ML20196K643

From kanterella
Jump to navigation Jump to search
Rev 1 to B&W Document 32-1203121-01, Fsplit Certification Analysis
ML20196K643
Person / Time
Site: Three Mile Island Constellation icon.png
Issue date: 07/20/1991
From: Bowan B, Heimer J
BABCOCK & WILCOX CO.
To:
Shared Package
ML20196K642 List:
References
32-1203121-01, 32-1203121-01-R01, 32-1203121-1, 32-1203121-1-R1, NUDOCS 9904050025
Download: ML20196K643 (143)


Text

i ATTACHMENT II B&W Document No. 32-1203121-01 "FSPLIT Certification Analysis"

)

~

9904050025 990326 PDR ADOCK 05000289 P PDR a, 1

f l

~

g.

8mit 20HT2(Itsen SW#&.1 i .

ex===uu== ==r em m ioptnnEn 32-1203121-01 Tent FSPLIT Certification Analysis P#tPARED BY: AfVONED St

,, J.A. Heimer , B.L. BoAen SCNATURE  ! b $4NATUAE y,gg Pri ipal Engineer g/2,J,y Engineer IV ,, y , , g ,

37g a Cost CENTER IIET. PAGE(!) TM STATEMDfT REVIDWEn DCDUsoDeCE [

PURPOSE AND SLIMMARY OF RESULTS: b This document comoletely replaces revision 00 and describes how to use the hydraulic code FSPLIT and veriffes its accuracy for use on the NPR project. The document l includes a users manual, water property verification, ten FSPLIT check cases, and a I computer code listing.

Pace 2A sumarizes the revision made to versten 01.

i l

l THE ronowlNG COMPUTER coots HAVE BEEN USED IN THt$ coCUMENT: j coot /vtassoas / Rtv coot / vtassose / Rtv THIS Met MWT CONTA8Ns ASSUMPnoMS TNAT MusT BC vtnantD Muon To ust oN SAFETYJitLATED Wodet Yts( ) Not I) ,

l g

1(..cyeptA) 211

o

)

! i BWNV-20005 3(10/09)

NUMBER RECORD OF REVISION , , _ , ., n , , , , _ n , ,

REV. NO. CHANGE SECT / PARA DESCRIPTION / CHANGE AUTHORIZATTQN 00 None Initial Release 01 See page 2A See Page 2A 1

l i

l l

i i

i

\

a l

l I

DATE 8/12/91 PAGE 2

n:-

32-1203121-01 Revision 01 includes many cosmatic changes and three calculational changes. The cosmetic changes include:

(1) The " ESC" key will terminate the calculation any time during the "run" time and save last set of results to viewed on the screen. This options saves time when models are not converging adequately or allows the user to view some intermediate output (see page 36).

(2) Version 00 would terminate an FSPLIT run when a full disk was encountered during a results save. This caused all results to be lost and the analysis had to be re-run. Error trapping was added in version 01 to prevent this.

l l (3) The default pump interpretation scheme was change from #2 to #1. #1 is the

more universal of the two (see page 33) t (4) A flag was added to the pump H-Q input file to acknowledge whether the H-Q data is based on stetic or total (no recoverable DP's included) pressure drop across the pump (see page 20).

(5) Additional discussion on errors and limitations has been added to pages 11 and 12.

(6) The flow dependent loss factors require the user to specify which path

.(velocity head) the calculated loss factor is based on. Version 00 automatically assumed the " denominator" path. See page 31 for more details.

(7) The option to view a list of files in the " current" directory (when loading pump files or FSPLIT files into program) has been expanded to allow user to view files in any drive / directory (see page 24).

(8) The actual convergence criteria met for all local, pump, and heat iteration is now printed out (see Attachment 2).

There are three calculational change associated with version 01. An error was found 1 in the enthalpy prediction for light water. This error primarily affected only the high pressure / high temperature cases and the maximum error in version 00 was =.03%.

The correction reduced this error to zero percent for all pressures below critical and all temperatures below 660*F. The second calculational change was in the sequence of the DP calculation. Version 00 calculated the static pressure drop in a path, then calculated the unrecoverable DP. The difference in these values was the recoverable DP. A very small error in : the unrecoverable DP would then show up as a small recoverable DP (typically .005 psi or less). Version 01 reversed this calculation where the unrecoverable DP is now the derived value. The static DP is calculated,

' then the < recoverable DP. The difference is the unrecoverable DP. Finally, a reference for heavy weter subcooled properties was found just as version 00 was released (version 00 water properties are based on saturated properties only). A

. check of the calculated density at subcooled conditions showed very small errors when

. compared to the subcooled table. However, the subcooled enthalpy calculation revealed

.1% to 3% deviation .from these tables. Therefore, the calculation of heavy water enthalpy was modified to better agree with the subcooled properties (Reference 7).

l as Page 2A

l DOCUMENT NUMBER 32-1203121-01 1

FSPLIT - Version SA Fluid Flow Network Code by D A Farnsworth J A Weimer i 1

I ABSTRACT FSPLIT is a PC based tLermal hydraulic computer code designed for the generation and solution of steady state flow networks. The networks can contain any combination of up to 100 nodes and 100 paths. The problem may be specified by imposing up to 10 flow boundary conditions (external paths) or one pressure drop boundary condition. Problems may be defined with either temperature inputs or heat inputs to paths. FSPLIT uses g.raphical on screen modeling and parameter specification to greatly ,implity model development.

I In addition to individual path flow rates, FSPLIT solves for momentun. G uion.

friction, and form loss pressure drops. The form loss pressure drop >:alraias..m.

can assume a constant loss factor (one for forward and one for reverso flow) or it can calculate a flow dependent lo.3 factor. Equivalent heat rates are calculated for temperature input problems and temperatures are ca.lculated Fcr heat input problems. External path input flows may be constant or specified as ,

a head-capacity relationship to simulate a pump. FSPLIT is designed to l accommodate water (H 3 0), heavy water (D2 0), any incompressible fluid, and gasses.

FSPLIT has internal property routines for light water and heavy water while gasses and incompressible fluid properties must be specified by the user.

0 4

3 ,

e~

l Document Number 32-1203121-01 l

l l

CONTENTS l

PAGE 1

1

1.0 INTRODUCTION

........................................ 5 2.0 DISCUSSION .......................................... 5 Code Description ............................... S General Input Requi rements . . . . . . . . . . . . . . . . . . . . 6 3.0 CODE ACCURACY .............t.......................... 6 i tai cul ational Errors . . . . . . . . . . . . . . . . . . . . . . . . . . 6 Code Operability ............................. 8 Certification Measures ....................... 9 4.0 CODE LIMITATIONS .................................... 10 S.O FUTURE CODE ENHANC EMENTS . . . . . . . . . . . . . . . . . . . . . . . . . . . . 12 6.0 COMPUTATIONAL MODELS . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 13 7.e CODE INPUT /0UTPUT DESCRIPTION........................ 16

0.0 REFERENCES

.......................................... 40 ATTACHMENT I FSPLIT BENCHMARK CASES....................... 41 ATTACHMENT 2 FSPLIT RESULTS OF BENCHMARK CASES ........... 71 ATTACHMENT 3 FSPLITSA CODE LISTING ...................... 143 ATTACHMENT 4 HEAVY WATER PROPERTIES ..................... 201 l

l l

4 .

I

T Document Number 32-1203121-01

1.0 INTRODUCTION

While there are several computer codes available that can solve for flow rates and pressure drops in a network, most of these are large system codes that require mainframe computers. These codes are usually expensive analysis tools often used for relatively simple network problems. The more simple hydraulic codes available typically do not have all the capability needed to perform hydraulic analyses of the most often . encountered networks without using many manual iteration and intermediate hand calculations. Many hydraulic networks require code capabilities that may include the use of pumps (head capacity relationships), flow dependent form loss factors (Y's and TEE'si, heat inputs, or fluids other than light water. Since there are many applications for an efficient computer code with all these capabilities, the micro computer based code FSPLIT was developed to meet these functional requirements. This document discusses FSPLIT, how to use it, its mathematical models, and presents a com-prehensive error analysis of the code's calculations.

2.0 DISCUSSION 1

Code Description l FSPLIT can be used for pressure drop / flow solutions for networks with water, heavy water, incompressible fluids, or gasses. The networks can be closed loop or simulated open loop. FSPLIT can analyze forced flow and natural circulation problems. Up to ten external flow rate paths or one pressure drop path will initialize the problem and the code will solve for flow distributions and pressure drops in the internal paths. The problems initialized with flow rate can utilize constant flow or flow versus system pressure drop (head capacity relationships). The fluid temperatures can be varied between nodes by specifying '

node temperatures or path heat rates.  ;

4 In addition to pressure and flow information, FSPLIT calculates and displays all the pertinent supporting data used in hydraulics analyses such as velocity heads, friction factors, velocities, densities, etc. It also calculates available net ,

5

l l

Document Number 32-1203121-01 positive suction head (NPSH) at the (simulated) pump suction ncdes for problems involving light or heavy water. The maximum Mach number in a gas flow nrcwork is also displayed. If heat rates are input, temperatures will be calculated and j if temperatures are input, equivalent path heat rates will be calculated. All the pertinent pressure and flow output parameters and input listings can be j displayed on the computer screen, written to a disk file, or printed to paper.

FSPLIT uses graphical on-screen modeling and parameter specification to greatly  ;

simplify moGel development and reviewing results.

General Input Requirements FSPLIT models require a network consisting of a set of nodes and paths connecting them. The node input requirements are (1) elevation, (2) flow area and (3) temperature (if heat rates are not used). The path input requirements include (1) temperature (or heat input), (2) friction length, (3) hydraulic diameter, (4) path area and (5) fvrm loss factors (forward and reverse). External path boundary conditions (flow or pressure drop) are also required to initialize the probl em. Other user inputs that may be required, depending on the type of problem, include the magnitude of various convergence criteria, and flags for different types of iteration schemes. Section 7 describes the details of all the inputs required to develop and run a hydraulics model with FSPLIT.

3.0 CODE ACCURACY Calculational Errors The code solves for an exact solution of the Bernoulli pressure flow relation' ship. Tha , . ' uncertainties that can be introduced in the results of an FSPLIT calcula' .n aae due to the water property evaluation, friction factor, andthemathematicalconvergencescheme.

The water property that most significantly effects pressure drop is the density calculation. Per Attachment 1, the error in the FSPLIT density evaluation is 6

J

Document Number 32-1203121-01 l RfiEfIisiH@j[f(3Ej]jjt3@' and on the order of .02% for heavy water.

This results in a pressure drop error of the same magnitude. The maximum possible error (at the more extreme pressure and temperature conditions) for l light.and heavy water is less than .01% and 0.2% respectively. Errors in the other water properties (enthalpy, C,, viscosity, and conductivity) range from 0%

to =3.0% depending on the property and temperature / pressure range. These  ;

i properties typically only have a secondary effect on pressure drop and flow j calculations. For example, the enthalpy error will only affect average density  ;

in cases where heat is input (compared to inputting temperatures). A 1% error j in temperature will alter density (and therefore pressure drops) by less than

.01%. Viscosity errors will have an effect on Reynolds number and therefore the I friction factor. The impact of viscosity error on total pressure drop will be j effectively zero in the high Reynolds number range (where friction factor is constant with Re Number). At laminar flow conditions, the error in the DP calculation could be as high as 1% (light water) or 2.5% (heavy water), however, I the uncertainty in the Moody friction factor itself will dominate the friction pressure drop this flow regime. Note, that this error is only on one part of the pressure drop and therefore, the error on the total pressure drop will be less.

Finally, the fluid property errors in problems involving incompressible fluids I and gasses are a function of the accuracy of the user input property tables and therefore up to the user to determine.

1 The friction factor is based on the equation describing the Moody friction vs Reynolds number (Reference 5). There is essentially no error in this calculation other than any uncertainty associated with the original Moody data.

The iteration scheme introduces error as a function of the magnitude of the convergence criteria used. A large convergence criteria will introduce more error. However, the default convergence criteria effectively introduced zero 1

l Y . kl$k NhA%EONUTYoliE55 Mh6d 9 'Ei J3(6i($ffeitfielj@2ero Pegggnt egyg , EL Jjg!!si 7

l

l l

Document Number 32-1203121-01 l

l percent error for all the test cases analyzed.2 Very simple networks may l converge to an effective zero percent error with relatively large convergence criteria and more complicated networks may requir: -11er convergence criteria.

However, the default convergence criteria ,. be acceptable for most I networks.3 1

Attachment I discusses the details of the water property comparisons and ten i benchmark cases used to evaluate FSPLIT calculational uncertainties. The benchmark cases were compared to exact solutions or known results. When the water property errors were eliminated, the calculational errors were effectively zero percent. In general, the hydraulic results of an FSPLIT calculated flow network are accurate, by any standards, and are acceptable for any nuclear or non-nuclear design application.

Code Operability i

Many test cases have been checked with FSPLIT. However, every possible combination of options cannot possibly be checked since there are virtually thousands of potential combinations. If a convergence problem arises with a  ;

particular analysis, check this document for some guidance (decreasing the convergence criteria, or changing the pump iteration scheme type, or changing the relaxation criteria will solve many of the problems). If the problem cannot be solved by changing various switches, better noding, etc., contact one of the authors for further help. If the code prematurely terminates (returns to DOS) when a problem is being analyzed, check the input for absurd conditions (like excessive pressures, temperatures, etc.) or check that an input file (pump head 2 ~

ncescRifrla canypfiIsy311g~MM7d{fferentWiYsit3

[$Ns#

Etdih%h(N$y$nartingiffomMrestMt3file ifey$tljisffTg(

Msl i ttingsfindtheistinRSPgifQe)Q3 diffirinuss Ms _ _ ,__ EEf#fRR5Eg 3

There is a code output that flags any one path with a greater than .01%

flow change or pressure change between the next to last and last iteration (even if overall convergence is met). This flag will help indicate when a smaller convergence is needed (see discussion in section 7).

8

Document Number 32-1203121-01 capacity, properties teble, or flow split tables) is correctly formatted. A DOS termination will' display one of the following numbers. This table will help determine the error.

Run-Time Error Codes Code Description Code Description 3 RETURN without GOSUB 54 Bad file mode 4 Out of DATA 55 File already open 5 Illegal function call 56 FIELD statement active 6 Overflow 57 Device I/O error 7 Out of memory 58 File already exists 9 Subscript out of range 59 Bad record length 11 13,Divgionbyzero 61 Disk full' thttg sfiniisticN 62 Input past end of file

- b_14"9&",diffor s@frTn~g~~siiEe 63 Bad record number 1

16 String formula to complex 64 Bad file name l 20 RESUME without error 67 Too many files 24- Device timeout 68 Device unavailable 25 Device fault 69 Communication-buffer overflow 27 Out of paper 70 Permission denied 39 CASE ELSE expected 71 Disk not ready 3 40 Variable required 71 Disk-media error 50 FIELD overflow 73 Advanced feature unavailable 51 Internal error 74 Rename across disks 52 Bad file name or number 75 Path / File access error 53 File not found 76 Path not found Certification Mea;ures There are two situations where an FSPLIT calculated result may not meet a minimum certification requirement. The first situation will occur if an obsolete version is used. Even if a new version involves only cosmetic changes or only adds new options to the program, the most recent version must be used for every calcula-tion. Therefore, the user is required to check with the appropriate sources (listed in the computer input and output sections) before each use to assure that the version being used is the most recent. Also, it is a responsibility of the calculation reviewer to re-check for the correct version.

9

l Document Number 32-1203121-01 I The second situation where certification may not be met could occur if a particular computer copy of the code has been corrupted during a file copy process or someone has tried to modify the executable (binary) file. In order to prevent this problem, FSPLIT does a checksum on the binary file prior to the first input request. If the checksum does not match a prescribed result, the code will not operate. Changing one bit in the coding will cause that copy of FSPLIT to be inoperable. Note that the program EXECHK.COM must be in the same 1 directory as FSPLITSA in order for the code to make this check. If this file is not included, the code will not run.

These two certification procedures should assure that all legitimate use of this code will be acceptable for nuclear standards.

4. CODE LIMITATIONS Because FSPLIT is PC based, there are some limitations (compared to mainframe codes). Most of these limitations are associated with the amount of memory available and microprocessor speeds. However, as future updates to the code will include the use. of expanded memory and faster microprocessors, many of these limitations will be minimized or eliminated. Some of the other limitations of FSPLIT (discussed below) are planned to be addressed in future upgrades of the l code as resources and time become available. j i

Most of the code's limitatioas are discussed in detail in the users manual (section7). The following is a summary and brief description of all the code limitations.

Computer Requirements to Run FSPLIT

- At least 555K byt' f usable RAM

- Math co-processor

- DOS 3.1 or greater 10 l

F Document Number 32-1203121-01

- FSPLIT.HLP and EXECHX.COM must be same directory as  ;

FSPLITSA.EXE in order to run the program. j Unless most resident programs (TSR's) are stored in expanded memory, they will  ;

likely need to be eliminated from most computers in order to have enough memory to run FSPLIT. A l6[{@MMtliiEIpj]MadsdiisiM{ay[gsi$filbj@y66@ld

]

F1MW0lTWidWiTficl3F5$)iTIlil General Model Limitations

- maximum of 100 nodes and 100 paths

- at least I and not more than 10 external paths (or pumps) l

- limit of one form loss per path (in each direction) )

- each node must have at least one path connected to it SN15NENk5NNN) (kN5EEN$

Fluid conditions

- single phase fluid

- subsonic flow

- water temperatures ;t 50*F and < Tsat'

- water pressures > 1 psia and < 3200 psia

,FSPLIT is presently limited to single phase fluids at subsonic velocities. If sonic velocities or saturated conditions (for water) are calculated, the code

. will flag the user that the results may be invalid. If the fluid is a gas, the user input property table defines the minimum / maximum ' temperature and pressure.

Exceeding these limits will be flagged by FSPLIT. Any calculated network Heavy water properties have been verified to 500*F. If a problem uses heavy water above this temperature, a warning will be printed that advises the user to check the water properties being used by FSPLIT.

