- changes to the extensive comments that we have there.
17. The capillary curve slope calculation in the unsaturated porous
sources). In Version 8.0.1. they will be computed all the time.
Post-Processor
30. The post-processor has been modified to eliminate some debugwrite statements.
31. Plotting labels for multi-block plots have been incorrect for certaintypes of plots. This has been resolved.
32. 2D multi-block plots in cylindrical coordinates have beenimproved.
Online Manual
33. Certain errors and inconsistencies in the online manual arecorrected. For example:
a. DIELE has been deleted.b. A correct default has been given for EPSVIS.c. The explanation of IPONLY=1 is updated.d. The defaults for DIELF1, DIEFL2, DIELPR and ODIEL are
corrected.e. A note added that all azimuthal coordinates in prepin file are
now defined in degrees, not arc length as before.
Platform/lnstallation Problems
34. Due to an optimization problem on the solver routine HTEVAL onthe HP-UX platform, this run was core-dumping. We havecorrected this problem.
35. In Version 8.0, there were noticeable differences in the results forthe mblock and dielectric example problems on single-precisionLinux and HP-UX. This was tracked to the optimization. We nowuse -01 optimization on those platforms (single-precision only).
media flow model as been corrected.
18. When IHELP=2 and the mentor activates the implicit viscousmodel. IMPVIS=1. it will now define all required variables.
19. A problem wherein some diagnostic variables for the heat transfermodel were only computed when LPR > 2. but were still writtenout to the flsanf file when LPR < 3 (as zeros) (thus possiblvconfusing the user). has been resolved.
Y/FLOW - 3D, VERSION 8.0.1
Volume in drive E is VOLUME_1Volume Serial Number is 17B4-9OC7
Directory of E:\
04/17/200204/17/2002
04:00p 10,809 LICENSEAGREEMENT04:08p <DIR> unix
1 File(s) 10,809 bytes
Directory of E:\unix
04/17/200204/18/200204/17/200204/18/200204/18/200204/18/200204/17/200204/17/200204/17/200204/18/200204/17/200204/17/200204/18/2 00204/17/200204/17/200204/17/200204/17/200204/17/200204/17/200204/17/200204/17/200204/17/200204/17/200204/18/200204/17/200204/17/200204/17/200204/18/200204/17/200204/18/200204/17/200204/18/200204/17/200204/17/200204/18/200204/17/200204/17/200204/18/2 00204/17/200204/17/200204/18/200204/17/200204/18/2002
04: 08p05:40p04: 00p03: 16p03: 16p03:17p04: 00p04: 01p04: 0lp03:1 7p04: 01p04: 06p03:18p04: 01p04: 06p04: 04p04: 00p04: 05p04: 05p04: 06p04: 07p04:06p04:06p04:21p04: 05p04: 06p04: 02p03:19p04:06p03:19p04: 04p03:20p04: 02p04: 07p03:20p04: 03p04: 03p03:21p04:04p04:04p03:21p04: 05p03:22p
41 File(s)
<DIR><DIR>
10,809 LICENSEAGREEMENT34,406,467 alpha-u4.Z34,410,915 alpha-u4_ev5.Z34,402,253 alpha-u4_ev6.Z16,663,005 alpha-u4_exe.Z1,257,269 alpha-u4_lic.Z14,337,935 helputil.Z27,842,263 hp9000_ulOpa7000.Z10,072,523 hp9000_ulO0pa7000_exe.Z1,642,047 hp9000_u0lpa7000_lic.Z
27,635,409 hp9000_ul0_pa8000.Z9,907,939 hp9000_ul0_pa8000_exe.Z1,641,871 hp9000_ul0_pa8000_lic.Z7,168,000 installalpha
549 installflow3d6,774,375 installhp
11,740,006 installibm5,088,844 installlnx
38,244 installscript8,660,636 install-sgi5,626,932 installsun
25,237,567 linux-pc.Z8,502,289 linux-pcexe.Z
797,263 linuxpc_lic.Z10,276,655 rs6000_u4_exe.Z26,226,331 rs6000_u4_generic.Z
993,009 rs6000_u4_genericlic.Z26,292,029 rs6000_u4_pwr.Z
11,268 runinstall30,233,865 sgi32_u6_mips3.Z9,955,355 sgi32_u6_mips3_exe.Z1,202,849 sgi32_u6_mips3_lic.Z
30,237,045 sgi32_u6_mips4.Z9,051,861 sgi32_u6_mips4_exe.Z1,202,171 sgi32_u6_mips4_lic.Z
30,430,407 sgi64_u6_mips4.Z1,217,384 sgi64_u6_mips4_lic.Z
9,588,341 sun_u5_exe.Z33,723,787 sun_u5_generic-ws5.0.Z
920,091 sun_u5_generic-ws5.0_lic.Z33,430,028 sun_u5_ultraws5.0.Z548,857,886 bytes
Page 1 of 2
Total Files Listed:42 File(s)3 Dir(s)
548,868,695 bytes0 bytes free
Page 2 of 2
MULTIPHASE FLOW SECTION
TO: Bruce Mabrito September 25, 2002
FROM: Steve Green d/c
SUBJECT: VAi~tiouof FLOW-3D Installation
FLOW3-D is a general purpose computational fluid dynamics software package that can be usedto support several projects and research efforts being conducted by the Center for Nuclear WasteRegulatory Analysis (CNWRA). This vendor of this code is Flow Science, Inc.(www.flow3d.com) of Santa Fe, New Mexico.
~~e~a4~a~e t.est Vea o-2This memo serves to document the va464ift of the FLOW-3D (Version 8.0.1) installation atSwRI. The approach taken here is to compare the results obtained with the SwRI installation ofFLOW-3D (on a LINUX server) with benchmark results provided by the software vendor (usinga PC running Windows 2000) for a single example problem. The example problem is titled'NCONXZ' by the vendor. It is a simulation of the convection heat transfer in a 2-D idealizationof a square cavity. The left and right sides of the cavity are specified as constant temperatureboundary conditions with the left side being hotter than the right side. The top and bottomsurfaces are perfectly insulated.