11

Document Number 32-1203121-01 pressures that are less than zero psia will be acknowledged in the FSPLIT output.

Pressures greater than =3200 psia for water will cause the FSPLIT code to terminate.

Mathematical

- no closed loops with zero pressure losses i

- various convergence criteria, relaxation criteria, and iteration schemes may need to be varied (user input options)

- If one external path is to be a head capacity relationship, all external paths.must be head capacity curves.

((_If[u]3 3 M i[il @ M EoW ${$5 M E D W 3 Ejjii$ 3{RlJf @}25 5fjj A closed loop with zero pressure resistance will result in an infinite flow rate  :

which will result in an overflow condition and FSPLIT will terminate. Also, networks with many parallel paths may not converge with the default relaxation criteria (0.5). The solution may flip-flop on each iteration and therefore, never converge. Lower relaxation criteria will alleviate this situatien (at the expense of running time). However, if convergence is not attained with a 0.2 relaxation, the model may need to be modified. For large models where few parallel paths are used, the relaxation can be increased to speed up the solution. This is particularly useful for model that may need to be re-run many times with modified inputs. Finally, the slope of the head capacity (HC) flow input may cause a problem not to converge. There are two HC convergence schemes (discussed in section 7) to choose from. If one does not wor,k for a particular case, try the other.

5. FUTURE CODE ENHANCEMENT 1 There are many additional capabilities and improvements that can be added to FSPLIT as the need arises and time and resources become available. This section 12

f 1

( Document Number 32-1203121-01 lists some of these possible modifications. As any of these are included in the i l

l code, this document will be revised to reflect the improvements and the code version will be changed. <

l 1. Compile code to make use of expanded memory in order to increase the l number of usable paths and nodes.

l

2. Allow for more than one external path with a specified pressure drop boundary condition.

4

3. Allow for a combination of pressure drop and flow rate boundary conditions.
4. Add two-phase flow ca to the drop calculations

}TMWj@@ligal@pabilitiesjjp@ @[fp W 3515 @pressurely@fkid 1

5. . Allow for quasi-transient conditions (varying pump discharge or suction boundary conditions with time).
6. Accommodate true transient problems.
7. Add built in fluid properties for common gasses (hydrogen, air, etc.).
8. Expand the heat transfer calculations to include different types of ,

correlations, flow dependent coefficients, and other hut transfer enhancements.

9. Include additional graphical capabilities to expand the display of the l output variables.

These upgrades will expand the capabilities of FSPLIT to be able to analyze larger problem and some additional types of thermal-hydraulic problems.

6.0 COMPUTATIONAL MODELS FSPLIT develops a set of equations, one for each unknown parameter. These equations are assembled into linearized matrix form. The solution of this matrix is obtained by using the LINPNCK general solution routines (Reference 3).

I l 13

g i

Document Number 32-1203121-01 Momentum Solution For a given network of N nodes and P internal paths, there are N

{

unknown pressures and P unknown flows. Each internal path provides one equation  ;

relating the pressures of the connected nodes. Each node provides an equation relating the entering and exiting flows (including those of the external paths). I The form of the internal path equations is (Ptot)in-(Ptot)out = HL where Ptot = P + VH + pgZ (total pressure)

VH = pVr/2 = W2/2pA8 (velocity head)

HL = (K + fL/D) VH (unrecoverableheadloss) l and W = path flow V = path velocity i

p = density (path or node)

A = path area t At any point the total pressure is the sum of the static pressure, the velocity head and the elevation head. The unrecoverable head loss between two nodes is the sum of the shock (K) and friction (fL/D) losses. The path equation is then put in matrix form as; (1] E P 3 7 3 {61t M [ Q Q J j y}]l where R = (K + f1/D)/2gpAt (flow resistance)

B = VHout - VHin + pgZout - pgZin ,

The form of the nodal path equations is 4

[2] Wout - Win - Wext where Wout = sum of internal flows out of the node Win = sum of internal flows in to the node Wext = net sum of external flows to this node 14

Document Number 32-1203121-01 Equations 1 and 2 are combined into a linearized [A] X = B system of equations. '

The solution vector, 8 (the nodal pressures followed by the path flows) is then appropriately under-relaxed and checked for convergence. If convergence is not achieved, the system is relinearized based on the current solution and this process continues until a converged solution is obtained.

{

Energy. Solution If the problem has path heat inputs, an energy solution must be obtained. ~Here there only N unknowns, the nodal temperatures. An energy balance is performed at each node to form the system of equations. l nodal energy - energy in (from flow) = heat added (path heat) j

[3] WN x HN - WP x HP = Qadded i where WN is the nodal flow HN is the nodal enthalpy l WP is the incoming (path) flow HP is the enthalpy of the incoming flow Qadded is the net heat added (+ or -) by all paths flowing into the node. (There can be multiple incoming path energies.)

A path heat input (either positive or negative) is considered to be added to the node if flow from that path is into the node.

The system from equation 3 is already linear (with enthalpy) and is solved directly. If the properties are a function of temperature, the momentum solution is repeated. If there is a change.in the momentum solution, the energy solution '

is redone. When there is no further change in the momentum and energy solution, convergence has been attained. -

External Pump Solution With external pumps (head-capacity equations or tables) included in the model, FSPLIT performs the solution procedure as follows. It finds a momentum solution 15 l

I Document Number 32-1203121-01 (including an energy solution if appropriate) based on the most recent external-path flow rates.

It then compares external path pressure drops with the input head capacity ir, formation. If they do not match, FSPLIT modifies the external path flows and repeats this procedure until the calculated head is equal to the input head (/or the same flow).

l Programming Considerations FSPLIT is written in a combination of Microsoft QuickBASIC and FORTRAN.All ca).culations are double (64 bit) precision and require a math coprocessor. The

aatrix solutions are performed using the LINPACK gensral solvers. The allowable problem size is limited by the size of the main solution matrix which, in itself, requires 316 kilobytes of storage. Presently FSPLIT requires approximately 555 kilobytes of free DOS memory. Consequently, all unnecessary memory resident programs should be unloaded prior to FSPLIT execution.
7. CODE INPUT /0UTPUT DESCRIPTION The following discussion describes how to develop models and perform analyses I with FSPLIT. 'All of the code input user option discussed below are menu driven and, for the most part, are self explanatory. Many of the questions a user may have are answered via two " help" menus in the program.

The general approach for solving a hydraulics network problem with FSPLIT is to; (1) outline the paths, nodes, and hydraulic inputs of the problem, (2) enter FSPLIT (with the relevant options) and sketch the flow paths and nodes on the computer monitor, 4 (3) at the appropriate prompts, input the needed hydraulic data for each node and path, (4) run the problem (typically with a large convergence criteria to save time),

l 16 i

Document Number 32-1203121-01 (5) review results, (6). make any modifications or corrections to the network and rerun with the default convergence criteria, and finally print /save the results.

  1. 1 Outline problem Sketch the network of your problem including all pertinent thermal hydraulic parameters (temperatures, valves, elbows, pipe diameters and lengths, elevation chtuges, pump locations, etc.). Define the portions of the network that are to be in each path. Sketch the paths and nodes assigning sequential numbers to each node and sequential numbers to each path. Determine all form loss factors in each path.- Since only one form loss factor per path is allowed, some components may have to be combined for shock loss calculations (for. example, a valve and an elbow may be in the same path and therefore their loss factors must be mathematically combined). Calculate a loss factor in the forward flow and reve'rse- flow direction (if they are different). If a flow restriction orifice causes a substantial flow area reduction in a network, make this component one separate path using the flow area of the orifice. This modeling technique will allow FSPLIT to determine if a sonic velocity (or cavitation) situation has been reached in a path. Next, determine the type of fluid and fluid temperatures (or i heat input or output) for each path. Decide if the boundary condition for the problem is to be (1) a pressure drop across one external path, or (2) a specified flow across one to ten external paths, or (3) flow determined by head capacity relationships in external paths (if pumps are included).

In addition to the network parameters, other inputs ma*y be required depending on the type of problem and the options selected. If pumps are modeled, a file containing a description of the head capacity must be developed. 'This file can

)

describe the flow pres'sure relationship by any of the three methods discussed I

below. The file must be in ASCII format which can be developed using any text ,

~

editor or word processor. The details of the pump file are discussed in #2 below. ~If the fluid to be used is a gas, either constant fluid properties or 17 l

I

Document Number 32-1203121-01 fluid properties as a function of temperature can be input. Finally, if any flow branches (TEE's or Y's) are in the network, flow dependent loss factors may be needed. In order to better model a variable loss branching flow, include three noss (one just inside each branch of the TEE or Y) in the network. The inputs for the flow dependent loss factors are discussed in detail in #3 below.

  1. 2 Enter FSPLIT After the model is developed on paper, type FSPLIT5A S (in the appropriate PC directory). A screen will appear that has a brief description of the code, some of the general limitations, and a request to check for the latest version release. The user must check with one of the two authors to assure that the latest version is being used. (This .is a Q/A recuirement to use this code for any analysis requiring certification.

Failure to do this may invalidate any analysis.) A SIGN WILL BE POSTED OUTSIDE THE OFFICES OF THE TWO AUTHORS THAT WILL SHOW THE LATEST VERSION AVAILABLE. Also, an internal checksum numbe'r will be calculated for the version being used. If this number is not correct, there may be something wrong with that particular executable file and that copy of the code will not execute. This is a check to assure that the particular copy of the code is correct.

After pressing " enter", the following messages will appear (one at a time) and the appropriate response should be given. Note, that most choices have a default response that will be chosen by just pressing " enter".

(E)GA, (V)GA, (C)GA, or (Hjerc Monitor (Def E) ?

Is this an external pump case (Y/N - Def N) ?

'Is there heat input in any path (Y/N - Def N) ?

5 NOTE THAT TYPING CTRL-BREAK (AT THE SAME TIME) WILL EXIT FSPLIT (T0 005)

FROM ALMOST ANY PART OF THE PROGRAM. THIS CAN BE USED TO TERMINATE A PROBLEM IF DESIRED.

. 18

i I .

I l

l l Document Number 32-1203121-01 If a VGA or EGA monitor is used, the system graphics model (on the screen) can l be larger (due to more screen pixels available). If the case involves heat input l (as opposed to temperature input) the format of the input files will be l different. The correct response to this question assures that the proper file l

format will be read (or written). (Note, that after a model is developed with I or without the heat option, this option cannot be changed when reanalyzing the model at a later time. Changing this option will cause FSPLIT to terminate when the file is being loaded.] If the network includes pumps ("Y"es to question 2),

the next screen FSPLIT displays will be; One head-capacity curve is required for each l pump. The three allowable forms are: The same form must be used for all pumps.

1) HEAD (Ft) = A + B
  • GPM
2) HEAD (Ft) = A + B
  • In(C - GPM]
3) HEAD (Ft), GPM data files (tables) ump coefficient file is: 1 The form for 0) the p$TYPERIIsM0/1{lpMi{p2[ol3aDj$

ffof{MPff@*

If' Type f/f - COEFFICIENTS ( A,B,' C) for each pump If Type 3 - # of' Data Pairs for each Pump, then Data l Note that if the case is run with recoverable l pressure drops, pump suction density will be  ;

used with the head-capacity curves - otherwise the density in Path I will be used.

... press any key...

This screen describes the types of head capacity (HC) files that are allowable.

The first is simply a straight line. This typically will be used where the i resulting flow - pressure drop relationship is known to be in a small range of the head capacity response where a straight line representation is sufficiently accurate. The second is a logarithmic representation (which can be generated automatically by the code FREELANCE) that was found to best fit a typical head capacity curve. This fit is best used when a high degree of accuracy over a wide range of flow rates is desirable. The third type is a table input of head verses flow. FSPLIT will linearly interpolate between input points.

i 19

1 l Document Number 32-1203121-01 i l l These three forms of HC responses describe system pressure drop (head in ft) as 1

a function of pump flow (in GPM). [FSPLIT will convert head and capacity to psia j and lb/sec based on the external path temperature and pressure]. HQifNgijQtj MIINdEk$ELDEd5513E15 SIN 3SIIINE$$AM336E2PEdidOfNE[A$@3 IHiroggsrtartr6g!!gtgiraggtTsaisirpIggsri~arreavagEEEmirgtgrgig  !

E74Encamssamriipsaswantarsiitwgla witsiirtrgrgwigpIpsroy t1lisriaisiPggBRistdgfewgJ. jripaEtEgitsiirguitJpsenTII1rdsIfg

~

!!MiffilittiiiiQiirt6siNiggMiEq3!sSEip3 age'c2ov4fa11'Egghd#desfelekTji

  1. ECHEnwaGFIME*e**sibEE!FId l The user must generate one of these types of pump files (in ASCII format) outside of FSPLIT.

For input type 1 (Head - A + B x GPM), the pump input file will look like the l following. l R@@lof@ggj8{j[jhIgy]$3}j]i[if))UdMg{f@@gl[eTQ A3 B, A, B2 A, B, For input type 2 (Head = A + B

  • In[C - GPM]), the input file will look like the following.

M[diii6ii@fMJjjfdffpM!?$pjM1(Mi[R@l@ig[fifsfapy[hiidllfy  : A, B, C3 Ag B3 C,  ! A, B, C, For input type 3, the input file will look like the following. i

f l 1 Document Number 32-1203121-01  ! I EM9Istli=NeEMireIFIEWelilii!PifiniEihl (number of data A3,3 B,,, 0102..o,pa irs for each pump)

       ^1,o1   B3 ,,,

Aa,3 B2,1 A,,og B2,02  ! A,,3 B,,5 Age, B,,,, l FIGURE 1 A set of "D" inputs (number of HC So" pairs) must be input for each of ,,, i "N" pumps. If a network flow c. . pressure drop solution occurs be- 9 eso - o i i fore the first input data pair or {"o after the last input data pair, ,,, _ FSPLIT will assume a H-C relation-ship per the dashed lines in '* ~

                                                                                             ~~~

Figure 1. This may not be a so; , , , , , ,,,,, 3,,,, realistic response, however, FLOW AATE FSPLif will at least execute and the user can then identify the input problems and correct them.

                                           }jf;[Ge@7nde~dffliitTajj HCidat(pai rj be                      .
 }$5EiiWi[SIRY##approYisife]J3 ego 1@Mla6WsREjlTn3H[E6e]jg!Used 2 Just befors the final execution of the network is performed by FSPLIT, the user will be asked for a pump multiplier. This will allow for all the HC input files to be adjusted to a different HC relation without altering the pump file. The new relationship will be;          Head =

f(Mult x Flow). For a given flow, the 21

m 4 Document Number 32-1203121-01 corresponding head will be increased (or decreased) by the multiplier. The next input requested by FSPLIT will be the fluid type per the following. (L)ight Water, (H)eavy Water Incompressible (F)1uid, or Incompressible (G)as (Default L) If incompressible fluid is chosen (fluid will have a constant density, viscosity, conductivity and specific heat for all temperatures and pressu es), the user will input the values for each of these parameters in the fc11owing screen. Reference Density, Lbm/Ft^3 ( Def 50 ) Reference Viscosity, Lbf-Sec/Ft^2 ( Def .000001 ) Reference Specific Heat, 8tu/Lbm-F ( Def 1 ) Reference Thermal Conductivity, Btu /Hr-Ft-F ( Def 1 ) If the user presses " enter" for each request, the code will use the default value. (Note that this option can be used for large problems that use constant

   ,    temperature water. This will significantly reduce the computer running time.

FSPLIT runs much faster if it does not have to do a property calculation for each path and node for each iteration.) This option can also be used for liquids other than water. If the "G"as option is selected, the user will have two sub-options for properties. The first will be to select a set of data from an input file. This will be an ASCII file (again, generated by the user) that defines specific heat, viscosities, and conductivities for various temperatures. The second option will allow for a set of constant pr6perties. Both options will require a user input of C/C, and the gas molecular wefght.. These inputs are used to calculate the limiting sonic velocities in the gas. Gas Molecular Weight (def 29) Specific Heat Ratio, CP/Cy (def 1.4) 22

                                                               ~

Document Number 32-1203121-01 The first option for an input file will prompt; File containing Gas Properties (Enter for constant properties) ? If a file name is input here, the gas will exhibit all properties as a function of temperature except density which will also be a function of pressure (using the perfect gas relationship). The pressure that the input table is based on (for the perfect gas relationship) will also be requested (see Base Pressure below). The first lias of the input file must have a 1 or a 0 in the first column. A "1" designates that the gas properties are in SI units and a "0' means British units. The remainder of line one can have a title (if desired). The subsequent lines must contain (in this order), temperature (*F, or *C), density (1bm/ft8, or kg/in8), viscosity (lbf sec/fte, or N sec/m r  ? thermal conductivity (BTU /hr-ftCr/.F, or W/m *K), and specific heat (BTU /lbm *F, or cal /gm *K). Each line in the file will contain an input for each of these values with at least one blank space between inputs. A maximum of 50 line of property input is allowed. Be sure that there are no blank lines at the end of the file as this will cause an error when the file is read. FSPLIT will flag the results on any gas temperature that is lower or higher than the minimum and maximum temperatures listed in the input file. (This attribute can be used to determine when superheated steam is cooled to saturation.) After the input file is read, FSPLIT will request the base pressure. Base Pressure (psia) If the user presses " enter", at the file request prompt, all the gas properties (except density) will be assumed constant in the network and the code will request the following inputs. 23

r-l l l l Document Number 32-1203121-01 l l Reference Viscosity, Lbf-Sec/Ft^2 ( Def .000001 ) ! Reference Specific Heat, Btu /Lba-F ( Def 1 ) Reference Thermal Conductivity, Btu /Hr-Ft-F ( Def 1 ) Reference Temperature, Dag F ( Def 0 ) As before, the density will be a function of pressure based on the perf<tet gas law. After a few seconds, the following message will appear on the screen. An incompressible gas is defined as one in which the properties are functions of of temperature only. In FSPLIT, only the density is a function of temperature.

                   ... press any key...

The next FSPLIT request on the screen will be for the pump file name. FR@[#AMiff{@j@[@@@@6/n61E(@[ESIMPitMWJjffiliiiB 1 i Be sure to include disk drive and directory that the file is located in (if it . is not in the same directory as FSPLIT). The code will not continue until it ' gets a yalid pump file. ffFJ1@31MEsli,1111sffilisifDig!(igheHdj{df$ij

  !NSE!EE$I5EENN[NtE$$M Next, the request for the problem file (or new case) will appear.

N00EL INPUT OPTIONS Type Filename _to 1oad disk file TypjjfREsitte2i,MEMEsEYi1E(E!!jlE5]QLXM Press ENTElf to create new modil RESPONSE 7 If an input file is to be used, remember to include computer file location if it is not in the same subdirectory as FSPLIT. After the file is loaded, a checksum number will appear. Thit is the checksum of the input file. The purpose of the check is. to verify that the input file has not been changed from a previous analysis. If the checksum is the same as the number printed out from a previous analysis performed with the same file, the file has not been modified. After 24

i Document Number 32-1203121-01 j pressing " enter" a diagram of the model will appear on the screen. l If a new model is to be developed, press " ENTER" after the above request. A screen will appear with the following note at the top. N0DE CREATION - F9 TO CREATE PATHS F10 FOR HELP A "<" symbol will also appear on the screen. This is the cursor position on the screen and is used to place the network's node location on the screen. Pressing "F1" will put a number at the < position. Pressing the arrow keys will move the

 < around the screen and pressing "F1" again will put the next squential number   j on the screen. Continue this process until all the problem nodes are on the screen in the desired positions. (Note that an EGA or VGA monitor will allow 380 ;

l node positions on the screen while CGA monitor will allow only 209 positions.) It is recommended to space the noding out as much as the model and the screen will allow in e eder to be able to better discern between node and path numbers. If some rearranging or changing some of the node is desired, the "F2" and "F3" keys can be used.' This procedure is described in the help screen "FIO" per the following. NODE CREATION HELP Use ARROW and DIAGONAL KEYS for cursor movement. Use F1 KEY to position a node at current cursor position. Use F2 KEY to remove any node at the current cursor position and then reposition it with the F1 KEY. Use F3 KF_Y to delete the last node only. Nodes created will be the next highest node unless a node has been removed with the F2 KEY; then that node will be recreated at the current cursor position by the F1 KEY. Press the F9 KEY to START CREATING PATES between nodes. Note that you can switch back and forth between the PATH CREATION MODE and the NODE CREATION MODE, but that you If a pressure drop is to be specified in an external path, assure that l the noding is such that the highest and lowest numbered nodes will bound the external path. This will be discussed in more detail in the next section. ! 25

4 Document Number 32-1203121-01 can QUIT or SAVE files only in the PATE CREATION MODE. Note also that you can RUN the completed and savs;! fil9 only from the PATE CREATION NODE. 4 NETWORK GEOMETRY is specified in the PATE 4:REATION MODE.  ! P8PLIT was written to be self documenting. With the use I' this help screen, the PATE CREATION help screen and some experimentation, model creation should be possible. If further problems arise, consult the manual. l

                                    ... press any key to continue ...         j l

Be sure that all the nodes on the screen will te included in at least one path otherwise the code will not execute. After all the nodes of the problem are sketched on the screen, press "F9" and the following note will appear at the top and bottom of the screen. PATH CREATION - TYPE "MORE" FOR NODE CREATION " HELP" FOR HELP PATH 1 - INLET NODE > FSPLIT will draw a path'(and label path number) between nodes by typing the " INLET" node and " OUTLET" node numbers at the prompt. The direction of the path will be from inlet to outlet nodes. Reverse flow and pressure drop in , a path will result in negative flow rates for this path. Also, different form loss factor inputs will be requested for forward and reverse flow. Like the node numbering, the path numbers will be sequential. DO NOT DEFINE A PATH BETWEEN NODES THAT ARE TO BE EXTERNAL PATHS. (Also, if at all possible, model only one path into and out of each external path node.) This portion of FSPLIT also has a help option accessed by typing " help". l PATH CREATION HELP i In this mode, paths between the nodes are created. Type either node numbers or commands at the ' INLET N0DE >' prompt. You can switch back and forth between the PATH CREATION and N00E CREATION modes until the model is finished. The CONMANDS in this mode are: HELP - Calls this screen SAVE or CREATE - To create /resave the FSPLIT input file. ERASE or DELETE - Prompts for a path to remove from 26

! Document Number 32-1203121-01 the model; paths will be sequentially renumbered. REDRAW - Redraws the graphics screen. N0 PATH - Redraws the graphics screen showing nodes only. NON0DE - Redraws the graphics screen showing paths only. SHOW - Shows the numerical model on screen. SHOWREC - Shows the recoverable information on screen. SHOWEXT - Shows the external path information on screen. MORE - Switches back to the NODE CREATION MODE. RUN - Returns to main FSPLIT program to RUN the model. QUIT or END - Terminates Program (must answer 'YES'). PRINTDOT/PRINTLAZER - Copies Screen to printer /lazer jet. NETWORK GEONETRY is specified when the SAVE command is used. 1

                                           ... press any key to continue ...

Paths can be deleted by typing " delete" or " erase" and the program will prompt the user for which path tc delete. The other commands available in the path creation mode are listed above and most are self explanatory. The "Show. . " commands disolay any previously input file data on the screen. (Note that the forward form loss factor will be shown unless the model was run and the flow was reversed. In this case, the reverse form loss will be shown.) If the nodes are placed on the screen such that the path locations are confusing, the user can reposition the nodes by typing "MORE". This will put the user back in the node creation screen. Using F2 and F1, the user can reposition the nodes on the screen for clarity. Typing F9 will put the user back into the path creation mode and automatically redraw the paths to the new node positions. PRINTLAZER or PRINTDOT will print the network drawing to the appropriate printer. If the print request does not work, the computer may be set up for a different port than the default in this program (parallel 1)7 . For large or detailed models, execute "nonode" or "nopath" before printing. This will print only the node numbers or path nuihbers on a page. l 7 If this is the case, a mode change outside FSPLIT in the DOS environment must be made. The format is MODE LPTI:-[YOUR PORT]: l A typical DOS command is MODE LPTI:=COMI: l l 27 l l

Document Number 32-1203121-01 A typical network diagram wi?1 look like the following. l l 1 8 5 9 7 b 6 z e i Q M 5 z Q s w s 9 2 4 d.' ,

                                        .       2      .1      3
                                   #3 Input Hydraul'ic Data After all the nodes and paths of the model are sketched on the monitor, type "3 AVE" or " CREATE" (from the path creation mode). The following will appear on the monitor.                                                                        !

1 - SPECIFY ALL NEW PATH INFO 2 - CHANGE ONLY SELECTED PATHS 3 - SAVE FILE AS IS AND RETURN TO MODEL (ONLY IF TYPE PREVIOUSLY SPECIFIED) ENTER 1, 2 or 3 ? Since this is a new model, select choice 1. Choice 2 will be used to modify models already developed. (Note that choice 3 will save the model to a disk file. It may be prudent to save the paths and nodes. created to this point as a safety factor and then retype "save" and then enter choice 1.) The following l screen will appear. 28

I l Document Number 32-1203121-01 NEW MODEL GENERATION l I TYPE SPECIFY SPECIFY VARIABLE RECOV FLOW DELTAP TEMPS DELTAP  ;

           ............................................_____                                             l 1            YES                                  YES         YES                         I 2                                   YES            YES         YES                         ;

3 YES YES i 4 YES YES l 5 YES YES 6 YES YES 7 YES E YES l X PATHS CONNECTING Y NODES WHEN A PRESSURE DROP IS SPECIFIED, ONLY ONE EXTERNAL PATH

                                                                                                         )

IS ALLOWED. THIS PATH MUST BE FROM THE HIGHEST NUMBER NODE l TO THE LOWEST (1). l Type of analysis (1 to 8) ? j Select the number that describes the problem to be analyzed. If flow is to be specified, it can be either a constant flow (in each external path) or a pump head capacity relationship. For these cases options 1, 3, 5, or 7 would be selected. If a pressure drop boundary condition is to be specified, only one external path will be allowed and it must be from the highest number node to node 1 (options 2, 4, 6, or 8 would be selected). [ Note that flow will be in the direction of the pressure drop. If a reverse flow is desired, input a negative , pressuredropwhenrequested.] Selections S through 8 are the same as 1 through I 4 except no recoverable pressure drop data will be input or calculated. This will allow large models to execute much faster in cases where recoverable pressure drops are negligible or not required in the problem solution' Alsc, if a constant temperature option is selected, the fluid density at every nod. and path will be constant (based on the external path tempe-ature) even if haat is a If recoverable DP's are not calculated, the calculated available NPSH l will be based on the total flow in the external path, the sum of all the path areas entering the external path, and the average of the path temperatures entering the external path. 29 l

Document Number 32-1203121-01 added to a path. It is not recommended to use the heat input option or gas option in conjunction with a constant temperature flag. l The next two messtges on the screen will be; Heat Input (Y/N - Def N) ? Case title (up to 65 characters) The heat input request is needed to sa'fe the input data in the correct position in the file and the case title is self explanatory. The next procedure will involve actual path and node data input. The'following set of data will be requested for each path. l Path 1 ofsX --- Entering node # 1 Ending node # 2 Path description (up to 60 characters - def PATH f) (If you desire a zero input, type 0, not 0.) Forward form loss in path ( Def 0 ) ? Reverse form loss in path ( Def 0 ) ? Area in path, Sq Feet ( Def 0 ) ? Temperature in path, Deg F ( Def 70 ) ? or (Heat input in path, BTU /sec (Def 0) ?] Absolute path roughness, p inches ( Def 0 ) ? Hydraulic diameter in path, Feet ( Def 0 ) ? Friction length in path, Feet ( Def 0 ) ? (S) ave Model, Show Flow (D)iagram, (C)ontinue, or (END) ? o " Enter" is the same as (C)ontinue FSPLIT will designate which path of x (where x is the total number of paths in l the model) it is requesting input. It will also show which nodes bound the path. l The first request will be for a description of the path. The second and third l inputs are the form loss factor, one for each direction of flow. These can be  ! any positive, zero, or negative number. The next request, area, must be the area that the velocity head will be calculated from. This means that both the form j 30 1 i

I Document Number 32-1203121-01 loss and friction velocity must be based on this area. If a negative sign precedes the area input, an additional form loss factor as a function of flow can l be added to this path. This is used for Y's and TEE's where form loss factors are flow split dependent. This split dependent loss factor will be used by 1 l FSPLIT for forward or reverse flow (in addition to the constant forward or j reverse k-factor)'. After a negative area is input'8, the following data will I be requested.

                                                                                       )

Numerator path # ? l Denominator path # ? WfBRpaWBiff6islyaWoE6sie376iii f'6f'VsT5Effy^Rit'io !TFEEt'6F51iil?s (Vr, K) ? Disk File for Vr, K pairs ? (If " enter", then) , Vr = K= l Per Reference 1, the flow dependent losses are typically dependent on the velocity ratio of one of the Y branches (numerator path) to another branch j (denomina,torpath). Inputting these paths and a set of velocity ratios (Vr) vs loss factor (K), FSPLIT will iterate to a flow split that satisfies the problem ) constraints. Q@$$$$e[sMN$fjlfdisj3@l@@yj}Efe}fM@lijgssed 57a7(Id1Riflhja[dMlaM$ElHinYtW!5@i[@@NtIQMM]M]fa@M NDi1W3IEiiEIEPHigitBlWiRHffdIWR[iTjiG1]f6(uisjIthi

    @@]@ygM[{$ 1&[g@@Ed@S@$$d@l@jg@{@lsidMtRth5 1EiE!EM3HsiffiUM3MENAMIEN#EEMEEWrca11FDutN6t jji@jjdilsgyl@@WiMi5[jWd[y[03))

FSPOT will request the number of input pairs N and then N sets of Vr,K data. A maximum of 15 data sets are allowed for each variable loss path. The Vr - K data pairs can be input from a separate ASCII file to be specified by the user. [This file would have N pairs of V ratios and K-factors.). In this case, the file

          ' Use flow dependent loss ratios only when the direction of flow is known.

These types of. flow factors are typically for one direction of flow only.

          'O ' Be sure that the absolute value of the area is the correct area for the paths form loss coefficient and friction velocity head.

31

Document Number 32-1203121-01 name would be input at the " Disk File for ..." prompt {@MQjQ]S6]didlifs {@H[5H[{@l[g[1((dy. If no file is used, press " enter" at this prompt and the flow split dependent data will have be entered directly into the FSPLIT file. The next request is for path temperature (or heat input). If temperature is input, the resulting density will be used for the path velocity head calculation. If a positive or negativc heat rate is input, the path density will be based on the average temperature of the path (half the path A enthalpy). Finally, the hydraulic diameter, material roughness, and friction length are input if friction losses are to be calculated. The hydraulic diameter is automatically calculated from the area input above as if the component was a circular pipe. If it is different, the user should input the actual valuo. A set of these hydraulic inputs will be requested for each path in the model. Each path will assume the default value of the parameters of the previous path, therefore, if many inputs are identical for subsequent path, the user need only press " enter" for the input. [ Keep this fact in mind when developing the nodes - and paths of the model, this could save considerable time when there are many paths with the same areas, temperatures etc.] After each path data is input, "D"iagram will display the network on the screen, "S" ave will save the model, and "C"ontinue or " enter" will begin the next path input sequence. After all the path data is input, the code will request node input data. Each node w"ill require an area input and an elevation input (if recoverable pressure drops are to be calculated). If the " heat input" option was not previously selected, the node temperatures will also be requested. An. example screen is shown below. NODE 1 of X Elevation of Node, Feet ( Def 0 ) ? 32  ! 1

fs Document Number 32-1203121-01 ) Flow Area at Node, Sq Feet ( Def 0 ) ? Temperature at Node, Deg F ( Def 70 ) ? (S) ave Hodel, Show Flow (D)iagram, (C)ontinue, or (END) ? l After all the node data is input, the external path data will be requested. If more than one external path is to be defined (steady state flow input or pump input problem only), the code will request the following data for each external path. How many External Psths ( Def 1 ) ? External Path 1 From Hode ( Def 0 ) ? To Node ( Def 0 ) ? If problem type 1, 3, 5, or 7 then; Flow in this Phth (Def 1000 Lbm/Sec) ? If problem type 2, 4, 6, or 8 then Pressure drop in this path (Def 0 psi)? f System pressure, psia ( Def 1000 ) 7 System pressure Node number ( Def 1 ) ? If there is more than one external (flow) path, these inputs will be requested i for each external path. (If pump head-capacity inputs are to be used, estimate the initial flow rate. The better the guess, the less time the problem will require to converge.) After all external path information is input, FSPLIT will ask if the file is to be saved. < Do you want to save this file (Y/N) ? If "Y"es is chosen, FSPLIT will request the file name. [It is strongly recommended that the problem be saved before attempting to run it.] If the user ues not want to save the model, type "N" or press " enter" ,

                               ,#4 Run the Problem After all the hydraulic data is input, the problem is ready to run.       From the path creation screen, type "run". The following set of inputs will be requested.

33

l

                                                                                      )

i Document Number 32-1203121-01 Restart Read File (ENTER'for none) ? , i Restart Write File (ENTER for none) ? Max Iterations (Def 50) ? [if pumps are used] Max Head-Capacity Iterations (Def 20) ? [if heat input is used] Max Energy Addition Iteration (Def 20) ? [if compressible gas is used] Max Density-Pressure Iterations (Def 20) , Convergence Criterion (Def .00001) ?  ! [if pumps are used] , Head capacity convergence criterion (Def.0001) ?  ! Pump 1 Correction Scheme (1 or 2 - Def 1) " l Multiplier on Head (def 1) ? Pump 2 CorrectionScheme(1or2-Defj) Multiplier on Head (def 1) ? Pump N Correction Scheme (1 or 2 - Def i) ~^ l Multiplier on Head (def 1) ? [if heat input is used] Energy Addition Convergence Criterion (Def .0001) ? [if compressible gas is used] Density-Pressure Convergence (Def .0001) This Case contains Recoverable Pressure Drop Informatio1. Should only the Unrecoverable Drops be calculated (Y/N) ? A " Restart" file can be writh > save all the path and node results for a later run. H6Ej,i]EiffER6][iif@HFj@lfjltloifleJMjjirl@jjjjiftKthe]r$

 ]f[j@l]Mjgj The advantage of using restart files is to reduce run time substantially because each path / node initial value is the calculated value from the restart file. This is a helpful option especially for problems with multiple pumps and'many nodes / paths.