The FLOW-3D input file for this problem is attached. This is followed by a set of graphscomparing the results obtained by the vendor to the results obtained from the SwRI installationof this code version. These graphs include the time histories of certain flow parameters (e.g.,pressure, temperature, velocity) at a point near the midpoint of the hot wall. Also, certain globalflow parameters (e.g., total fluid thermal energy, fluid kinetic energy) and some simulationparameters (e.g., time step size, iteration count) are compared. Finally, two-dimensional contourand flow velocity vector graphs from the benchmark results are compared to the SwRI results forthe state of the fluid at the end of the simulation time.
It is seen that the two sets of results presented in the time history graphs agree closely. In fact,all the predicted parameters from the two simulations are essentially identical up to a simulationtime of about 0.3 sec. After that time the details of the single-point pressure and fluid velocitypredictions differ, but both sets of results oscillate about the same mean value. This is related tothe difference between the way that the two different computers accumulate and handle round-off errors as the computer code runs. Other single-point parameters, such as temperature andlocal energy are essentially identical. Finally, all of the global flow parameters predicted by thetwo simulations are in very close agreement.
The con from t exerc se is that the Swi installation of FLOW-3D has beenP'>.o2,success I a a IV h e the enchMaresusp ro ided by the vendor, Flow Science, Inc.
SO UT H WEST RES EAR CH INS T IT UT E
/1/&/
prepin.inpNATURAL CONVECTION IN A SQUARE CAVITY (RA=1.OE+3, 20x20 MESH)
This test calculation is based on a benchmark defined in:
G DE VAHL DAVIS AND I P JONES, "NATURAL CONVECTION INA SQUARE CAVITY: A COMPARISON EXERCISE", INTERNATIONALJOURNAL FOR NUMERICAL METHODS IN FLUIDS, VOL. 3,PP 227-248, 1983.
we have chosen the low rayleigh number case to reducecomputational cost.
The problem definition has been "non-dimensionalized"selecting lengths and physical properties equal to 1,for the viscosity and thermal expansion coefficient.rayleigh number is controlled by varying the value of
byexceptThegz.
$xputipdis=1,i fenrg=3,ifrho=1,itb=O,nmat=1,iwsh=1,gz=-7 . le+4,delt=0.001,twfi n=O. 5,
prtdt=O. 5,pItdt=O.1,ihtc=1,
$end$limitsSend$props
remark=' equaticvl=1.O,rhof=1.0,thc1=1.0,mui=0.71,
thexfl=0.01,
tstar=O.O,
remark='remark='remark='remark='remark='remark='remark='remark='remark='remark='remark='remark='remark='
initial hydrostatic pressure distribution',solve for internal energy and temperature',evaluate density from temperature ',no free surface ',one fluid',include wall shear ',9z gives rayleigh number = 1000 ',initial time step sizethis is a short final time '
it is chosen to reduce cpu time ',only print at end of calcu aiton ',only plot at end of calculation ',evaluate wall heat transfer ',
on of state parameters ',remark=' non-dimensional specific heat ',remark=' non-dimensional density ',remark=' non-dimensional conductivity',remark=' viscosity set to give the ',remark=' correct prandtl number ',remark=' thermal expansion coefficient ',remark=' must be small ',remark=' reference temperature for thermal',remark=' expansion',
Send$bcdata
remark=' walls on physical edges of the box ',wl=2,
tbc(1)=1.0,
hwal 11(1)=1. e4,
wr=2,tbc(2)=1.Oe-10,
remark='remark='remark='remark='
remark='remark='remark='remark='remark='
non-dimensional left boundary',temperature ',
effective heat transfer ',coefficient for conduction ',
a non-zero value is needed',because zero indicates an ',insulated boundary ',effective heat transfer ',coefficient for conduction ',
hwall1(2)=1.e4,
wb=2, rwall(5)=O.O,wt=2, rwall(6)=O.O,
remark=' insulated bottom boundary 'remark=' insulated top boundary',
Page 1
,;2./
prepin.inp
remark=' symmetry conditions for 2d approximation ',wf=l,wbk=l.
Send$mesh
remark=' define a simple, uniform meshnxcelt=20, px(2)=1.0,nzcelt=20, pz(2)=1.0,
SendSobsSend$fl
flht=1.0,Send$bfSend$temp
remark=' start from uniform temperature ',tempi=0.5,
Send$grafic
remark=' place a history probe in the lowexloc(l)=0.025, zloc(l)=0.025, ylocnvplts=2, remark='2 velocity ve,contpv(l)='tn', remark=' fluid temperacontpv(2)='p', remark=' pressure ,ncplts=2, remark='2 contour plot:ictyp(l)=5, remark=' color contourcontyp(l)='tn', remark=' of fluid teiictyp(2)=5, remark=' color contourcontyp(2)='p', remark=' of pressure
Send$partsSend
2
r left corner ',(1)=0.5,ctor plots ',ture ',
s ',lines ',mperature ',lines ',I
Page 2
or.d-4-1
U
to m
H
VC)
H
O-4i0 0 N
'-(NX
OC
) 0
0 C
N~
~~
~~
~~
~~
~~
~~
~~
~~
~~
~~
~~
~~
~~
~~
~~
~~
~~
~~
~~
~~
~~
~~
~~
~~
~~
~~
~~
~~
~~
~~
~~
~~
~~
~~
~~
~~
'
(N
xIf)
X
C
Ho
0~~~~~~~~~~~~~~~~~~~~~~~~~~~4C
)~~~~~~~~~~~~~~~~~~~~~~~~~~~~U
o 3o
H4
LO~
~~
~~
~~
~
0Z
C)~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
z
.~ ~
~ ~~~~~
_ H
(N
'O
O
4.OC
)04
-Hr- ~ ~ ~ ~ ~ ~
~ ~
~ ~
~ ~
~ ~
~ ~
~ ~
~~~~~~~-o
0~
~~
~~
~~
~~
~~
~~
~~
0
C-~
~~
~~
~~
~~
~~
-
.oH
UmH
ID
H
UQ)
H
O
U)
C
CN'0~~~~~~~~~~~~~~~~~~~
0 NH
0
C,'
(N~
~~
~~
~~
~~
~~
~~
~~
~~
~~
~~
~~
~~
~~
~~
~~
~~
~~
~~
~~
~~
~~
~D
CU) C
o .