The maximum number of local iteration input (50) is the maximum number of times the network will be solved. After each solution, the resulting pressure drops and flow rates are compared to the previous iteration. When they are all within the " convergence criteria" of.each other, the problem is solved. [Notethatall paths with a flow rate or node pressure'that have a .01% or greater change in the between the next to last and last iteration will be noted on the first page of the output. If this message appears, the problem may need to be re-run with a smaller convergence criteria. If a path in the network has a very small flow 34

Document Number 32-1203121-01 rate (.1% of the maximum flow), this path could have a relatively large percent change between iteration but a very small absolute change. In this case, a smaller convergence criteria may not help. If pumps are in the model, additional calculations are performed to compare the final calculated flow rate and system pressure drop to the head capacity (HC) values. If they do not match, the system flow rate is adjusted and another set of local iterations are performed. This process continues until both the local and the pump HC convergence is satisfied. Fifty local iterations is typically more than enough for most problem. In cases where Y's and Tee's are used, more than 50 may be needed to adequately converge the problem. Also, 20 HC iteration are usually adequate for most pumps but some very steep or very shallow HC curves may require more. The fractional difference (between iterations) in system flow and system pressure drop is shown on the screen for each iteration. If the problem does not converge in the number of specified iterations, FSPLIT will retain the ,results of the last iteration. The next input is a choice of either of two pump correction scheme. The slope of some HC curves are such that they will not converge by the default convergence method [@K41M@yLJhl@c]{KNjjj[$6f@lG6Mj@M@jylfil this scheme does not work (ie., the problem does not converge), try the other scheme (choice #2). RESC@]MMij{alRthi!@jj]jpgri?igjljglfi[!~pfoblim[tRb's L81sI61asmqMgigirgangshuisthagaiiiE(65vygggjisth3#D Next. the pump head multiplier is required. This allows for the user to modify the HC curve by effectively raising or lowering the curve by a multiplier. The default value for this input is 1.0. If heat rates are input, the local iterations will calculate a path flow rate which will generate a new outlet node temperature (and densities). The model will iterate until two successive " heat input" iteration calculate the same path flows and nodal pressures (within the convergence selected). 35  !

p . Document Number 32-1203121-01 The final user input in this menu is for eliminating recoverable pressure drop calculations. This option is primarily used to speed up analyses that have previously shown that the recoverable pressure drops are negligible. NOTE THAT IF THIS PROMPT IS ANSWERED WITH "N", AND THE FILE IS RESAVED AFTER RUNNING, ALL THE RECOVERABLE HYDRAULIC INPUTS WILL BE LOST. THE FILE MUST BE RE-RUN WITH "Y" TO THIS PROMPT BEFORE RESAVING IN ORDER TO RETAIN THE RECOVERABLE PRESSURE DROP DATA (SEE DISCUSSION ON THE RELOAD OPTION BELOW). A typical output of the convergence of a problem with pumps and heat input will look like the following. FSPLIT 5A- FLOW SPLITS by DA FARNSWORTH & JA WEIPER - Light Water Option Local Iters Sec/ Iter Convergence Parameters 1 0.21875 1.C100000 0.9999996

        > H/C Iteration 1 took 50 Local Iters, Max H/C Deviation 0.79763
       > H/C Iteration 2 took 50 Local Iters, Max H/C Deviation 0.00000
       > H/C Iteration 3 took 6 Local Iters, Max H/C Deviation 0.00000
   === Heat Addition Iteration 1 - Max Deviations : l'00000 and 1.00000===
 ^
       > H/C Iteration 4 took 20 Local Iters, Max H/C Deviation 1 00876
       > H/C Iteration 5 took 25 Local Iters, Max H/C Deviation 0.00000
   === Heat Addition Iteration 2 - Max Deviations : 1.00000 and 0.00441===
       > H/C Iteration 6 took 13 Local Iters, Max H                   0.00005
   === Heat Addition Iteration 3 - Max Deviations        0.00574 :/C Deviation and 0.00000===
       > H/C Iteration 7 took 1 Local Iters, Max H/C Deviation 0.00005
   === Heat Addition Iteration 4 - Max Deviations : 0.00000 cnd 0.00000 ===

Press < ENTER > to continue ... The first line shows the convergence comparison for' each local iteration (along with the computer time required to do the iteration). When the convergence comparison are within the specified limit,s, the maximum number of iteration and the head-capacity (HC) or energy addition convergence will be printed on the screen. This process will continue until both.the local and HC and/or energy convergence are met. Note that the local iterations converge to a flow or pressure ratio (1.00000) and all the other convergence calculations converge to an absolute difference (.00001). Jgj@ Te pd 6fisi[GIIofc E rj @ E I K fis [tWMterruptithe BJEdfat]Efg(iMEE*2Riigy!1][stM3fiWDEGlit15iFIRtEtfie' nextsi t era tiori

   @jg@ltgygvleERRR{sE@@g[AHjfeg@{[{ff[thMjiults((see next!

36

i l ! l l i Document Number 32-120312I-01 l l 5ESENS133!$$3N!$5EEENNbkN5555EiNI556bkk$$N!k[$$!hlNNNSNN E @ NJHi i

                                           #5 Review Results After pressing enter (from the above menu), the following screen will appear.

1 - Print Results to (D)iskfile 2 - Print Results to (S)creen i 3 - Print Results to (L)ineprinter 4 - Show results on (DIA) gram I 5 - (SH) ell to system (' EXIT' to return) 6 - Write (RES) start file 7 - (R)eload current model 8 - (NEW) problem (restart FSPLIT) 9 - (END) to terminate FSPLIT CH0 ICE (Number or letters) ? Since the model was analyzed, the user can either print the results to the screen (option 2) or to a printer (option 3) or to a disk file (option I). It is recommended that the results be reviewed on the screen, especially'if this is the first solution. The output can be reviewed to determine if this case is adequate. During the screen review the above menu can be reaccessed by typing " menu" and the l

     -  results can then be saved, printed or the model re-run. BE SURE TO CHECK THE FIRST  l l

PAGE OF THE OUTPUT TO DETERMINE IF ANY LIMITS HAVE BEEN VIOLATED THAT MAY HAVE THE RESULTS INVALID. THESE LIMITS INCLUDE SATURATION CONDITION $ EXIST, VELOCITIES j GREATER THAN THE SPEED OF SOUND, $_EgQ[$TEjgEdP]$TjREf[GiREAIEQHAN;500*FQGAi FSPLIT RReefgujyllBn0@BMlLgGM{g OR PRESSURES LESS THAN ZERO PSIA. will calculate and show these results even if some of them are physically impossible . to attain. This allows the ' user to ' determine where the model is deficient and modify the case. Also, a note specifying if any path flow rate or  ! 37 i

r . Document Number 32-1203121-01 node pressure drop had a greater change than .01% between the next to last and la iteration (see discussion above). Samples of the output are shown in Attachment 2. The screen output and printed results have the same format. I If the results are acceptable, they can be printed or saved (options 1 and [It 3). I is recommended that the results be saved to a file before attempting to print because a printer error may cause FSPLIT to return to DOS. The results would then be lost and the problem would have to be re-run.] In addition to the screen du option 4 allows for the model nodes / paths to be displayed on the screen with selected results shown at the bottom of the screen for any path. This is a convenient method of a quick review for large models. Options 5 and 9 are self explanatory while option 8 will restart FSPLIT from the first screen (and all previous results will be lost). If a restart file was not opted to be written in the previous run input menu, it can be written at this point if option 6 is selected. Option 7 is discussed below.

                                      #6 Modify Network If the Reloadoption (7) is selected, the model appears in the node creation mode.

Typing F9, gets to the path creation mode. In either of the path or node creation mode:;, changes can be made to the physical model as previously discussed. If only input data needs to be modified, type "save" or " create" from the path creation mode and the following menu (discussed in #3 above) will appear. l 1 - SPECIFY ALL NEW PATH INF0 (AREAS, LOSSES, ETC.) 2 - CHANGE ONLY SELECTED PATHS 3 - SAVE FILE AS IS AND RETURN TO MODEL * (ONLYIFTYPEPREVIOUSLYSPECIFIED) ENTER I, 2 or 3 7 1 i Type "2" and follow the instructions to change any of the input data. Each change option will have a default of the original value (or descriptien) and the user will 38

c Document Number 32-1203121-01 l have the option of changing the value or keeping it the same by pressing " enter". ) f i[QEJg@Sg)EXINjT}fS3"gfNR$Qg(gfg~ fsgITEWI[@TERfR$TfA^

  $pTEjjJT[l 59sIEMBEfeIMTMILENTpj3%I@EiVggMQLNg[C]HMGQ The first change will be the case title. After this, the following message will appear.

Input the path number to change, or '999' to end path changes, or a negative value to view the flow diagram ? Typing a path number will show the default inputs of all the path and adjoining node data and the user will have an opportunity to change any of the data. Note that a change in node data will also impact any other path connected to that node. Typing any negative number from this menu will show the network on the screen. { After all the path / node changes are made, type "999" and the external path input l data will be presented for change. Again, pressing " enter" for each request will  ; maintain the original values. After all changes are made the user will be prompted to save the model to a disk file and the path creation mode will appear on the  ; screen. At this point, the user can re-run the model and repeat this process as 1 often as needed. 39

r= 1 l Document Number 32-1203121-01

8. REFERENCES )

! i

1. Handbook of Hydraulic Resistances Second Edition by I E Idelchik Hemisphere Publishing Co. 1986
2. ASME Steam Tables Fifth Edition 1983
3. LINPACK Users Guide by J.J. Dongarra, et. al. Philadelphia,1979
4. AECL-1763 Tables of Thermodynamic Properties of Heavy Water by'J. N i Elliott Chalk River Ontario,1963.
5. BAW-230 " Core Hydraulics and Thermal Analysis (CHATA)" 1975
6. Crane - Flow of Fluids. Fittinos and Pioes Technical paper #410 Crane Co. 1988
7. AECL-7531 Table of Thermodynamic orocerties of heavy water in SI units Atomic Energy of Canada 1981 I

l I 40 .

f I i ATTACHMENT 1 FSPLIT BENCHMARK CASES This attachment discusses all the details of the check cases evaluated by FSPl.IT. The accuracy of the results is highly dependent on the fluid properties used. Since FSPLIT calculates the fluid properties for light and heavy water, the FSPLIT results are compared to'. water properties' from referencable sources. Finally, the results of -all the other calculation performed by FSPLIT are compared to-known values. Water Properties and Friction Factor 1 1 The flow and pressure drop calculations can only be as accurate as the water l properties used. The following Tables A2 and A3 shows FSPLIT =alculated properties compared to References 2, 4, 7 and Attachment 4 (for light and heavy water). The reference properties for heavy water viscosity and specific heat are compared for saturated liquids only while @sff}R[GiitRigfdaMiddictivity

                                      ~

Ee]6MisidinifsUtiE@R@ltMjj. s The heavy water saturation properties are I calculated at saturation conditions and then ratioed by the same variation that l the light water exhibits from its saturated condition. T]hMidijjsh]((s@powrj [E@M))(($lffE@M(jE@$$$f81(E(@HFy}@((jjgNflhypD4ef l EtENE!MRinME!!!~WFI5NNeay1MNBlininI!D11tHE:m6ri t 5N}L}M E 8 E E A E @ 3 MI M i @ @ $1{j @ g K @ f % jg y m6(ifiy ' 51EE1M~MI3dEEld@MPNENN1] Reference 4 (compared to-Attachment 4 per Table A3) shows that there is some disagreement in the heavy water properties. ,$)5jMiifM~jj[@Tjtil@1W@Fj6Elfir7 agreement)

     ^

L 'tjig{sjrM g j g g @ M S W J g fie~pflT@ (except when the on1y data l 1 available was from Attachment 4). Density _ is the most' important property used in the pressure drop calculations since it is used directly in Bernoullie's equation. As shown in Tables A2, it deviates typically less than .01%'(from Reference'2) for light water with the maximum Saviation of .03%. $JEd@j($}}[sR$$$Edja]ijloQheEfact that ME N S3A @.ggggf@{dE@jjgffal@Mfdglo25]Midisi)fplnes '(4 4

1 l l l 1 i l r l Document Number 32-1203121-01 555((N N E03N N E[5d E5 EilS N 3 E E E N13MlM[]ji@ {Fj QTjMIMf

.. .a n n .n lyln.n-nn,',f.,li,sjgnJ a tp3sjusiugi ot f can jn~.,tdi_gures,d7 h,ei [ff,egnce;jj@,pjwggvaluestcculd d

[fff((@fal[$p5_G]E[ej@{HfF_Sjl]JJsij{@]lj{c"uHijfthin'itERjfggEE!2

   ,1]@j@[f631 Dip @j$@EEd))MR@)Ms3p353M[Ql[ijllyH@] For heavy I   water, the typical deviation is less than =.1% and the maximum deviation is
   =0.2%8 (see Table A3-3 and A2-4). Summarizing, the potential error due to water density in all the pressure drop components calculation is effectively i'if6 jfrEffg{@jDj@Mty
          !!               and zero to a maximum of 0.2% for heavy water. The nominal real error for heavy water will be on the order of .02% to .03%.                l Friction losses are dependent on density and the friction factor. The friction factor is dependent on Reynolds number which in turn is dependent on density and viscosity. In the highly turbulent region (high Reynolds number) the friction factor is relatively constant and therefore an error in the Reynolds number will not affect the friction pressure drop.        In the lower turbulent and transition region, the relationship for large diameter pipes is approximately 1% error in friction factor for 1% error in density or viscosity. The smaller the pipe, the smaller the error.        In the laminar flow region, a 1% error in density or viscosity will cause =1.5% error in friction pressure drop. From Table A1, the typical viscosity error in light water is less than 0.1%. The heavy water error is less than 1.0% in the pressure / temperature range where most heavy water            !

problem are analyzed (see Table A2). The maximum viscosity errors for light and heavy water are less than .2% and 2.5% respectively. Therefore, the total error in friction pressure drop (at low Reynolds numbers) due to water properties could l be in the =0.1% to 0.2% range for light water and =1.0% to 2.5% range for heavy water during laminar and transition flow. The error will be effectively zero for 1 A large portion of the .1% error is due to the use of only 4 significant figures in Reference 7, similar to the light water error. Therefore, the real error is nuch less than values shown on tables A3-3 and A3-4. 2 Heavy water properties were based on fits from =50*F to l =500*F. Temperatures outside this range may result in larger l errors. FSPLIT will flag results with temperatures greater than 500'F. 42 e>. m

Document Number 32-1203121-01 the.high Reynolds number cases. Friction losses are also dependent on the " friction factor". Friction factor has been shown to be a function of material roughness and Reynolds number (see l Reference 6). Typically, the friction factor is described in graphical form as shown in Reference 6. Reference 5 describes a mathematical fit of this data which is used in FSPLIT. This equation results in the following friction factors as a function of Reynolds Number and relative roughness (c/D). A comparison to the friction factor in Reference 1 is also shown. TABLE Al FRICTION FACTOR CALCULATED IN FSPLIT REYNOLDS NUMBER 1.0E+03 1.0E+04 1.0E+05 1.0E+06 1.0E+07

 -          RELATIVE    ----------------------------------------------------

ROUGHNESS FSPLIT 0.05000 0.0640 0.0737 0.0717 0.0714 0.0714 REF (1) 0.05000 (0.073) (0.072) (0.072) (0.072) FSPLIT 0.01000 0.0640 0.0430 0.0385 0.0379 0.0379 REF (1).0.01000 (0.043) (0.038) (0.038) (0.038) FSPLIT 0.00500 0.0640 0.0375 0.0313 0.0304 0.0303 REF (1) 0.00500 (0.0375) (0.0315) (0.0300) (0.0300) FSPLIT 0.00100 0.0640 0.0322 0.0222 0.0199 0.0197 I REF (1) 0.00100 (0.032) (0.023) (0.020) (0.020) FSPLIT 0.00050 0.0640 0.0314 .0.0203 0.0172 0.0167 REF (1) 0.00050 (0.0320) (0.0205) (0.0165) (0.0165) FSPLIT 0.00010 0.0640 0.0308 0.0186 0.0135 0.0122 REF (1) 0.00010 (0.032) * (0.019) (0.013) (0.012) l FSPLIT 0.00005 0.0640 0.0307 0.0183 0.0127 0.0109 REF (1) 0.00005 . (0.032) (0.019) (0.013) (0.011) j FSPLIT 0.00001 0.0640 0.0307 0.0181 0.0119 0.0090 REF (1) 0.00001 (0.032) (0.019) (0.012) (0.009) 43 l l

[- Document Number 32-1203121-01 These results were compared to page 85 of Reference 1 and agree very well as

              -shown above. Since the Reference 1 data were also a result of interpretation of the Moody friction curves (experimental data), they are not necessarily better results than the Reference 5 equation. Therefore, no e/D error will be assumed to be introduced to the FSPLIT friction factor calculation based on Reference 5.3 The enthalpy property is used only when heat loss / gain (instead of temperature) is input to the problem. An error in this property will create a calculated temperature error and consequently a potential error in actual density. N$[vgl

[@X@@%efeitRj{3Fd(({jf(ajjjp@M[Qt [j[@[j{Qjfe?@$[{j W i 15EIE!NsIEEeI!MtiBEEinEfFlGMifit900XsHH The heavy water enthalpy error will affect the resulting density calculation by much less than 0.1% and therefore, it will not have a significant impact the , pressure drop calculations. Finally, the specific heat and conductivity are used to calculate a hea't transfer coefficient which is not used in any other calculations in this program and the errors are shown for information only. Specific heat comparison to Attachment 1 4 is for saturated conditions only. However, Reference 4 and 7 show a pressure l dependency for conductivity. l

            . The .propertie's for " fluids" and " gasses" are input by the user as previously     l discussed. Therefore, any errors in these properties are the responsibility of       I the user.                                                                            i l

3 The Reynolds number region where laminar flow becomes ,

              -transition flow is typically shown as a step function in friction                    l factor.      FSPLIT models this as a smooth transition within a very small Reynolds number change.