O
-Hs>
f
X
t~~~~~~~~~~~~~~~~~~~~~>1
10
-2
=
0+
HO
0
C;~
O0~
~~
~
~~~~~~~~~~~~~~~~~~~~
LO
C,
o~~
~~
~~
~~
~~
~~
~~
~o
10~
~~
~~
~~
~~
~~
~~
~~
~~
~~
~~
0
r-I "I~~~~~~~~~~~~~~~~~~~~-U
.oo
S-i
U)
H
43
U)
I:')~~
~~
~~
~~
~~
~~
fH
~~
~~
~~
0C
d)~~
~~
~~
~~
~~
~~
~~
~~
C0 ~~
(NI~
~~
~~
~~
~~
~~
~~
~~
~~
~~
~~
~~
~~
~~
~~
~~
~~
~~
~~
~~
~~
~~
~~
~~
~~
~~
~~
~~
~~
~~
~~
~~
~~
~~
~l
a C
'- X
~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~'0~
~~
~~
~~
~~
~~
~~
~~
~~
~~
~~
~~
o N
o rz@
HH
Oo
H
C)
@n <
cq~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~cS
( O
N
HI
Ol
o~
~~
~~
~~
~~
~~
~~
~~
~~
-~~
~~
~~
~~
~~
~~
~~
~~
~~
~~
~~
~~
~
~~
~~
~~
~~
~~
l0"-IH
W
0
~ ~~~~~~~~~~~~~~~~~
0,
O
0~~
~~
~~
~~
~~
~~
~~
~~
~~
~~
~~
~~
~~
~~
~~
~a
O
0~
~~
~~
~~
~~
~~
~~
~o
oX0 E
HF I
v awC
i2mH
0H
H
a 00
0
0( N
:
o N
d
0
10
04-)
-
H~
~~
~~
~~
~~
~~
~~
~~
~~
~~
~~
~~
~~
~~
~~
~
U
I H
O~
~~
~~
~~
~~
~~
~~
~~
~~
~~
~~
~~
~~
~~
~O
S
>-I--)~~
~~
4F-
l~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~t)
0~
~~
~~
~~
~~
~~
~~
~~
~~
~~
~~
~~
~~
~~
~~
~~
~~
~~
~~
~~
~~
1W
z1 E
Eox
* vn
In
H4-lD
*
0N~
~O
f
v HE..
O
~ ~
~ ~
~ ~
~ ~
~ ~
~ ~
~ ~
~ ~
~ ~
C
NC
) E0
I l
4-
(D
w
Es
LO0
0
0.'I
-4H
(00)
al)
0qH
1
HU)4
0
Cl
0z10LC)
rIn
01
w00Ln)
00Cl
a)
-r00
N0002HO
0>1
ao
U
S
*N 0
0+cn4 0
O>
14
0II~
CO,)
NH0Nq O
(S H
t0®
U412:
i Ii
0 m
l0
0F
) .0
Y0
= M
U
z O
c U
O
4- U
H r)
44
tO m
~~
~~
~~
~~
~~
La
0)
0~~~~~~~~~~~~~~~~~~~~~~H
0~
~~
~~
~~
~~
~~
~~
~~
~~
~~
~~
~
0~~
~~
~0
0o
O~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
+
a~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~. X
ON >
0~~~~~~~~~~~~~~~~~~~~~~C
lH C
ON
I H
~~
~~
~~
~~
~~
~~
~~
~~
~~
O*
v -)In
0+C?)
toI
U~~~~~~~~~~~~~~~~~~~~~~~~~~~~C
\,
Z0~~~4 04-H
L
2~
~~
~~
~~
~~
O~
~~
~~
~~
~~
~~
~~
Il
,27
0.H-
-44-iV
I
H
'1)~42
0)
-14HU
)
U)
LO0;
I.-N00MC
N0~In00(N0U
10(N
a)r=.-H4-l
0
N00CD
-4Ox*
0
02(N
0+
-.
$4 H
10
(NH
0N O
~ HN
D E
-(N
O
a) :
0
00
0)H
0>
u
()r--lr--l
0.4 J4--)(oHHN
J
U4
0401)
Ual)
0H
LA
L)0
Cl
00clH000Ln
Ulq00U
-)
CNI
-H0
.~~~~~~~~~~~~~~~~~~~~~~00Nxcl)O
xN0 +
CJH
*,~
mE
-'F:~-H10~
N14H
0NO
N..0)) >U
LO
Z
4-)
0Lnm
-W
4 M
$L
L
I4 WU
O
U)4-4-
04-)
1-lH4-)U
i)
HxHCl)
1)a)Ua))
0H4
EL
.
U)m)
C;
CN
0l
W0LO(NH000
l00(N
a)
4-)
0~0>
.~ ~ ~ ~ ~ ~ ~ ~ ~ ~~~~~ ~~~~~~~~
C/
HO0.-4
0'+z
wC
>H
0 >
44 U3
c0'0
.H(N
C)
L~o
zC
N H
0o~
~~
c
(N0kD
Hto
E
U
AC
)
Ss0
ON
Z
11) 0
..~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~/
IV
U
,c: )
rd 4-U
O
() 4444
3 0
o X
H
Ut
cqmH
C
1)
V:2
H
V
0~~~~~~~
ro ox~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~T
H
0 +~~~~~~~~~~~~~~~~~~~
0 90
U
_
U)
HO
0 +
D E
-4
4-)
%1 1
2 ^
t @
@
@
h
m
S
^ ^
>
O
<
v O
h
S
aft~~~~~~m
o~ ~ ~ ~
~ ~~~~~~~~~~o
al*
~0
S
0
3,~
~~
~~
~~
H
!
H
m
HU
C)~
~~
~~
~~
~~
~~
~~
rX
:H
Oa )
HH
o~~
~~
~~
~~
~~
~~
~~
~~
~~
:~~~~~~~~~~~~~~~~~~~~~ _
--
--
----
-..-
---
_-
a L
)
-------------~~~~~~~~~-
----- -- -------
-----
----
---
--
-- -- -
-- -
--
---
---
-
c
--
-----
---
----
---
----
------
-.
A_ _ _ _
_ __
_
--------------_ _
-_-_
-_ _
s~~~~~~~~~~~~~~~~~~~~~~~~-- -
--- --- ---
--
C')
O
H'
J1-) H
O+
U)
H~
~~
~~
~~
~~
~~
~~
~~
~~
~~
~~
~~
~~
~~
~~
~~
~~
~~
~~
~~
.
U)
'IaW
z I
~~~~~~~~~~~~~~~~~~~~~~~~~~~~>t
(L) z IH
04~~~~~~~~~~~~~~~~~~
z III-
cl H
*
~~~C40
O
(.
~~
(Ok H
C)0
r-H4J a)
~4 (
J-HQri 0O
4-)i
3 )
0 a4
o1 (
vm:
4U)
X4
0H
U
:{H
U
0)
a1 44
H
3
TO
N
:~~~~~~~~~~~~~~~CU
).0.~~~~~~
H~~~~~~~~~~~~~~~
UE
U)
C~~~~~C
1
w
C
l l
H
O~~~~~E
0]
-H ~~
~~~ U
X
.1) @
@ z
m SO
*
g44S
-H~
~~
~ -
'0 >z*
~0
0 *
c~~~~~~~~~~~~~~~~~~~~~u~~m
0~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~C
0(i)
U)
4-) -rI
(d 4J
() 'd
4-) H
0 (L
) 4)
~4 L
) >
0
v0H4-)
'-4
'-4
HH3oq
C)
0)m
.
.,,0)U
:
C1)
C):
0 :
H
U)
0n
(N4
01
W0Ln
(NH01
W000I1
0(N
a)-.H-W
-
00
(N
00C(N
x"-XC
)
HO
+
0+"
-4(0
0 C
>
11
..
d>
4
-H
XS>zC
J
0E-4
(NH
('oO H
(U0O
Uo~~~~~~~~~~~~~~~~~~~~o~
~~
~~
~L
.~~~~~~~~~~~~~~~01
i I
U-'
0Hi
H4
a,0
OQ
U
) -r-
(1) U
LW
( d U
U
)4-)
) ESQ.
.0
31V
0HH(tIH
i:-3
C)
0)Va)
0a)I
LO
0nCo
N.HcoU)
a)4,
a)
.H4i
0~
CD
00rlN
x
oz1
HOCD
N
P* 0
0 +
uoo
.1 -
> 11-4
N4 H..H
'0NO-
zvo q08
00+04410
00InU-)
CD
JJ
a) 04
U)-H
N
W
00+000)
0~
.I- H
S
U)
0
or. U
HU
0 -H
44.)
mH
4
H
U
:0:
H
0:
H
l 3
3 i
H
-1.
04
0 ~
.
0nU)
C;
cl0z1010H010
0010cl
-H0
(N0NxO>
10N
00 040
Oz
NO
Dt
O +
~4 H*N
O0>
U
CN
H3
08
W0
o
IIna'0)a'
0r°a'
a)
4-4
r- -H
'
-H
4 4V
o 4-4
4Jo V
o
0
34
04.)
21-
HN'
V0).V
~:
0 :
.L:
VQ
:
cn:
0H-
*n
UL
)
0
.-I
01l0Cl
H00000LC)
(N
a).00
0x.0
0.11xC
>
1
H -
4CY)
cl
H-4HC
z0
4-ICJ
E-I
CS
H0oZ'10N
O
H(Z
D
E2
10)
0Cn
m4
H-
0CH
u (1) r
r H
31
sor."-:
x0
H
VH
0
(0
*n
4J
U
Z~~~~~~~~~~~~~~~~~~~~~
U~
~~
~~
~~
0~
~~
~~
~~
H
v H
e
cO44
0 +
~~
~~
~~
~~
~~
~~
~~
~~
~
00~~~~~~~~~~~~~~~~~~~~~~~
F
N ~~~~~~~~~~~~~~~~~~O
f4U
4H
024~
~~
) 0+
0~~
~~
~~
~~
~~
~~
~~
~~
~0
4~
~
~ ~
~
~~
~-
-:
4r
zmE
I
--)r
1 -
4>E
~4
waC;
eq0~
00In0
v1�0
H
0,1
00
D
N
10 0
o o
0
0
V1�0
CD
N
H
1O
m000a4)kh4)h4)'0-1k
00U)
r
I; I
1 .
4 .-
I
'
14
:' C
N'4
I
0
N
40
0
q
O
N
o o
0
0
0 0
0
I .
..
..
F7
04
-.
0
:No 44
40 H
eq
0NIno
eq
v)
0u4)00
44.
a +4hh
11
4)E _
V
,1
0
I� *
� t
A
4 � t i'4
4 �A
�
4'
I ,44
�4�
4-
4
4 A
c IC
p II
44N4
090
N
44
x NC
40
e 0
, .
O~
>4
O
N.