44

1 Document Number 32-1203121-01 TABLE A2-1 FSPLIT LIGHT WATER PROPERTY COMPARISON TO REFERENCE 2 ........................................ENTNALPT - 8TU/LS*===-- ==*--* -==- .- - ====== PRESS TEMP FSPLIT STEAM PERCENT ABS PRESS TEMP FSPLIT 5744 PER0ENT ABS PS!A F TABLES OlFF DIFF PS!A F TANLES OlFF O!FF 15 50 18.10 18.10 0.00% 0.00 1500 500 487.60 487.63 0.01% 0.03 15 100 68.04 68.04 0.00% 0.00 1500 550 548.41 548.47 0.01% 0.06 15 150 117.98 117.98 0.00% 0.00 1500 590 602.43 602.57 0.02% 0.14 15 200 168.09 168.09 0.00% 0.00 15 210 178.15 178.15 0.00% 0.00 2000 50 23.80 23.80 0.00% 0.00 2000 100 73.26 73.26 0.00% 0.00 100 50 18.34 18.34 0.00% 0.00 2000 150 122.83 122.83 0.00% 0.00 100 100 68.26 68.26 0.00% 0.00 2000- 200 172.60 172.60 0.00% 0.00 100 150 118.19 118.19 0.00% 0.00 2000 250 222.70 222.70 0.00% 0.00 100 200 168.29 168.29 0.00% 0.00 2000 300 273.32 273.32 0.00% 0.00 100 250 218.74 218.74 0.00% 0.00 2000 350 324.71 324.71 0.00% 0.00 100 300 269.77 269.77 0.00% 0.00 2000 400 377.18 377.19 -0.00% 0.01 100 320 290.42 290.42 0.00% 0.00 2000 450 431.20 431.21 -0.00% 0.01 2000 500 487.50 487.53 0.01% 0.03 500 50 19.50 19.50 0.00% 0.00 2000 550 547.49 547.56 0.01% 0.07 500 100 69.32 69.32 0.00% 0.00. 2000 600 614.30 614.48 0.03% 0.18 500 150 119.16 119.16 0.00% 0.00 500 200 169.19 169.19 0.00% 0.00 2500 50 25.22 25.22 0.00% 0.00 500 250 219.57 219.57 0.00% 0.00 2500 100 74.57 74.5 7 0.00% 0.00 500 300 270.51 270.51 0.00% 0.00 2500 150 124.05 124.05 0.00% 0.00 500 350 322.32 322.32 0.00% 0.00 2500 200 173.74 173.74 0.00% 0.00 500 400 375.38 375.38 0.00% 0.00 2500 250 223.75 223.75 0.00% 0.00 500 450 430.27 430.27 0.00% 0.00 2500 300 274.27 274.27 0.00% 0.00 2500 35J 325.53 325.53 0.00% 0.00 1000 50 20.94 20.94 0.00% 0.00 2500 400 377.82 377.82 0.00% 0.00 1000 100 70.63 70.63 0.00% 0.00. 2500 450 431.58 431.58 0.00% 0.00 1000 150 120.39 120.39 0.00% 0.00 2500 500 487.48 487.50 0.00% 0.02 1000 200 170.33 170.33 0.00% 0.00 2500 550 546.73 546.79 0.01% ' O.06 1000 250 220.61 220.41 0.00% 0.00 2500 600 611.93 612.08 0.02% 0.15 1000 300 271.44 271.44 0.00% 0.00 2500 650 691.10 691.49 0.06% 0.39 1000 350 323.11 323.11 0.00% 0.00 1000 400 375. % 375.96 0.00% 0.00 3000 50 26.64 26.64 0.00% 0.00 1000 450 430.54 430.55 0.00% 0.01 3000 100 75.88 75.88 0.00% 0.00 1000 500 487.77 487.79 0.00% 0.02 3000 150 125.28 125.28 0.00% 0.00 1000 540 536.66 536.69 0.01% 0.03 3000 200 174.88 174.88 0.00% 0.00 3000 250 224.81 224.81 0.00% 0.00 1500 50 22.37 22.37 0.00% 0.00 3000 300 275.22 275.22 0.00% 0.00 1500 100 71.95 71.95 0.00% 0.00 3000 350 326.35 326.35 0.00% 0.00 1500 150 121.61 121.61 0.00% 0.00 3000 400 378.47 378.47 0.00% 0.00 1500 200 171.47 171.47 0.00% 0.00 3000 450 431.99 431.99 0.00%

  • 0.00 1500 250 221.65 221.65 0.00% 0.00 3000 500 487.51 487.52 0.00% 0.01 1500 300 272.38 272.38 0.00% 0.00 3000 550 546.13 546.16 -0.01% 0.03 1500 350 323.90 323.91 0.00% 0.01 3001 600 610.01 610.08 0.01% 0.07 1500 400 376.56 376.56 0.00% 0.00 3000 650 684.94 685.10 0.02% 0.16 1500 450 430.85 430.86 -0.00% 0.01 3000 690 775.95 775.15 0.10% 0.80 45
 )

Document Number 32-1203121-01 TABLE A2 , PRESS TEMP STEAM FSPL1T........................................ PERCENT ABS DENSITY LB/CUFT

                                                                                              --  -      - ~~- -- - ---

PSIA F TA8LES

           ............................O!FF        O!FF      PRESS-PSIA     FTEMP FSPLIT STEAM PERCENT ABS
                                           ............                             TABL       01FF     O!fF 15 ~ 50 62.413 62.422 0.01% 0.009              .........................ES ..................

15 ~ 100 . 61.997, 61.996 0.005 -0.001 1500 450 51.908 51.921 -0.03% 0.013 15 150 61.190 61.200 0.02% 0.010 1500 500 49.384 49.383 0.00% 0.001 15 200 60,107 60.096 0.02% 0.011 1500'590 550 4.321 44.318 0.01% 0.003 1500

15. 210 59.862 59.880 0.03% 0.018 43.193 43.197 0.01% 0.004 100 50 62.431 62.422 0.01% 0.009 2000 50 62.813 62.814 0.005 0.001 10" .100 62.013 61.996 0.035 0.017 2000 100 62.367 62.383 0.03% 0.016
          .00 150'    61.20760,132 61.200 0.01% -0.007            2000 150 61.564 61.576. -0.02% 0.012 100     200 60.124             0.01%               2000 200 60.503 60.496 0.01% 0.007 0.008~
        ~100 250 $8.817 58.824           0.01% 0.007         2000 250 59.230 59.242 0.02% 0.012 100 300 57.303 57.307 0.01% 0.004                   2000 300 57.766 5 7.770 0.01% 0.004 100 320 56.639 56.625           0.02% 0.014         2000 350 56.107 56.117 0.02% 0.010 2000 400 54.234 54.230 0.01% 0.004 500 50 62.512 62.500 -0.02% 0.012                  2000 450 52.105-52.110 0.01% 0.005
      '500'100 62.089 62.073 - 0.035 0.016                  2000 500 49.641 49.652 -0.02%              0.011 500 150 61.283 61.275 0.01% 0.008                  2000 550 4.688 46.685 0.01% 0.003 500 200 60.205 60.205 .0.001 0.000                   2000 600 42.880 42.882 0.00% 0.002                          !

500 250 58.906 54.893 0.02% 0.013 500 300 57.403 57.405 0.005 0.002 2500 50 62.912 62.933 --0.035. 0.021 500 350 55.691 55.679 0.02% 0.012 2500 100 62.458 62. 4 1 -0.00% 0.003 500 400 $3.740 53.735 0.01% -0.005 2500 150 61.656 61.652 0.01% 0.004 2 2500 200 500 450 51.493 51.493 -0.00% 0.000 60.600. 60.606 0.01% 0.006 2500 250 59.336 59.347 -0.02% 0.011 1000- 50. 62.613 62.617 0.01% - 0.004' 2500 300 57.883 57.870 0.02% 0.013 1000 100 62.182 62.189 0.01%. 0.007 2500 350 56.241 56.243 -0.00% 0.002 1000 150 61.378 61.387 -0.02% 0.009 1000 .*>00 2500 400 54.392 54.407 0.03% 0.015 60.305 60.314 -0.01% 0.009 2500 450 52.297 52.301 ~-0.01% 0.004 1000 250 59.015 59.032 0.03% 0.017 2500 500 49.888 49.900 0.02% 0.012 1000 300 57.526 57.537 0.02% 0.011 2500 550 47.032 47.037 0.01% 0.005 1000 350 55.832 55.835 0.00% 0.003 2500 600 43.438 43.440 0.01% 0.002 1000 400 53.909 53.908 0.00% 0.001 2500 650 38.180 38.183 0.01% 0.003 1000 450 51.704 51.706 0.005 0.002 1000 500 49.116 49.116 0.005 0.000 3000 50 63.010 63.012 0.005 0.002 1000-540 46.632 4 .642 0.02% 0.010 3000 100 62.549 62.539 0.02% 0,010 3000 150~ 61.747 61.767 0.03% 0.020 1500- 50 62.713 62.696 0.03% 0.017 _ 3000 200 60.696 60.680 0.03% 0.016 1500 100 62.275 62.267 0.01% 0.008 3000 250 59.440 59.453 0.02% 0.013 1500 '150 61.471 - 61. 4 3 0.01% -0.008 3000. 300 57.999 58.005 0.01% 0.006 1500 200- 60.404 60.423 -0.03% 0.019 3000 350 56.373 56.270 0.015 0.003 1500 250 59.124 59.137 -0.02% 0.013 3000 400 54.545 54.555 0.02% 0.010 1500 300 57.647 57.637 0.02% -0.010- -3000 450 52.483 52.493 0.02% 0.010 15CO 350 55.971 55.960 0.022 -0.011 3000-500 50.125 50.125 -0.005 0.000 1500" 400 54.074 54.083 0.02% 0.009 3000 550- 47.357 47.348 0.02% 0.009 3000 600 43.940 43.937 0.01% 0.003 3000 650 39.241 39.24 0.01% 0.005 I t 46

y J Document Number 32-1203121-01 TABLE A2-3 1

      . --
  • CONDUCTIVITY - BTU / HR FT F -----

0YNAMIC VISC08 TTY LBf SEC/SQFT" - PRESS TEMP FSPLIT STEAN PERCENT ABS PRESS TEMP FSPLIT STEAM PERCENT- ABS

       ' PSIA F               TABLES DIFF                  PSIA F                     TABLES DIFF          O!FF
         ...................................'DIFF-15 50 0.3360 0.3363 0.08% 0.0003                  100     50 2.73E 05 2.73E 05 0.03% 8.0E 09 15 100 0.3614 0.3617 0.06% 0.0002                 100 100 1.42E-05 1.42E-05 0.01% 2.0E 09 15 150- 0.3788 0.3790 0.04% 0.0002                100 150 8.99E-06 8.99E 06 0.02% 1.7E-09 15 200 0.3896 0.3896 -0.001- 0.0000               100 200 6.34E-06 6.34E 06 0.05% 3.1E 09 100 250 4.80E-06 4.80E 06. -0.00% 1.0E 10 100- 50 0.3362 0.3365 0.08% 0.0003                 100 300 3.83E 06 3.83E 06 0.00% 1.CE 10 100 100 .0.3616 0.3618 0.05% 0.0002 100 150 -0.3790 0.3791 -0.03%        0.0001       500      50 2.72E-05 2.72E 05 -0.01% 4.0E 09 100 200 0.3898 0.3899 -0.0!X 0.0001               500 100 1.42E 05 1.42E 05 0.00% 1.7E 21 100 250 0.3950 0.3951 0.04% 0.0001                500 150 9.00E 06 9.00E-06 0.02% -1.4E 09 100 300 0.3950 0.3951 0.02% 0.0001                500 200 6.35E-06 6.35E 06 0.06% 2.4E 09 500 250 4.82E*06 4.82E 06 0.09% 4.5E-09 500 ' 50 0.3371 0.3374 0.081 0.0003               500 300 3.85E 06 3.85E 06 0.12% 4.7E 09 500 100 0.3624 0.3626 0.04% 0.0002               '500     350 3.20E-06 3.20E 06 0.15% 4.9E 09                1 500 150 0.3798 0.3799 0.02% 0.0001                 500 400 2.74E 06 . 2.74E-06 0.16% 4.4E 09                  J
        $00 200 0.3907 0.3908 0.03% 0.0001                 500 450 2.39E 06 2.39E 06 0.04% -1.0E 09 500 250 0.3959 0.3%0 0.02% 0.0001 500 300 0.3961 0.3 %2 0.03% 0.0001                1000      50 2.72E 05 2.72E 05 0.01% 2.0E 09 500 350 0.3915 0.3917 0.05% 0.0002                1000 100 1.42E 05 1.42E 05 0.04% 6.0E 09 500 400 0.3822 0.3825 0.07% 'O.0003               1000 150 9.02E 06 9.02E 06 0.02% 2.1E 09 500 450 0.3661 0.3685 -0.11% 0.0004'              1000 200 6.37E 06 6.37E 06 0.03% -1.6E 09
                                                       ' 1000 250 4.83E 06 4.84E 06 0.11% 5.1E 09                     .

1000 . 50 0.3382-0.3385 0.08% 0.0003 1000 300 3.86E 06 3.86E 06 0.11% 4.1E 09 l 1000 100 0.3635 0.3636 0.03% 0.0001 1000 350 3.21E 06 3.21E 06 0.10% 3.3E 09 1000 150 0.3809-0.3810 0.04% 0.0001 1000 400 2.75E 06 2.75E 06 0.14% 3.9E-09

     ~1000 200 0.3918 0.3919 0.03% 0.0001                 1000 450 2.41E 06 2.41E 06 0.00% -1.0E 10                   ;

1000 250 0.3971 0.3972 0.02% 0.0001 1000 500. 2.13E 06 2.13E 06 0.15% 3.1E-09 ' 1000 300 0.3975 0.3976 0.04% 0.0001 1000 350 0.3931 0.3932 0.04% 0.0001 2000 50 2.70E 05 2.71E 05 0.03% 9.0E-09 1000 400 0.3841 0.3843 0.06% 0.0002- 2000 100 1.42E 05 1.42E 05 0.05% 7.0E 09 1000- 450 0.3703 0.3707 0.10% 0.0004 2000 150 9.05E-06 9.05E 06 0.01% 1.3E 09

     -1000 500 0.3512 0.3519 -0.19% 0.0007                2000 200 6.41E 06 6.41E 06 0.00% 2.0E 10                    j 2000 250 4.87E 06 4.87E 06 0.07% 3.4E 09                     '

1500 50 0.3394 0.3395 0.04% 0.0002 2000 300 3.90E 06 3.90E-06 0.03% 1.3E 09

     -1500. 100 0.3645 0.3646 0.02% 0.0001                2000 350 3.25E 06 3.25E 06 -0.02% 6'.0E 10 1500 150. 0.3819 0.3820 0.03% 0.0001                2000 400 2.79E 06 2.79E 06 -0.01% 3.0E 10 1500 200 0.3929 0.3930 -0.03% 0.0001              -2000 450 2.45E 06 2.45E 06 -0.11% 2.7E 09 1500 250 0.3983 0.3984 0.02% 0.0001                 2000 500 2.17E 06 2.17E 06 0.19% -4.2E 09
     ,1500- 300 0.3988 0.3909 0.03% 0.0001                2000 550 1.94E 06 1.94E 06 0.20% 3.9E 09 1500 350 0.3946 0.3947 -0.03% 0.0001                2000 600 1.70E 06 1.70E 06 0.111 1.8E 09
     .1500 450 0.3725 0.3726 0.02%            0.0001.                                                               ,

2000 50 0.3405 -0.3407 0.07% 0.0003 2000 100 0.3655 0.3657 0.05% 0.0002 < 2000 -150 0.3829 'O.3830 -0.02% 0.0001 l

     '2000 200 ~0.3 % 0 0.3941 0.03% 0.0001
  • 2000.300 0.4001 0.4002 0.02% 0.0001 2000 350 0.3961 0.3962 0.03% 0.0001 5

2000. 400 0.3876 0.3878 0.05% 0.0002 2000 450'O.3747 0.3750 -0.09% 0.0004 2000 500 0.3568' O.3573 0.14% 0.0005 47

p Document Number 32-1203121-01 TABt.E A2-4

         ................................ SPECIFIC HEAT STU/LS-F - - - -               -

PRESS TEMP PSIA F FSPLIT STEAM PERCENT A8S TA8LES OlFF PRESS TEMP FSPLIT STEAM PERCENT ASS

            ...........................DIFF     .............

PSIA F TA8LES O!FF 15 50 1.0018 1.0020 -0.021 0.0002 .....................................DIFF 15 100 0.9980 0.9980 0.005 0.0000 1500 50 0.9921 0.9920 0.01X +0.0001 15 150 1.0000 1.0000 0.001 0.0000 1500 100 0.9919 0.9920 -0.01X 0.0001 15 200 1.0049 1.0050 0.01% 0.0001 1500 150 0.9947 0.9950 0.03% 0.0003 1500 200 0.9996 1.0000 0.04% 0.0004 100 50 1.0012 1.0010 0.021 0.0002 1500 250 1.0079 1.0085 -0.061 0.0006 100 100 0.9976 0.9980 0.041 0.0004 1500 300 1.0209 1.0220 0.101 0.0011 100 150 0.9997' 1.0000 -0.03% 0.0003 1500 150 1.03 % 1.0415 -0.18X 0.0019 100 200 1.0045 1.0050 -0.051 0.0005 1500 400 1.0661 1.0670 -0.085 0.0009

           ?90 250 1.0135 1.0140 0.051 0.0005                     1500 500 1.1651 1.1690 -0.33% 0.0039 100 300 1.0279 1.0290 0.11% 0.0011                     1500 550 1.2697 1.2790 0.73% 0.0093 600 60 0.9965 0.9960 0.051 0.0005                      2000 50 0.9890 0.9890 0.001 0.0000 600 100 0.9955 0.9960 0.051 0.0005                     2000 100 0.9900 0.9900 0.00% 0.0000 600 160 0.9986 0.9990 -0.041 0.0004                    2000 150 0.9930 0.9930 -0.005 0.0000
        '600 200 1.0027 1.0030 -0.03% 0.0003                     2000 200 0.9979 0.9980 0.01% 0.0001 600 260 1.0138 1.0140 0.02% 0.0002                     2000 250 1.0060 1.0065 -0.05% 0.0005 600 300 1.0253 1.0260 0.07K 0.0007                     2000 300 1.0186 1.0190 0.04E 0.0004 600 360 1.0505 1.0520 0.15X 0.0015                      2000 350 1.0365 1.0375 0.09E 0.0010 600 400 1.0746 1.0760 -0.13X 0.0015                     2000 400 1.0618 1.0630 -0.125 0.0012 2000 450 1.0982 1.1010 0.25% 0.0028 600 480 1.1558 1.1590 0.285 0.0032 2000 500 1.1538 1.1570 -0.27% 0.0032 1000 50 0.9952 0.9955 0.03% 0.0003                      2000 550 1.2469 1.2580 0.881 0.0111 1000 100 0.9939 0.9940 -0.011 0.0001                     2000 600 1.4372 1.4530 -1.08% 0.0158 1000 150 0.9965 0.9965 0.01% 0.0001                      3000 50 0.9831 0.9835 0.04% 0.0004 1000 200 1.0013 1.0020 0.07K 0.0007 .

1000 250 1.0099 1.0110 0.11% 0.0011 3000 100 0.9862 0.9860 0.021 0.0002 1000 300 1.0233 1.0240 -0.06% 0.0007 3000 150 0.9897 0.9900 -0.03% 0.0003 3000 200 0.9946 0.9950 0.04X .0.0004 1000 350 1.0428 1.0440 -0.111 0.0012 1000 400 1.0707 1.0720 -0.121 0.0013 3000 250 1.0023 1.0030 -0.07X 0.0007 1000 450 1.1120 1.1150 -0.271 0.0030 3000 300 1.0141 1.0150 -0.09% 0.0009 3000 350 1.0307 1.0320 -0.13X 0.0013 ) 1000 500 1.1776 1.1810 0.29E 0.0034  ! 1000 540 1.2653 1.2720 0.525 0.0067 3000 400 1.0534 1.0550 0.131 0.0014 3000 450 1.0862 1.0880 0.165 0.0018 , t 3000 500 1.1344 1.1370 0.23% 0.0026 3000 550 1.2106 1.2160 -0.441 0.0054 3000 600 1.3490 1.3580 -0.67E 0.0090 e l 48

V 1 1 Document Number 3 A l203121-01 1 1 FSPLIT HEAVY WATER PROPERTY COMPARIS0N TO REFERENCE 4, 7, AND ATTACHMENT 4 Some.of the heavy water properties are generated for saturation condi-tions only and therefore the comparison is at saturation pressures. Density, enthalpy and conductivity were compared at subcooled conditions. ENMlW95U2fM83I970 6fiftW@l$p_e,g(WW5IastalH B5miasMGMEMEm TABLE A3-1 U;Ti G ~ P PRESSF'4EWENTHALPWBTd/lBd TEMP QSPLIIM's REEs4*"f. REFl7nV2PERCENMO.':! ABS

                                                                                                      'ar '"fG7MWifTF-W-E-9 I(p%:'TPSIM'$;.'i
               ~%&fM=$ C-i- CR%--u&

LT 1.0*d:Yl,68&'29;604 g b 29/47C Eg*::i':'#:'d?$RESULTSdRESU

                                                                            -: '^ L' =9 '2"-% ; '5 - .--9 102:01h;dc19167:$y:229147k
                                                                      .           :(102;'22W:                  & 0.44% W -O.'li 80' 215 M 0.21 fpf$13.9ud21276173;713:'373:38VF174:

M 6.4?S176fl${I37195 4 5 8137.572:E'$2 138:29:17;3! % 255Q-O.34 9: 6W140$ res fiy0i26% P '0:46 209:117'pT209.91D ?O;23%g ) P's0.49 [cy'd$3lf32$@281145R!L280b87CRi:l'c

             #ESh l 28 0 6 245i2flT 28tT4$5;
042Q$ ~fCOO:49 18%WMiJ0!51 ll27.6E rx 143 #@888**7s :cA3FJi9943L7;96AC0a3 pT2241;'6Av /v G 30.53 25 W 0.'61 p$42U1M4649443204@48tt92W,~

t 39667G39243865 430:99kO;@i2155+^;b,0.90 W 0.73 392i40150. N 470',90"QMO.93%; J 0.14 p?!" 9 685M6

       .-.39..9_A.536W!

_ - - '500?in46%WH"473;t3;% 5.1. k. O6s4516.'60~~ -

                                                                                                   + - ~ - ~ ~

5..12;61M < 0; 295, W

                                                                                                                        ~

TABLE A3-2 E remg! W M Fereasucr f #4FTE@sittiistrontSimatecriekipsHWIWierconnecnous tcPR _  :. yeesspannut;wWE unset . greacraignes ~' M.7 snata anansa h?O$i i. '.. Y1si A$. ,MNW stuAst W or# C otre y . TWJU!k%38EW0.11 gy@M % 030 %@ 11 s s4K1s tc4M gg g s 22

                                                                                                           ?47.ss!# 6,,3 ,16:6c.10 g ,,3 MA%;

h20 $ l l 405kW

                                                                ,.D3 n
                                                                                                         . NiQ90&

p#yg p%je.3st"AB hOVAQ C.10 klo rj . 444 --i y y- - wu .JM g ~ v%o;82215X osse;.eo.es 0.38 r.3 i 1 g g ~T 'f% ,.g Q gl.@; 9 fee . . 6 w.m...a a:tssMo 4 .:. g, 7(ms:ndygoasspso an m E ljy N ,,

                                                                                                             ~

13g 9 og %g % Nh6119E M 0 48 y{!# W 2 4RI29.55 Ds4M $m e g' gg;dag y$340 6 "'

                                                   .teS F 2208.S $ 101:35MS,9ft# 9,;93r 9tMast                   i nooJ @ gt                                                     WiM;8PjstRole965$3jds$58E                               Q.

w1MeestemA21Ertaoisa Qjs.esE@mo'1

                                                                                                                           @U.00          !

i 49 l 1 1 l l

(F 1 Document Number 32-1203121-01 [usi sMst342s&7twME2s5:#Fass:tyMe pMW s eeTQ3MA%5.c3

                                     %h a ggg 299 gggg.:                         y;gg.sW 299.9the:2Wu7rcssW24s74t!E                     ygggy
       &. #; f$gmyy i<                                    s&I5419df38Miy30,no#[i;ggg            9stMa :              ptes.at:

g g!N$ 4 9 31sp&gia;tt [t00! ' 1435420 Q185:2610tagteI4ctqAs$pm71 m gD20s g43534 M 75 15MM0;635N1T;446tN;f5$ 4 LHok 30es#. .f pas ' 4284$435410hase;4$e251sIWo;213McMua4Ee&dgM15tWO'26

                                                                          ;3 g}sanoegmsgwp3sawsmagsta:mem "25t        glF)h                 W                  jt3Nf                                                    asp                     mgoas
                                                                                                                                +separmgq!ks1231 e.

09

                                     %,1$gegg;84410444W;Kg4 d

q ggg$p944139,

y. #w"#""*ggggggg ggggg sei ~'I@ 6~ 7495;$] MS;2
                                                                           'n ggggg r' EPR.09 W450
                     . g .= ?"%-

Men @ i:

                                                                             ' r 9                         .;.
                                                                                                                                               @oeas wagg
               .gy # t ( gf.l<4 9              !

g J r:

            , i;.y**y WW                                                                           '

nm. m , l l

                                                                               ,                                                                  g ww m
                                                                                                                              "  "~
                                          ,                                                                                                     ^

l - - 6 M sm% ex fa90E a,h258;. p258.5D f g M f140jt K J .. E85W ' ^' wSR ?n#*t:16 a {pangano0e g m %g< pasosasWitsy@w.w@e 22Sn3000Ddp98FMm43Ew2900;&E3005V30$Mpttago. = .i .

                                                                                                         %2           ste; g 8;00sd ea v3R$Mlcep$@heattsp$0              .e

{2eG@2000Di st f  ;' c,35 E(sosgiesse%je199E0$2500$ja90s;aGW0kM41083$e5FrLapos sass N e u.340 A *2_000$._K151t9nJMs n - F.2_90._0;e5hM._667.:sM a 5032%1M..~ 7;

e. sm- * ' . ~3aua b d b _e;:as u39t_M2.5.J-TABLE A3-3 y

er.v.v.+=., rs rn u.,=.rm., oruswv~ wcur8=n..-- .--. c a -x =7 W?$ .-

h. ~ $5NUY +g-. - ,..-

x hSN ...- .:kkU

                                                                                                          ,.g.=,W5 6
      % Cu;M         MM$     WlQ.$;'Ck                     W k hi~SUb                                                                                     .

DR2MK50df0,. f.(44E9EstRRERW#{REPAF%Ip4FW

                                         .4ata.a4-.Ma%.@M4%%=huas,M MMY4Pild hM*4@SAswas:stWas.w$a$M3s@0$0IMME@es#!i517.a, Ma s11E                                                    ss.amemamam                                               i N13;9M2ta$sFIC4F;3Mg4030M54L31tNi!EesEGfE057: s:no "                                                               -H y#2F.4                    M 40sde$.3BI4J535 h At;35 ss.eE2epR6Cet$A014EBes:# e r
                                                                                                 !a$1113 0ett$46;4M AMiuM36 281 a83E,                         gdE39 % MASW n 14314 M356hys43418gg4r;mgamm%ap;eo$$}8 is7;ssimpsasut                     MM;391M41tSPF/es005g&            .not gem g e.n                        e.cR l

W226.1Q39tpg&5tADM w:59.atgi-0.113thaE00get;46gweg eitt%5SJtM5936f

                                                                                                          #A4136f                                           '

b336ftg e Mn58 # 8% 58.#hr-o t.0t#5meti! LM7Jspp466Wkt _.54.'2t#No.p#IIpi40GM5ei;2S%qss.ap Mme sastat jl$f3hfNdN2($ f48R15500 . d sis SCB92f 5COPEREStr$70;01M 5009?k 33M dd2ME21@lsaISA33)$#d#$k:(410ajf.$532ygMg TABLE A3-4 g m e ,--- - -- - - w - % - -.. ~ ..-,. ..u .. , I $1 $ b~f p Y k"32h"df$$LM%jaggg$NEREEAEApsj gmg q sW[. o Vn%n h m m _ 3ak t;d;PtMd M .bi k e* m ep%w wa. &4 E aI ~ d&.k iw'"a s ,. f29pjjy h. n4BD k jQ135 d58kginag7f. ge.saven-agenss.$8$$$030HN0th4P,ggEkhMkCfM n&deAMea!Lsobesww#qamgo;ggesrit@m!M&mm mun.ousm%s.mEksaa#Aemme $esh.nu ag 50

r Document Number 32-1203121-01

        ) .~ ~ ~            -m.-      ; r~y~~              ;3 -~ge3,                          3, ,          ,
                                                                                                                   -> y ~-;q
        'f 20       '.400   ' y 9.0450E    04   ,+   68 ^ 58.01        c   0.014489    ' 69.019    '  ,  69.149            -0.10% ' -0.130 i,60'S t.00e ' 9.1680E 04' ? 140 K 58.01                  <

0.0a4686 68.093 / ' 68.140' '-0.07% -0.047 h100 400'

  • 9.4030E-04'/
  • 21k'"' '.58.01, ' n 0.015062 66.392
  • 66.413 0.03% 0.021
                                                                                                                          . 0,00% ' 0,002

[ 140 i' 400 ; u'" 9.7320E ~ 04 f ,f284 f^358.01 ' ~ 0.015589 , 64.147, ' ' , 64.145

         ;       ; , y, ?          u         s' 3 < .. ,                       ; ' ~ :       >.

g, 1,20 ,,1000 u 9;0430E-04 ' 68 -145.03 0,014485 " 69.035 ' s 69.167 ' ;0.195 -0.132 '

  • 60 : 1000 ' 2 , 9.1640E-04-: ' ' 140;2 145.03 ' O.014679 68.123 68.159 ' -0.05% -0.036 f 100 'r 1000204 9.4000E 0W'J 212, ' 145i.031. ' O.015057, 66.413'* ' l 66.433, < -0.035: -0.020 y 140 % 1000 l % 9.7280E-04 ' ,2844 ^ 145.03 ^D 0.015583 v> M 173 n'> 64.168 ' O.01% ~ 0.005 k

141.03)*t 0.016092 L 62.142 J' % ,,62.332 - 'O.02% 'O.010 {170'1:004*A-05 6'1000, TM338J'N,O'id :33 p, JbfQt7 f;2CM ' , K'? y ' L 7 : ',,' '

                                                                                                                                          '}

b'20 b 3000c)~? 9.0240t-04M 688 435,10:" 0.014455 ' 69,'180L c 69.230 o '-0.07% J-0.050 9.tS80s-cap;14st435;10y -0.052 0.015045 - u.676 y d 66.500 G ,' 0.08%4 ' 68.220 % ' 0.04%' 0 fy140605 3000A J100L 3000 300sW*C {9 7170E-04/N,;ttaf S.3910EiO6& 43530 M, "0.014670 68:168'1 306 d 435;10 IV 0,015565 Y 44.246/ N 64.245~ , 0.005 ' O.001 t 180 Y3000 'O s' 1.0tS2Er-03d-:',,35&v '435.10 N, 0.016262 U5 61;495 5' 7u 61.479 L s 0.02% 0.014 l 230 ff~3000 #1.0916E-63 %';s' 446 3'435.10 N s 0.017486 ' 57.189 C 57,151' ' 0.075 - 0.038 gy c a:m '[^q , m ',^Q y ~,1;,;,'t 7 f

                                                                       ,' O.014425
                                                                                    ; - [f r
                                                                                          < + 69.324 l"      -
                                                                                                                                ^ 'y
                                                                                                              ,y*f 5, !-0.18% s 0.122
           ' 20 ,10000 + % 9.0050E-04 > '{ ' 68 1450.33                          'yN%
                                                                                                    " T 69.448 60 510000. <,; ' 9.1290E-04 ' *',140 01450.33 ' a k 0.014623 68.384 > '68.431' ^-0.07% -0.047 100 '10000'v l 9.3590E 04  ' 212+   :     1450.33/,         0.014992 ' ^66.704(, ^W 66.729 ' ,-0.04%. -0.025                      >
        !140 710000 Y' 9.6780E-0A #284 01450.33 , ~ ' O.015503f: 64.5055 %64,510  0.01%                                              0.005
         ~ 180 ^ 10000(l' / 1.0100E 05^ !356 ' 1450.33 Y ' O.01617v ' 61.810 "' ,< ' 61.801, 0.01%' O.009 e220 <10000 '< 1.0660E 03 t f. 428' 1450.33:                      0.017076V $8.563 ,e : 58.533'                   ;0.05% 0.030 260 ',100007 ' 1.1441E-03 ' ,500,1650.33 i O,, 0.018327 '54.565 :: ' 54,539  O.05% 0.026 300 .10000 ' + 1.2685E-03' 5'> 5725 1450.33 '                  O.020319s 49.214 +' > >49.346 ' -0.27% , -0.132
, l *3, ;' , V <' y : '

a ;": l C p^' 3 :,  ; t' 20 , ; y',i y,,s8.9630E-04,5 20000m _v ,, y 68 '2900.65 ,, ,0.014357' 69,651 ' ' ' 49,753, i'-0.155 -0.102 t 60. 20000 ~ J9.0900E-04 w 140J 2900.65 ;t 0.014561: ' 68;678N ', 68.725 ' ' ,-0.07% 0.047

        '1 00 s' 20000 ' , ; 9.3150E 06 :f 212 l 2900.65 W
  • 0.014921' 67,019 y'; l 67.047' . -0.04% 0.028
       }140 20000 ' ^ < 9.6240E-04 *Ji 284 x 2900.65 ',^ l 0.015416 0 64,867T' ' 64.875 - ' -0.01% -0.006 i180~ 20000;             1;0C29E-03 '_ 356 ' 2900.65;
  • D.016065 c 62.247 "'n ,i62.241
  • 1 0.01% 0.006 l220 ' 20000 Q 1.0558E 03 ;',, 428 i 2900.65 ' : 0- 016912 e < 59.129K '/,59.096, ;.' O.06% 0.033 4 260 ; 20000 J ' < 1.1275E:03w < 50012900.65!> 2: 0.018061J 55.368i s 55.335. O.06% 0.C33

[300i 20000$fj;2332E-031, 572f 2900.6510.019754.50.623,M' ; 50.69L -0.1510.074 TABLE A3-5

 -------------DYNAMIC VISCOSITY LBf-SEC/SQFT----- --------------------

PRESS TEMP FSPLIT REF 4 PERCENT ABS ATTCH 4 % DIFF PSIA F RESULTS DIFF DIFF RESULTS REF 3/4 LT 1 68 2.615E-05 2.606E-05 0.35% 9.0E-08 2.951E-05 -13.25% 2.6 140 1.146E-05 1.153E-05 -0.59% -6.8E-08 1.163E-05 -0.87% 6.4 176 8.662E-06 8.690E-06 -0.32% -2.8E-08 8.702E-06 -0.14% 13.9 212 6.885E-06 6.870E-06 0.22% 1.5E-08 6.868E-06 0.03%  ! 27.6 248 5.666E-06 5.629E-06 0.66% 3.7E-08 5.635E-06 -0.11% 53.0 284 4.780E-06 4.741E-06 0.83% 3.9E-08 4.76'.E-06 -0.41% . 87.8 320 4.109E-06 4.081E-06 0.70% 2.8E-08 4.li3E-06 -0.79% " l 143.6 356 3.585E-06 3.575E-06 0.28% 1.0E-08 3.618E-06 -1.21% l 224.1 392 3.164E-06 3.174E-06 -0.30% -9.6E-09 3.229E-06 -1.73% j 336.1 428 2.819E-06 2.850E-06 -1.10% -3.1E-08 2.917E-06 -2.36% 487.1 464 2.528E-06 2.580E-06 -2.03% -5.2E-08 2.662E-06 -3.17% 685.1 500 2.300E-06 2.350E-06 -2.16% -5.0E-08 2.450E-06 -4.22% 939.9 536 2.117E-06 2.145E-06 -1.32% -2.8E-08 2.272E-06 -5.85% 51 1

L Document Number 32-1203121-01 I l

                                                              ~^
            }7?'ra d"TC~I~ '?TABf.EIA'3 6
                  '   ~    '      '

y3-------- .- ,- LSPECIEIC: HEAT 1 BTU /LB-FCR *'M

               - PRESS 9TEMPNFSPLIT' ATTCH!4s REF176 % DIFE p'$ PSIA}% F$p;LQ('f._RESULTS j'M';g , :REF;7,

{(?t~4 c LT;l %^68M 0,9988;' l'.0124

                 .---------.----:,---,--------------_e--------
                                                                              '? r.0134 -1:144%     4 i :'s . 2',6 / 140 4 1'i0000'.'0.99891.0048 '20.48%

d'!Q'5 6'.4 - '176:t1L9007" 2 0.9958 $0',9986T';!0.21% l f $,13;9TT'Mrh001550;9933^? 0 !9938V,; 0'. 77% f'a:x256i'f 248$^t003150.9926G0.99196:'lli13% NN53WOf28 450056%0;99514f;0.99331j:!I23% W #87t8 19 82 7.0@$!c,a 12.9031'" .9972::: 'I.27%