OO
g m
I4O
N0N
NC
*(S
on*o
ODOxIM
1;N
0
0N
0
0
No~
~~
~~
~~
~~
~~
1
LX
10
11tH
a
.ID
.
..
1J.
40vaa:3I?V4),1044
.,4
agN0
0
Neq,
N
O0
Hw
H0
eq
0>4
IO440I0
0 +
4)2
0
0 0
4.4
O
>q
N
0O
OV
* eq0
0
44
z
V
Z
ol
H
0
N O
oq
08
0
0 O
O '
O
,
SO
1�
N_ c I
I~0
0
r0
mN
N
O
N
H1
en en
0en
.r4 0
w4
erm
rc
r
N
,
vIer1
en e
0
w)
UN
,H
0
U)
N
WU
' en
-40
40
0 )
U )
N
U )
4 0
enU
H
' en N
..4
V)
0440
rD0044o4404
014I
r
0
U)
U)
~ ~ ~~~ ~~N
C
)40
e
0
.0
o 4V
40 v-
X
eN~x*
.,4
U)
en_ O
N
>4
o 1e
0
+
°g, I
0O* -
tla0
en
T440.
--
--
� �
� )'
I * $4
p �
-#
I' ,�
FA. XA
. F
.A�
A4
.--
-4*
1o
I
N
00
N~
011
0
40*
.X
0°
0
en0
* +
o
0
H O
.4
cocN
Oa+
0)0
r>
1
In- E
00
>s
0,1
N
H
04.4
SC
N40 -4O
04
'4
040
.X
O O
on
0,,?
U E
o
oS
*
eN
.0
U)
r I
en en
rn
H
0
U)
N
0.
Ul
U)
fN
N
,1I
4
,1
wrVUl
3U)
tr44400.
*0
0F.
' 0N
0
4.N40-4
° V
-X
en
0
004
J 0
I V
~II
0IO,v0U)IU)04.
or4
+
I44
V ,1
'a
1'
rJ t
0*00=1,4
� q I
I iT
IMM
ER
IA,
1oN
Oc0
N
MULTIPHASE FLOW SECTION
TO: Bruce Mabrito October 4, 2002
FROM: Steve Green
SUBJECT: Installation Test Results for FLOW-3D version 8.0.1
FLOW3-D is a general purpose computational fluid dynamics software package that can be usedto support several projects and research efforts being conducted by the Center for Nuclear WasteRegulatory Analyses (CNWRA). This vendor of this code is Flow Science, Inc.(www.flow3d.com) of Santa Fe, New Mexico.
This memo serves to document the installation test for the FLOW-3D (Version 8.0.1) softwarepackage at SwRI. The approach taken here is to compare the results obtained with the SwRIinstallation of FLOW-3D (on a Dual AMD 1600+MP server running LINUX Mandrake 8.1) withbenchmark results provided by the software vendor (using a PC running Windows 2000) for asingle example problem. The example problem is titled 'NCONXZ' by the vendor. It is asimulation of the convection heat transfer in a 2-D idealization of a square cavity. The left andright sides of the cavity are specified as constant temperature boundary conditions with the leftside being hotter than the right side. The top and bottom surfaces are perfectly insulated.
The FLOW-3D input file for this problem is attached. This is followed by a set of graphscomparing the results obtained by the vendor to the results obtained from the SwRI installation ofthis code version. These graphs include the time histories of certain flow parameters (e.g.,pressure, temperature, velocity) at a point near the midpoint of the hot wall. Also, certain globalflow parameters (e.g., total fluid thermal energy, fluid kinetic energy) and some simulationparameters (e.g., time step size, iteration count) are compared. Finally, two-dimensional contourand flow velocity vector graphs from the benchmark results are compared to the SwRI results forthe state of the fluid at the end of the simulation time.
It is seen that the two sets of results presented in the time history graphs agree closely. In fact, allthe predicted parameters from the two simulations are essentially identical up to a simulationtime of about 0.3 sec. After that time the details of the single-point pressure and fluid velocitypredictions differ, but both sets of results oscillate about the same mean value. This is related tothe difference between the way that the two different computers accumulate and handle round-offerrors as the computer code runs. Other single-point parameters, such as temperature and localenergy are essentially identical. Finally, all of the global flow parameters predicted by the twosimulations are in very close agreement.
The conclusion from this exercise is that the SwRI installation of FLOW-3D has beensuccessfully completed, including comparing SwRI results against the benchmark provided bythe vendor, Flow Science, Inc.
SO UT H WEST RES EAR CH INS T IT UT E
9
prepin.inpNATURAL CONVECTION IN A SQUARE CAVITY (RA=1.OE+3, 20x20 MESH)
This test calculation is based on a benchmark defined in:
G DE VAHL DAVIS AND I P JONES, "NATURAL CONVECTION INA SQUARE CAVITY: A COMPARISON EXERCISE", INTERNATIONALJOURNAL FOR NUMERICAL METHODS IN FLUIDS, VOL. 3,PP 227-248, 1983.
we have chosen the low rayleigh number case to reducecomputational cost.
The problem definition has been "non-dimensionalized" byselecting lengths and physical properties equal to 1, exceptfor the viscosity and thermal expansion coefficient. Therayleigh number is controlled by varying the value of gz.