            @44143469                             o183J1.01276-0:33%

7 n',42244t& Al.016 1 0255s,n @bliO426" 110301 6 0,45% b'n336FId428' 5i50396@ l';0770i l'.0542^ 91:38% F O 487:1 %:464 Al^;0607 /A1.'1214' ' I .0884^ v -2 ' 545 2:64% L939.9 ;;' 53621 1 1658,y?;1l.2388.'1L1;.2057, a ' 685i11500 41.1001'%'J.1744 TABLE A3-7

       ------------CONDUCTIVITY - BTV/             HR-FT-F-----------------------

PRESS TEMP FSPLIT REF 4 PERCENT ABSOLUTE ATTCH 4 % DIFF PSIA F RESULTS DIFF DIFF RESULTS REF 3/4 1.0 68 0.3380 0.3361 0.58% 0.0019 0.3262 2.95% 1450.0 68 0.3311 ------

                                                   -1.79% -0.0060 0.3371 2.6       140     0.3590     0.3611 -0.59% -0.0021 0.3626 -0.41%

1450.0 140 0.3617 ------

                                                   -0.60% -0.0022 0.3639
          '6.4       176     0.3638     0.3657 -0.53% -0.0019 0.3692 -0.95%

1450.0 176 0.3666 ------

                                                   -1.11% -0.0041 0.3707 13.9       212     0.3657     0.3675 -0.49% -0.0018 0.3723 -1.31%

1450.0 212 0.3686 ------

                                                   -1.39% -0.0052 0.3738 27.6      248      0.3649     0.3669 -0.55% -0.0020 0.3721 -1.42%

1450.0 248 0.3680 ------

                                                   -1.52% -0.0057 0.3737 53.0       284      0.3619     0.3634 -0.42% -0.0015 0.3691 -1.56%-

1450.0 284 0.3652 ------

                                                   -1.55% -0.0057 0.3709 87.8       320      0.3568     0.3582 -0.41% -0.0015 0.3637 -1.53%

1450.0 320 0.3603 ------

                                                   -1.43% -0.0052 0.3655 143.6       356      0.3499     0.3507 -0.23% -0.0008 0.3562 -1.56%

1450.0 356 0.3535 ------

                                                   -1.22% -0.0044 0.3579 224.1       392      0.3413     0.3421 -0.22% -0.0008 0.3466 -1.33%
   . 1450.0       392      0.3451     ------
                                                   -0.897 -0.0031 0.3482 336.1       428      0.3311     0.3317 -0.17% -0.0005 0.3351 -1.04%

1450.0 428 0.3350 ------

                                                   -0.41% -0.0014 0.3364 487.1       464      0.3194     0.3195 -0.04% -0.0001 0.3218 -0.71%                          ,

1450.0 464 0.3234 ------ 0.24% 0.0008 0.3226 l 685.1 500 0.3066 0.3057 0.31% 0.0010 0.3066 -0.31% l l 939.9 536 0.2921 0.2901 0.72% 0.0021 0.2895 0.19% 52 4

Document Number 32-1203121-01 Benchmark Cases I tiThe previous section discussed the water property and friction factor unce er,. This section will evaluate only the calculational errors. Therefore, in order to separate the calculational errors from the water properties and fric factor errors, the benchmark hand calculations cases will. use the same w l properties and friction factors predicted by FSPLIT. The resulting errors will be due to calculational mythology only. The following discusses ten different benchmark cases. Each case is compares calculated or known results with the FSPLIT results. Each case is designed to check one or more of the FSPLIT calculational models. The final two cases are a summary cases involving most of the FSPLIT options. These cases will be .e-run i after each version change to assure that all calculations are correct.If a version change includes a new option (not previously checked in this docum the check cases may be modified to incorporate the new option or a new che will be added. added. The check case number will describe which version the ca For example, all initial check cases will have a 4A following the check case number. If any cases are added to the list (at a version change) they will be labled as 4B or SA etc. New FSPLIT versions with cosmetic changes oniv will be numbered 48, 4C, etc. New versions with calculational changes will be numbered 5A, 6A etc. l 1 The benchmark cases analyzed are tabulated below. These analyses address all of the present options available in FSPLIT. I i 53 I'

f . l l l i Document Number 32-1203121-01 ! TABLE A4

              . CHECK C SEI;      J

[h[ D BOUN'ARN$3TIPE)(~  ; FEATURES' DEMONSTRATED-i

                       [ d CONDITIONIb M
                                                                   ,       s.
.                1-4A AP SPECIFIED   (8)        FLOW SPLITS, UNREC AP'S 2-4A FLOW SPECIFIED (7)        FLOW SPLITS, UNREC AP'S 3-4A FLOW SPECIFIED (7)        FLOW SPLITS, UNREC AP'S 4-4A FLOW SPECIFIED (1)        FRICT, MOMENTUM, ELEV AP'E 5-4A FLOW SPECIFIED (3)        PUMP H-Q, NPSH 6-4A FLOW SPECIFIED (1)

HEAT ADDED, REVERSE K-FACTR 7-4A AP SPECIFIED (2) _ NATURAL CIRCULATION 8-4A FLOW SPECIFIED (3) 9-4A FLOW DEPENDENT K-FACTORS FLOW SPECIFIED (1)

                                                       )GASFLOWPROBLEM 10-4A t

FLOW SPECIFIED (1) COMBINATION CASES 1-8 b CASE 1-4A Verify FSPLIT-parallel resistances calculation. The first case ill be a simple parallel flow example. specified across two parallel flow paths with arbitrary flow areas a able shock loss factors. unrecover-to show that the initially assumed pressure culated. drop wil This example psia (shown (at inlet). in Figure 1) will assume 200'f water at a nomina re of 1000 The flow rate for this example is; AP - (k x Wr)/(p x 29 x Ar) or W =[AP x (p x 2g x Ar) / k]o.s g = 32.1740 ft/seca (Reference 1) - 54

l Document Number 32-1203121-01 FIGURE Al PARALLEL CHANNEL FLOW LOSS FACTOR = 1 CBASED ON 1.0) AAEA = 1.0 SOFT O NODE 2 NODE 4

                                  $      NODE 1                 $

P1 = 1000 PSIA P'2 = 950 PSIA. NODE 3 s-LOSS FACTOR = 3 CBASED ON D.5) AREA = 0.5 SOFT p = 60.295 lb/ft3 (see FSPLIT analysis node 4) . AP= 50 psid =7200 psf (given) A = 1.0 ft (arbitrary) k - combination of parallel path loss factors kl and k2 adjusting k's to 1.0 fte kl = 1, k2 =(1/.5) x3 - 12 k = 1/[(1/k!) 5 + (1/k2) 5]2 k = 1/[(1/1)*5 + (1/12)*5]2 = .602161 W =[7200 x (60.295 x 64.3480 x 12) /.602161]D5 W = 6811.10 lb/sec Since.W1 = [12/l]'5 x W2 and W1 + W2 = 6811.10 W1 = 5285.37 and W2 = 1525.75 - This case analyzed in FSPLIT (see Attachment 2 for input / output of code analysis) results in the following. W = 6811.10 lb/see W1 - 5285.37 W2 = 1525.75 55

F l Document Number 32-1203121-01 bI55[IEN$There is no error in this calculation and {@i[Eh] CASE 2-4A the same pressure drop when re-input into FSPLIT. Verify that ai 1 The second test for FSPLIT will to reanalyze the above case input verify that the code will predict the same pressure drop (50 psid). The results (shown in Attachment 2) were 49.9997 psid and 5285.35 to 1525 75 flow splits to each branch, the same as the AP calculation in caseThis . 4A1 EN!5IN1}id3E5U2$N15Eindicates again, virtually zero CASE 3-4A Verify parallel path flow with a cross connect The third benchmark will be similar to case 1-4A except a cross conn be added per Figure A2. i i This case will show that two identical parallel paths i FIGURE A2 PARALLEL FLOW WITH CROSS CONNECT i LOSS FACTOA = 1 (BASED ON 1.0) AREA = 1.n SOFT

e i

NODE  ? NODE 1 NODE 4

  • 6814. 6 LB/ SEC----- g g-I NODE 3
G LOSS FACTOA =
                                                             ~

1 (BASED ON 1.0) AAEA = 1.0 SOFT 56

r l- Document Number 32-1203121-01 will not permit any cross flow between them. The FSPLIT results (shown in Attachment 2) show 3407.3 lb/sec in each of the parallel paths and .298E-6 lb/sec in the cross connect path. This is effectively a zero flow rate.

CASE 4-4A Verify elevation, momentum, and friction AP Figure A3 shows the configuration of this test case.

It consists of water flowing uphill through a pipe with area changes and a temperature increase in the fluid in the last section of pipe. Assuming the heated portion increases the temperature linearly, the average temperature in this section will be 200*F. The FIGURE A3 PIPE FLOW i o 50 " A96b SOO fEMP.200F LEMg G ELEv. 3s 57

                                                          $ SoM                                                 {

ELB' _, AhT3

  • g sH - T N * '00 5
                      .   . s. ,,e n             o,   -

get0i* ggg IN 13 FT fewp . 100 e O t l< *

          #       ELEV         TEup . 100 P oipe will also. be assumed to be clean commercial steel pipe with a relative roughness o'f 500 micro inches (.0005 inches)

Elevation AP This pressure drop is simply p x Z where's is the density and Z is the elevation i change. From the FSPLIT analysis p = 62.182 lb/ft3 (100*F,1000 psia) [ path 1, 2 node 1, 2 3) i 57 9 %/

Document Number 32-1203121-01 p = 60.305 lb/ft3 (200*F,1000 psia) [ node 4] p = 61.378 lb/ft3 (150*F,1000 psia) [ path 3]

          ~ AP(EI) = 13 ft x 62.182 lb/ft3 = 808.366 psf = 5 6137 psi l

AP(E2) = 10 ft x 62.182 lb/ft3 = 621.820 psf = 4.3182 psi ' AP(E3) = 13 ft x 61.378 lb/ft3 = 797.914 psf = 5.5411 psi The results (see Attachment 2) of the FSPLIT analysis were the same as . Momentum AP i This pressure change (due to fluid velocity change) occurs at either an change or a water density change. The area changes will effectively cause an immediate momentum ch:inge at a node and the temperature change wil gradual change over tne path length. Momentum change for path #1 l AP = AP = W2/(pl x 2g x A12) - W /(p2 x 2g x A22) or (W2/2gp) x (1/1A12 -1/A2r) At 1000 lb/sec AP = -749.7587 psf = -5.2066AP = 1000r/(64.3480 x 62.182) x (1/1r - 1/.52)  ; psi. (pressure decrease) i S%ce the expansion back to the I ft pipe is identical to the contraction (i it n a pressure increase), the momentum pressure will increase by 5 2066 at expansion. . { i The momentum AP in the heated portion will be node densities; AP = 10002/(64.3480 x 12) x (1/62.182 - 1/60.305) AP = 15540.50 x .0005006 = -7.7787 psf = .0540 psi ' i The FSPLITSA calculations for these momentum losses are 5.2066 and .0 ' respectively or zero percent error. Shock losses 58

\ Document Number 32-1203121-01 These losses will be due to the contraction as 1.0 for both). and expansion (k arbitrarily defined 1 AP = (k x W8)/(p x 2g x A8) or AP = 1.0 x 10002/(62.182 x 64.3480 x 0.52) AP = 999.678 psf = 6.9422 psi ' This is the same result FSPLIT5A calculates for path I and 2 . I . Friction losses The friction length. losses in a pipe are based on the pipe hydrauli i The pressure drop due to friction are; c diameter and pipe AP = (fL/Dh x Wr)/(p x 2g x Ar) The friction factor (f) is based on pipe diameter, Reynolds number. relative roughness and (previously discussed).The Reynolds number is based oner viscosity calculation is assumed to be correct for a specifie r and relative roughness. Therefore, following pressure drop check. the FSPLIT calculated friction or the factor wi factor calculation will be checked to confirm that they friction factor for one case.) c a reasonable Friction Factor check Per Reference 2, the viscosity of water at 100 *F and 1000 psia i sec/ft2 or 0.6807 centipoise. s .14210E-4 lbf-For a 1 ftt area pipe, the Dh = 1.128 ft = 13.54 in Re - 6.31 x 3600 x W / (d x g) = 2.2716E7 / (13.54 x Re = 2.4647E6 .6807) , (compared to 2.4639E6 calculated in FSPLIT) c/D = .0005/13.54 = .000037 Per Reference 1 page 85 f = .0113 This compares well to the FSPLIT calculation of s.01129 example for path I 59 l

Document Number 32-1203121-01 4-4A~in Attachment 2. Therefore, the friction factor is considered correct. (See Table Al for further verification of the friction factor accuracy.) The friction DP is; AP = (fL/Dh x Wr)/(p x 29 x A2) AP leg 1=(.011289 x 100/1.1283 x 10002)/(62.182 x 64.3478 x 12) AP leg 1/144 - 1.7365 psi For leg 2, the friction factor is .011412 (per FSPLIT) AP leg 2 =(.011413 x 80/.7979 x 1000r)/(62.182 x 64.3478 x .52) AP leg 2/ 144 - 7.9433 psi For leg 3, the friction factor is .010894 (per FSPLIT) AP leg 3 .010894 x 100/1.1283 x 1000r/(61.378 x 64.3478 x 12) AP leg 3/ 144 = 1.6976 psi These friction loss calculation agree very well with the FSPLIT calculated values of 1.7364, 7.9437 and 1.6976 ((C@sMijffiR((MMg(@@_{j((1A?Fe 3 !#5pjilj l Case 5-4A Head Capacity Check Case 4A5 is a simple case to check the pump head capacity option. A recircula-tion loop consisting of a one square foot pipe was analyzed for cold water (62.323 lb/ft3) at 100 psia. Arbitrary system loss factors of 40, 60, 80' and 100 velocity heads were analyzed against an arbitrary head capacity response. The sytt .a resistance vs the head capacity curve is shown below (the H-C input , is also tabulated below). The point where the resistance curves intersect the head capacity curve is the point that FSPLIT should iterate to. The results of the FSPLIT analyses (shown in Attachment 2) are; for k = 100 (form loss) 60

i Document Number 32-1203121-01 flow = 147.41 lb/sec = 1061.60 gpm AP = 3.7627 psid = 8.694 feet of water for k = 80. flow = 163.54 lb/sec = 1177.76 gpm AP = 3.7048 psid = 8.560 feet of water for k = 60 flow = 185.02 lb/sec = 1332.45 gpm AP = 3.5567 psid

                                           = 8.218 feet of water for k = 40
                    . flow - 217.52 lb/sec = 1566.51 gpm AP = 3.2773 psid
                                          = 7.572 feet of water These results agree quite well with the hand calculated intercept point on Figu A4 shown below, thereby confirming that the head capacity option is iterating the correct value.

The' I 3 head 4 capacity file data used in this analysis is shown below. 9 796.26 8.5 '1230.58 7.5 1592.51 5.5 2026.84 Hand calculated HC operating points AP = (k x Wr)/(p x 29 x Ar) A-1.0 p=62.323 2g=64.3478 AP (ft of water) = k x W2/(p2 x 64.3478) (W = lb/sec) AP (ft of water) = 4.0010E-6 x k x W2 (W = lb/sec) AP (ft of water) = 4.0010E-6 x k(W/(60x7.4805/62.323)):(W -gpm) AP (ft of water) = 7.7144E-B x,k x W (W=gpm) AP is calculated and plotted for various values of W (flow) and k (system loss factor). The results are shown in Table A5 61

n- 1 l i 1 I Document Number 32-1203121 01 1 l FIGURE A4 ' HEAD CAPACITY DATA 9.000 -

                                                                                                                                             .(

h; 1 i i

                                                         . 4 .,I I                                       f                                          I

_4 -4 4- + 8.875 . t f i f i 1 i

                                                                                                                                                                                    !                        l I#                           !
; 5; .-
                                                                                                                                                                                  .i. ..
                           .N..%s.,' g 8

T .~ -1 t 8.750  :  ? f l -.'N 8 A s 1  !

                                                           ,i,f i
                                                                ?"

s' l ~ a l s. l s t s i 8.625 't -

                                                                                                                                                      -                            i                        !

i 3  :  !!

                       .;                                  .s..                             .
                                                                                                      -                 l
                                                                                                                       .. ..                       :.                              l                        j
                                                                                     .%}~ w7..%*

s E . . . 8 500 .- .- -

                                                                                                                                                                                  '                                      i s                                                                                       y                             ;                                      -
A x .

i 4 _.* --t-r ---Y' -f---- ---F- ------ 1 i

s:  :: ri  !

B.375 - 1  : i l i i \ i 1 fi i' I '

                                  }..                     ["}i.-- ....{........                                     .L..          3,z         p ......               .            .-                       L 8.250                     !                     8I                                 I~                                         ' #                                       I                                      '

s.

                                                                                                                                       .'..sn
                                    .                                                        s
                                                                                           ,f.....                                            f,     7--
                      ......{'                           1.f..                                                .. . ......

9 8.125 ~ '

                                                                                                                                                          ^                      '                       '

2_  : '1 i -

                                                       *1                          m
                                                                                         ?                                      _          l \ \%                               !                        i m
      ] 8.000                                c'8
                                             -                                    m]:.                                         e          [ t
                                                                                                                                                             .\ ,O                                       ,

u ,, E p .. ... II : I p ... ... .ip n... .p.. 5  ! p '\ J I p-

                                                                                                                                                                                                                          \

I" 7.875 2 i V: , l lV p  ! g.g.s e. .  ! 1 l '# l

i 1 -
      'l                                           I                                                                                   I          !                            P.. V                     i 4              .'3             L                      .:. 4 .                                                 . 4                                                   .4.