Lt/
$xputipdis=1, remark=' initial hydrostatic pressure distriifenrg=3, remark=' solve for internal energy and tempeifrho=1, remark=' evaluate density from temperatureitb=O, remark=' no free surface ',nmat=1, remark=' one fluid'iwsh=1, remark=' include wail shear ',gz=-7.le+4, remark=' 9z gives rayleigh number = 1000 ',delt=0.001, remark=' initial time step size ',twfin=0.5, remark=' this is a short final time ',
remark=' it is chosen to reduce cpu timeprtdt=0.5, remark=' only print at end of calcul aiton ',pltdt=0.1, remark=' only plot at end of calculation ',ihtc=1, remark=' evaluate wall heat transfer ',
$end$limits$end$props
remark=' equation of state parameterscv1=1.0, remark=' non-dimensionai specific heat ',rhof=1.O, remark=' non-dimensional density "thc1=1.0, remark=' non-dimensional conductivity',mui=0.71, remark=' viscosity set to give the ',
remark=' correct prandtl number ',thexfl=0.01, remark=' thermal expansion coefficient ',
remark=' must be small ',tstar=O.O, remark=' reference temperature for thermal',
remark=' expansion',
buti on',rature',
$end$bcdata
remark='wl=2,
wr=2,
walls on physical edges of the box ',
tbc(1)=1.0, remark=' non-dimensional left boundary',remark=' temperature ',
hwall(1)=1.e4, remark=' effective heat transfer ,remark=' coefficient for conduction
tbc(2)=1.Oe-10, remark=' a non-zero value is needed',remark=' because zero indicates anremark=' insulated boundary ',
hwall(2)=1.e4, remark=' effective heat transfer ',remark=' coefficient for conduction ',
rwall(5)=O.O, remark=' insulated bottom boundary ',rwall(6)=O.O, remark=' insulated top boundary',
Page 1
wb=2,wt=2,
.
prepin.inp
remark=' symmetry conditions for 2d approximation ',wf=1,wbk=1,
Send$mesh
remark=' define a simple, uniform meshnxcelt=20, px(2)=1.0,nzcelt=20, pz(2)==1.0,
SendSobsSend$fl
flht=1.0,SendSbfSend$temp
remark=' start from uniform temperature ',tempi=0.5,
SendSgrafic
remark=' place a history probe in the lowexloc(1)=0.025, zloc(1)=0.025 ylocnvplts=2, remark='2 velocity ve,contpv(1)='tn', remark=' fluid temperacontpv(2)='p', remark=' pressure 'ncplts=2, remark='2 contour plot:ictyp(1)=5, remark=' color contourcontyp(1)='tn', remark=' of fluid teiictyp(2)=51 remark=' color contourcontyp(2)= p', remark=' of pressure
SendSpartsSend
I
r left corner ',(1)=0.5,ctor plots ',ture ',
s 'Ilines ',nmerature ',; ines ',I
Page 2
V.e
0
To~
~~
~~
~~
~~
~~
~~
~~
~~
-44)
45)
C)~
~~
~~
~~
~~
~~
~~
~~
~~
~~
~~
~~
~~
~~
~~
~~
~~
~~
~~
~~
~~
~~
~~
~~
~~
~~
~~
~~
~~
5
0~~~~~~~~~~~~~~~~~
(N~
~~
~~
~~
~~
~~
~~
~~
~~
~~
~~
~~
~~
~~
~~
~~
~~
~~
~~
~~
~~
C
Lfl ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~00
C
(N
X>
CD4i~~.4
-ox~0
Xl
> ~
>
A
D~
~~
~~
~~
~~
~~
~~
~~
~~
~~
~o ~ ~ ~
~ ~
~ ~
~~~~~~~~oLO
0+
~~
~~
~~
~~
~~
~~
~~
~~
~~
~~
~~
~~
~~
~~
~~
~~
~~
~O
44 0
0~
~~
~~
~~
~~
~~
~~
~~
~~
~~
~
(N~
~~
~~
~~
~~
~~
~~
~~
~~
~~
~~
~~
~~
~~
~~
~~
~~
~~
~~
~~
~C
4-i~~
~~
~~
~~
~~
~~
~~
~~
~~
~~
~~
~~
~~
~~
~~
~~
~~
r
04 ~~~~~4 U) U
)~ ~
~ ~
~ ~
Cd
0
C'JH
~~
~~
~~
~~
~~
~~
~~
~~
~~
~~
~~
~~
~~
~~
~~
~~
~~
~~
~)L
*
0CI
S~~~~~~~~~so
'44:'U
cn~~
~~
~~
~~
~~
L
C)~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~a
UU
I~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~I
C\1~
~~
~~
~~
~~
~~
~~
~~
~~
~~
~~
~~
~~
~~
~~
~~
~~
~~
~~
~~
~~
~~
~~
~~
~~
~~
~~
~~
~~
~~
~~
~-
C'1x0
oc
.40
o -H
~~
~~
~~
~~
~~
~~
~~
~~
~~
~~
~~
~~
~~
~~
~~
~~
~~
~~
~~
~~
~~
~~
~~
~~
~~
~~
~~
~~
~~
~~
~~
~~
~~
-*
4....LO
an
C~~~~~
(l0
0~
~~
~~
~~
~~
~~
~~
~~
~~
~~
~LC
)~~
~~
~~
~~
~~
~~
~~
~~
~~
~~
~~
~~
~~
~~
~~
~~
~~
~~
~~
~C%
D -4C
i
0 L
n~~~~~~~~~~~~~~~~ ~
tOo
0 0
4
* ~~~~~O
E>
f02
uy
0 0
0.,4
k4
-H'iXH U
4.) 0)
H
0)rI)
UU
Ln
C12 ~ ~
~ ~
~ ~ C
0 ~ ~
CN
J~~
~~
~~
~~
~~
~~
~~
~~
~~
~~
~~
~~
~~
~~
~~
~~
~~
~~
~~
~~
~~
~~
~~
~~
~~
~~
~~
~~
~~
~~
~1
0
CN
J~
~ ~ ~ ~ ~ ~ ~ ~
~ ~
~ ~
~ ~
~ ~ ~
~~
~~
~~
~~
~~
~~
~~
~~
~~
~~
~~
Cox
In ~~~~~~~~~~~~~~~~~~~0+0
0
(Nq
>1
en -4H
(N
zL
I) C
O
N(t0
0 C
' l
*
~~0a)
r-I
_ U
-r-I
4U
o~~
~~
~~
~~
0O
N >
@
<
o U
^
v >
Q
o;~~
~~
~~
~
C.
NZ
H
0
4l
JJ
(UH 0); U
)
UU
C,
-4:~~
~~
~~
~~
~~
-
0~~~~~~~~~~~~~~~~~~
to
if)~~
~~
~~
~~
~~
~~
~~
~~
~~
~~
~~
~~
~~
~~
~r
4-)~~~~~-
O~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
M
Ln o1
m
Ex
O
t>~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~4O
D
0~~~~~~~~~~~~~~~~~~~
\4 H
D
la >04
4~~
~~
~~
~~
~~
~~
~~
~~
~~
~~
~~
~)
~~~ ~
~
~
~
~
~
~
C
o C
' H*
~~0
o4)4 W
on
0
00(aHHHR
@S~
4
a@UC)
Cl)
0t4 3Oi
LnIn0D
N010L
O
H01W00L0
L-)
a)rl;4-H
00
00H-cnE
-(0
HO0N
O +
U14-1
@' H
g >
4
@ U
4-)(O
a0)
0
.
-0U
V
O
W V
O
4-H U
rO
a) 2
U
-H4
-) 1
0.H4-)4-)
HH
a)U0-L4
U-)
0
(N0Iw0C~
rHi
CIWCCLO
(NCDIwLC)
CN
Q)
r-H4i)
CN
00CNx0O
>o
0
0 +
C"0
>
1-op
H
0U
)10
,~4otO
4 Z
1,0 E-4
01
rn0
4 i00~
rI rN
o 0
o~
o
: U
( VO
V
CO
rH
V
) r.
a) w
0) >
1
0.HH,--I
IV.)
a3)
C-)
0H-
Ln
U).n
0
(-JQ(NJI
0LC)L
nC
NI
W00Ln.N
aI)
4-H
0
C"4
00
C,0
*( C,,'
0 +I,qulw
4>k
H
0oNO
N H
10 0C
-0' Eq0W0
IUd
Q)L
2
4J
0
'-4m'-I
( (1) r--l
r--iL
WH
4 r-
<
X
S
0.Hfri4-)
HUi)
.0a)a)
U)0H--
F'..
U)
U)
In
NI0OLn.
000Cq
00I
wLO
(N
U1)
.rA
0
(N00
CD
*rI
C)* N
ox0,{0+.qo
l X
*0
a>*rI
C'1
(nonE-4>
H
m-.
..4
4-U
)
z(N
H0
O Z
CS
H'H
C,
U
N)0
@C
) ~
it.i-( O
Z
4
4-i
0In
Inm
t W
C 0(O
4
-U
O
Q
) 4
4
S-/
0.L-)-i~
4
I-4HU)
0 )a-)
'3)
U)
0H4
FS4U
)u-.
ClIW0Ln.00ClI
W0Lncq
a)
4i
0
0,4
0 +
OD
C.
(N+
*H
U)
>
14H
O
ro >
^4
Eq
44 CA
q
CS*(N
J0-co 04
(Z)
O2
0) Z4
'-02
4 -)
0U)
,0,W)
O -)
U
0 (1) 44 44
S~
S5
To4
a
'-i
H
0)C)he
0)~~
~~
~~
U
C)~
~ ~
~~
~~
~~
~~
~C
Cl)~
~ ~ ~ ~
~~
~~
~~
~C
r-*4;
D~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~O
V
N
0 +
0~~~~~~~~~~~~~~~~~~~~a,~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
I _
_ __
..
__
_
_ _
__ _
H~
~~
~~
~~
~~
~~
~~
~~
~~
~~
~~
~~
~~
~,-
4-)4.)~~~~~~~.-0 0Z
4.
o~
~~
~~
~~
~~
~~
~~
~~
~~
~~
~~
~~
~~
~4.)
~ ~ ~
~
4
4;J-
)~ )
-Q
.4
J0
~
1U
5-3
0.-H4J4..)
HU)
0)0)0)r.)a)0.-1
U')
U)
0~
.jJ0u04-H(iia1)4
-)
---- ---
--- ---
---_
_
_
_
_
_ _
_
_
_
_
_
_
_ _
_
_
_
_ _
----
------ --
----
----
----
---
----
---
-
a)U)
'I)
a)-H4-I
0
00T-0 0
0+>
4-4U
CN0)L
I---U
L)0
0>0
-H
4J() -4(O
4-)- H
0 r
U0
~ -14
0 mo~
~~
~~
~~
~~
~T
L3)
0)
@~~~~~~~~~~~~
U'3
)~~
~~
~~
~~
~~
~'
g >
1
t~~~~~~~~~~~~ a
@~~~~~~~~~~~~
..0)~~~~~~~~~~~~~~~~~~~
U-) ~
~ ~
C0 ~~
U)
00 +
Q)
IH
---
ro~
~~
~~
~~
~~
~~
~~
~~
~~
~~
~~
~~
~~
-1-) (4
12
U)~
~~
~~
~~
~~
~~
~~
~~
~~
~~
~~
~~
~~
~~
~~
~~
~~
~~
~~
~~
~~
~~
~~
~~
~~
~~
14U.0
H ~~~~~o
-
0+(00r~E-l
Z4i
S
0HH4i
0tS4M0
:Q
):
ci)u
:r.
U]a):
,-q!
L)
L)
O
QcI0CDIWOIz1c-IHOC
I
-H4-ii
0
.~~~ ~~~ ~~~~~~~~~~~~~
o~~~~~~~~~~~~~~~~~~~~
cI-,4
o ~ o0
+.,,w
04
c n
*(N
0zo
H
o +s~
o
4i)
L)01-4
00
orQ
U
) -H
r 4J
W
-444 (d U
()
4J
a)
r:� Q
4
.
0H.41)4-)02
H3HN-
I@0)04@0v
Aua)0a).,,C
D3n
U7
UL
).
a)N.HU)
a)(/2U)
.H4iJ
a1)r4).H00
O
00-4>
U)
oO
x.o
0 +
h -i
E-1
..
'0>
s(N
H(SH
AC
,)
08
)2(
0U)0
00U-)
Lfl
4J-H~
~~
~W
(fl-H
N
O)~~~~C
1
000000
4i -r-I Fi
Q)
(12 -N
Q
)
/S~~~~~~~~~~~~~~~~~~~~-
0~~~~~~~~~
.Ii
>c, Q
i A
H
431
U)~
~ ~
~~
~~
~~
~~
C
U-)~~~~~~~~~~~~~~~~~~~~~~~~~~~U
u-i
H~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
CD
o 0c
LO
t
_~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~ +O
C)~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~U
oD \o no
C~D
E
ILn A
MC
) 0~~U
0
(N~
~~
~~
~~
~~
~~
~~
~~
~~
~~
~~
~~
~~
~~
~~
~~
~~
~~
~~
~~
~~
~C
0 )r
0)~~~~~~ co
a%0~
~~
~~
~~
~~
~~
~~
~~
~~
~~
~~
~~
~~
~~
~~
~~
~~
~~
0In
~~
~~
~~
~~
~~
~~
~~
~~
~~
~~
~~
~~
~~
~~
~~
~~
~~
~~
~~
r
-4 I-
-H
-H
a 4
(O
U
W
Z m
o (C
M)
0//
0VHH4U)
a)mU
)
0'-iF
L4
nIn
N0Ln(N0
121
pq
000
4-I
0~
0
00CA
U)
0+(n
NQ
h-
00
E
-E-4
C'4
cOh HW
U
,:~4-J
0m-
0
( (D
r--l r--lU
~~
r-Ir'P
I Q
W (d
4 J ir~
44 r--l �J
X
0 5
y4/~
~~
~~
~
oX
U£1:H
34U
) v
H
U')
CN
Cl)~
~~
~~
~~
~~
~~
~~
~~
~~
~~
~~
~~
~~
~~
~~
~~
~~
~~
~~
~~
CC
l)~~
~~
~~
~~
~~
~~
~~
~~
~~
~~
~~
~~
~~
~~
~~
~~
~~
~~
~~
~C
Cd
~~
~~
~~
~~
~~
~~
~~
~~
~~
~~
~~
~~
~~
~~
~~
~~
~~
IC
-H
Dox+
-H
1~ ~~
~ ~
~~
~~
~~
~~
~~
~~
~~
~~
~~
~~
~~
~~
~~
~~
- hD
(1
n CJ
0~~
~~
~~
~~
~~
~~
~~
~0
4~~
~~
~~
~~
~~
~~
~~
~~
~~
~~
~~
~~
~~
~~
~~
~~
~~
~~
~~
~~
~)
~ ~ ~ ~ ~
~
~
~
~
~
~
~
~
~
~
~
~
~
4
4-)~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~441
0~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~04
o~~
~~
r 0
®
C
0III
0c N
W
0 0
w
CO
(N
o4
0o o0
o~
o
i I
0o
o 0
0
0 0
0
w
(N
W
o 0
0
O
t cO
N
1o1 i
I 0
00
..
0I
IIM
OM
MEM
ERIM
2I
0
---'
O
44"'-
0
'' '
.
4O0
0 _
_ _
U ~ ~ ~
~ ~ ~ ~ ~ -
_____
3:
0' i
44
~ ~
~ ~ ~
44N
W
O0 ~
~ f
un r
o
o
o
o
'0 ,
-'-4~
~~
~~
~~
~~
~__
_ _.
A-|
44 '
-
X .-
0 ''.,""
0
,.,--0
s '"
X
V
0
0>
0
N
;* 0o 4II
.uN
H
X0
H0
044x
g
0
0O
0
0
o (N
*00.It
0~40P>4
~ O
p
w W
21'0H
30
40
N
D
N
W
0
0
D0
04
04
40
U ~
4'-)
o 0
0
0
0
0
L
1
-o
(N
04
*
0*o
4(N
NN
.4'*0 o
*
(
N
0 ~
0
II w
*
(N
U1
O >
o +
4o
*d->
o (N
*
0
HO
-o N
o Htrfl
0+m
C
lq 0II
0~
~
044
Z
o '-o-H
o V
n E
-i
II IIII0W
H-; 0
o
N°0
H0
0X
X
.4000903.
0
o V04
HiI 00
*0AI
11
Ha4t
II
_ _
-
iIC9*H
4N
II..
964t
III
0I,-
N
N
0
InN
,1
0
Go
03 0
I n(In
3
'4 m
N
(
co
I'1'-
(0
10
0'4
(H
.
r00
NInUm
N-1
0m
o 0o
co r-
o co
r" 3O
H
co
(0
r- In
'4 A
(N
(N
H
I
04.
433443
43
34
0Hi
03.
'
(N
_ ..
N
O
V33x
, '4
0
I0
0N
0
*
01
O
(N
0 00
_b
U)33
41
1111111111117.y�I
O o;
4.3
u +
43-
0.f
I
0'
NN
H
*00C
0
4.3
N* (
0
04.3
z
X
N
0
H
Cl
0
'310
0
Qw
C~4.1 M
3.zl
I
(N
m
H
0
Cl
N
(N
H
0
co3"`
((3
(
H
0D
0
03
0D
r'
(f
r um
w
(N
-r
10
1i
rN03tn
,I
H0
0
03 0
r0
0D
3-
'1H
0
10
(N
(N
H1
��w
�m
I01_
HH43I44
-H
33.4
N
O
0(N
N.34O
H04
.3
3 41X
Ix
H00(N
0
0
(*0
33
4.3
hH
14I14.
N
0
OH
fia1
3
.
143~
'0'-p~H
m.
IIIN
II