2 1

l .. ..h I 4.\.

D 7.750 '  !

1. I I' ...f a

8 i p

1
                                                                                 -3. .y                               .p..J. .'.._

3 p .; .....

                                                                                                                                                                    .......ip.
                                                                                                                                                                                        \ @C'.
                                                                                                                                                                                            .'... A
                                                                                                                                                  .                            .                    N 3..

7.625 ~  ! ' ' ' ' ' ' 1 L.s L......

                                                                               ~
1 L . .. .... 4.. i
j. j.=- 4 l

N/ S

          ,.500                            .!                                 ..                                             .l                                              !,                         I,,ls, s                                  y                                              f                  .                                                           <
                      ......p..t -<                                           *          .                         +-I-- - - - - - + - - - - -                                                                   \

4- ---- 7 375 i s ------H--- i 3 l  !!  !  !

                                                                                                                                                                                                                   -i
                               . .                                                                                   ;;                                                                      o 1.Y I                  I f..              1.. .. . !g ..                               ..      i --                           I               Y                     . .I 7.250                ;8                                       -

y ,[..

                               .' 8                                     :                                           3                                                                     y
                           .. ! n                                      ~..                                         E
                                                                                                                                                                                                     ..L...

s  ! - 7.125 l8  ?

  • i f: !  !

if i : l  :

3 . 4. ..........a..... . ... ,f_ A. .a . .

l l

                                                                                                                                                                                                    ...A.....

I  !: I

                                                                                         '                      !!                              1                            1                         1 7 000                -                                   -                                            "'                              '                            '                         '

900 1,000 1,100 1,200 1,300 1,400 1,500 1,600 SYSTEM FLOW CGPM) 62 4

m Document Number 32-1203121-01 TABLE A5 FLOW FLOW SYSTEM PRESSURE DROP (FT OF WATER) (gpm) (1b/sec) k = 40 60 80 100 720.17 100.0 1.600 2.401 3.201 4.001 792.18 110.0 1.936 2.905 3.873 4.841 864.20 120.0 2.305 3.457 4.609 5.761 936.22 130.0 2.705 4.057 5.409 6.762 1008.23 140.0 3.137 '4.705 6.274 7.842 1080.25 150.0 3.601 5.401 7.202 9.002 1152.27 160.0 4.097 6.146 8.104 10.243 1224.28 170.0 4.625 6.938 9.250 11.563 1296.30 180.0 5.185 7.778 10.371 12.963 1368.32 190.0 5.777 8.666 11.555 14.444 1440.34 200.0 6.402 9.602 12.803 16.004 1512.35 210.0 7.058 10.587 14.116 17.644 1584.37 220.0 7.746 11.619 15.492 19.365 1656.39 230.0 8.466 12.699 16.932 21.165 1728.40 240.0 9.218 13.827 18.437 23.046 1800.42 250.0 10.003 15.004 20.005 25.006 These cases also demonstrates FSPLIT's net positive suction head (NPSH) calculation. All external paths with a water flow boundary condition (constant flow or pump head capacity) will have an NPSH calculation at the suction node of the external path. NPSH is simply the stagnation pressure minus the saturation pressure at the pump suction node. For the case with a system resistance of 100 (case 5-4A.100k in Attachment 2), the static pressure is 100 psia at node 4. The velocity head at node 4 is .0376 psia (see FSPLIT output). Since the temperature is 70*F, the saturation pressure is .3629 psia (Reference 2). Therefore, the NPSH is 100 + .0376 - .3629 or 99.6747 psia which agrees with the FSPLIT calculated 99.675 psia. j[@$3Esl@METrjQj@ljji @ij~@8istMTTiFishifi9 l I i CASE 6-4A Heat Addition and reverse form loss factor '

 . This case involves a simple configuration of a heat source and a heat sink per   -

Figure A5. This will change the fluid tegerature and therefore density thereby  ; changir.g the fluid velocity. This will affect all the unrecoverable and 63

Document Number 32-1203121-01 FIGURE A5 HEAT ADDITION MODEL K= 1 20 FT 20 FT O=500000 BTU / SEC -- K= 1 K= 1

                         - O=500000 BTU / SEC 10 FT   -

10 FT K= 1 K= 1 l.-- 70 F EXTERNAL PATH 2000 L8/ SEC recoverable pressure drops in the path. The last path in this example was input in the reverse direction of flow (path direction is from node 6 to node 5). A forward and reverse form loss factor of 3 and I respectively were input. The analysis should use the reverse factor and the flow in this path should be negative. Five hundred thousand BTU /sec was put into 2000 lb/sec flowing water or 250 BTU /lb was added to the water. The initfal temperature was 70*F (01000 psia) l which results ir, an enthalpy of 40.82 BTU /lb (note from table A2 above that this range of enthalpy predictions has a zero percent error). The resulting water energy level was 290.82 BTU /lb or 318.86*F and 56.91 lb/ft3 (per Reference 2). This matches the FSPLIT predicted 318.9'F and 56.91 lb/ft3 64

   ~

I

i Document Number \ 32-1203121-01 The average temperature in the path is (318 9 this temperature is 60.44 lb/ft).

                                                       + 70)/2 =194.45'F. The density at 10 ft x 60.44 lb/ft3 /144 - 4.197 psiTherefore, the elevation                      s pressure pressure change.                     .

FSPLIT predicts 4.1969 for this elevation j The momentum AP in the heated portion will b  ! e node densities; AP = 20002/(64.3478 x 12) x (1/62.498 - 1/56 9 AP = 62162.19 x .0015705 -97.624 psf =

                                                 .6779 psi, tha same answr/ as FSPLIT.

These calculations show that FSPLIT is using th e heat addition correctly. Finally, the case in Attachment show that th that the 1.0 form loss factor was used (co e flow was negative in path factor). mpared to the 0.0 forward form loss CASE 7-1A Natural Circulation \ 1 This case is similar to case 6 except that th  ! psid boundary condition and the heated path is chae external path w nged to path #1. and cooling effects (10000 BTU /sec) will causeYne n theating The natural circulation flow rate will be based a ural con , head in the two vertical sections on theand the total difference in elevation pressure psi (14.1408 drop will be the sum of the elevation he dunrecoverable The loss fac! psf) per the FSPLIT analysis . a s or 4.2419 - 4.3401 = .009820 elevation head calculations.] [See case 6-4A for verification of the average density is 61.708 lb/ft)The total loss factor is and 5.0 based on I f Therefore, the flow rate is; *  ; W = [AP x 2g x p x Ar k]* -  ! [Lrsg4~A]. This compares to 105.9555 s 105.956 lb/sec [t][If"h193i@eT@j@s))@Qij lb/sec calculated by FSPLIT. 4 65

6 Document Number 32-1203121-01 CASE 8-4A Flow Dependent Loss Factors (Y's and TEE's) FSPLIT has an option that allows a table input to iterate on form loss factors in Tee and Y branch flow. Per Reference 1, the loss factor on the various Tee and Y branches is a function of the flow rate. In most codes, this functionality would require multiple code analyses adjusting the loss factor by hand in order to converge on a network solution. However, FSPLIT has an internal iteration scheme to accomplish this. network. The following is a pictorial example of this type of In this example (see Figure 6A), the flow splits in paths 1 and 2, 5 and 8, 3 and 7, and finally into path 13 are dependent on the network loss factors. These loss factors are in turn dependent on the flow splits. FIGURE 6A FLOW SPLIT DEPENDENT MODEL b  ; ...e..@ ~}:...@. _ * .e t K= 1.0 I l Q) b a

                                                         ?

Ill K = 30 elg9. te/ sec. _, _ AREA "A"m1 SOFT < P AAEA "B"=.5 SOFT PATH '~~~~ b""hl * *'25 ~h**" K = 3.0 The loss factor in path I will be the sum of the flow split loss plus 3.0 (arbitrarily.addedresistance). Assuming an initial 7000 lb/sec flow rate, the rlow split to path I calculated by FSPLIT (Attachment 2) is 2745.07/7000 =

    .39215.

Since tJie path areas are the same, the mass flow ratio is also the velocity ratio. Per Reference 1 page 371, the loss factor in the branch pipe for this split is 1.04843 (linear interpolation of table at bottom of page in Reference 1). Since this is based on a velocity head in path 12, the loss factor must be multiplied by (V12/V1)r to base it on the velocity head in path 1. This is (1/.39215)2 x 1.04843 - 6.818. Adding the other form loss in the path (3.0 66 l l

1 Document Number 32-1203121-01  ! per above), the total loss is 9.818 which comparesy to (see Attachment 2).

                                                                 .                         9 818 ca FSPLIT factor equation is used vs using linear                                        interpola u ar points, as FSPLIT does. The larger the number of input points                             from Refer
                                                                              , the smaller the error will be.)                                                                            !

Path on 2 has page 371 a flow of Reference 1). ratio of 4254.93/7000

                                                                               = .60785 s factor (based       or a :

(a velocity ratio of 1/.60785),Correcting this to the appropriate path velocity the loss factor is 3.013. path resistance of 1.0 was added to the path resulting actor of in a total 4.013, the same value that FSPLIT calculated. Paths 4, 9,12, and 13 all have an arbitrary zero loss example. factor to simplify this i Path 6 has an input .1.0 loss factor. this section since they have no flow split dependent lossesThese! Path 3 is the top of a converging TEE configuration flow splits per page 347 of Reference 1. Its loss is dependent on The flow dependency is based on the l flow ratio of the middle of the TEE (path 7) to the or 2674.71/5419.79 = .4935. a pathcombined 9) will result in a loss factor of .5255.Per curve 2 of the reference, a flo 1 path, the result is (V9/V3)2 x .5255. Changing this loss factor to the correct Since the areas in these paths are the ' { Multiplying this by .5255 results ines exactly a 2.048 lo with the FSPLIT calculation. Path 7 (the middle of a converging TEE) will develop a loss fact Reference I page 347. or based on be based on the same flow ratio calculated above or . This will result in ka=loss

      .4168.factor of .7485 x A where A = .55 (see page 336 Refere                                    !

Correcting this to the appropriate path, the loss factor will be multiplied by the velocity ratio (or mass flow ratio)resulting of (1/.4935)2 in a loss of 1.692 velocity heads in path 7. This is compared to the FSPLIT 67

l Document Number 32-1203121-01 calculated value of 1.690. l Path 5 is the middle of a diverging TEE configuration. Its loss is dependent on flow splits per page 361 of Reference 1. The flow dependency is based on the l flow velocity ratio of the middle of the TEE (path 5) to the cambined flow path (path 4) or 85.807/68.250 (see FSPLIT results in Attachment 2) or 1.2572. Per Reference 1, a velocity ratio of 1.2572 will result in a loss factor of 1.4758 j times a multiplier. The multiplier used in this analysis was .55 so the total loss factor is .8117. Changing this l'oss factor to the correct path, the loss j factor is (V4/VS)2x.8117 or (1/1.2572): x .8117 = .5135 loss factor which! exactly with the FSPLIT calculation. l Path 8 (the top of the diverging TEE) will develop a loss factor based on Reference I page 363. l The loss in this path will be based on a flow ratio of V8/V4 or 25.347/68.250 = .3714. This will result in a loss factor of .1571 (per Reference 1). Correcting this to the appropriate path, the loss factor will be multiplied by the velocity ratio (or mass flow ratio) of (1/.3714)2 resulting in a loss of 1.139 velocity heads in path 8. This agrees well with the 1.14 value calculated by FSPLIT. Path 10 will have a total loss factor due to the combining TEE and an arbitrarily added 15 velocity heads. Per Reference 1 paae 371, the loss factor in the branch pipe for combining flows is based on V10/V13 or 25.347/112.28 .2257. This results in a loss factor of 1.4815. Correcting this to the velocity head in path 10, the loss factor is (1/.2257) x 1.4815 - 29.08. Adr' ng the fixed 30 velocity heads, the total loss factor is 59.08 which compares to the FSPLIT prediction of 59.07.

  • Path 11 will have a similar calculation except the flow ratio is 86.935/112.28
   .7742 which results in the same loss factor of 1.4815. Adjusting this to the correct velocity head, the total loss is (1/.7742)2 x 1.4815 = 2.471. Since path 11 had an additional 0.5 velocity head loss, the total loss is 2.971 which is identical to the FSPLIT results.

68

F l Document Number 32-1203121-01 This example shows that FSPLIT can accommodate multiple flow dependent loss factor paths and that the loss factors were calculated correctly. Mfis^'@ Eggg gj yTMhfy@ Ks]{@l[M] CASE 9-4A Gas flow density and Mach number calculation This is a case where air was assumed to be cooled and then reheated in a loop

  .(ie. and air conditioning system). The properties of air were input to FSPLIT from a file with the data shown below.

Temp *F Density Viscosity Conductivity Pr

      *F   lbm/ft3     lbf-sec/ft2   BTU /hr-ft *F 8       .085       3.4625E-7      .0133        .2403 44      .079       3.6697E-7      .0142        .2405 98      .071       3.9719E-7      .0155        .2406 152     .065       4.2569E-7      .0168        .2410 206     .060       4.5246E-7      .0179        .2416 260     .055       4.7922E-7      .0191        .2424 The base pressure was 14.7 psia.
  • A pressure drop was induced in the loop to verify that FSPLIT calculates the density correctly. Nodes 2 and 6 will be checked. Node 2 temperature and pressure are 8'F and 27.5554 psia. From the table above, the density of 8.0*F and 14.7 psia is .085. Therefore, 27.5554 /14.7 x .085 is .15933 lb/ft3 which is the same as the FSPLIT calculation. Node 6 is at 206*F and 15.0 psia. The corrected density is .060 (see table above) x 15/14.7 or .06122 lb/ft*, again the same as the FSPLIT result.

The Mach number (maximum velocity) of a gas is defined as; V(ft/sec) = [g x R x k x T]' where g = 32.174 ft-lb/(lb, -sect) R = Gas constant = 1545/MW ft-lb,/(lb, *R) MW = molecular Weight of gas 69

Document Number 32-1203121-01 T = gas temperature (*R) k - C/C, of gas For this case, the path with the highest velocity was path 5 at 476.05 ft/sec. The velocity of sound was; V=[32.174x1545/29x1.4x(179+459.73)f V = 1238.133 ft/sec Therefore, the Mach number was 476.05/1237.979 = .385, the same value calculated by FSPLIT. N W W Q ?'!E S N 5 0 N5$ U 5 5 Ts))fiff G f C Q } CASE 10-4A Combination case This case contains a combination of flow dependent loss factors, all types of pressure drop calculations, a pump, and heat input. These results are not discussed in detail other than all future code updates will require that cases 9-A4 and 10-A4 be reanalyzed to determine if the results are the same. This section will discuss any differences that may occur. The pump H-C file used for this case is listed below. 1 3 4 500 200 300 3000 200 3500 l 10 5000 1 Per Attachment 2, the resulting pressure drop and flow are 420.98 lbm/sec and 127.56 psia (at 62.401 ' Ibm /ft3 density). 127.56 psia 1.s 294.36 ft which results in a 3028.2 gpm flow rate per the input H-C table above. This is the same flow predieted in Attachment 2. yi[@[da!4fLrid@@j3Ni(93WAM311])h j lj  ! $.{lfEES W DPEI M E 2 @ @Y N d N $ dI$ $25 M 6 M K NIfIf851Eisl 5115M RferisqLiid!!!iiisinuR4@H@shvi#Bs1EstauisI@siDTeisissitnggi200Is

                                                            ~

HIMIDJ!it!NENGIGifa35EMHa71EE3Fi5iiittaliEEBiiEInMa~IH iiresTitwr415Eii!!!iigggerirgtE11ri!5EMZi!?iewfsnygggig 70

Document Number 32-1203121-01 ATTACHMENT 2 FSPLIT RESULTS OF BENCEMARK CASES Per Attachment 1, analyzed with FSPLIT and' compared to known resultsten diffe output of these cases is listed in this attachment The input and i checksums except the cases were.all the same as the revision The 00 FSPLIT5A where data had to be added (FSPLIT4A) cases support the new input for flow dependent loss to factors) the files (to of the and 5A. benchmark cases were virtually identical between . The results ve rsions 4A calculation in' cases where heat was added ensity(on the These differences were due to the enthalpy calculati .002% onal change). The ten cases include; . 1-4A. 2-4A. Parallel channel flow 3-4A. Network flow (from a DP case) 4-4A. Parallel channel flow with cross connectwill predict same DP 5-4A. Recoverable / unrecoverable losses 6-4A. Heat input caseHead capacity cases (4 separate cases) 7-4A. 8-4A. Natural circulation' case 9-4A. Flow dependent form loss factors 10-4A. Gas flow case Summary combined case with most of cases 1 through 8ons op'ti CASE 1-4A PARALLEL CHANNEL F'AW I FSPLIT SA - FLOW SPLITS by DA FARNSWORTH & JA WEIME for the generation and solution of steady state The networks 100 paths. canma The problem contain any combination of up to 100 nodes a flow boundary conditions (y be specified by imposing up to 10 boundary condition. Problems may be defined with eitherexternal pa temperature inputs or heat inputs to nodes and paths . FSPLIT greatly simplify model development.uses graphical on screen m In addition to individual path flow rates, FSPLIT solves for momentum elevation friction and form loss pressure drops. The form, loss pressu,re drop ca,lculations can assume a co loss factor or can calculate a flow dependent loss factor. Ecuivalent heat rates are calculated for temperature input ppoblems and temperatures are calculated for heat input problem . i 71 j i l j

l 1 Document Number 32-1203121-01 External head-capa. path input flows may be constant or specified as a city relationship to simulate a pump. FSPLIT is designed to accommodate incompressible water (H2O), heavy water (D20), any fluid, and gasses. routines for light water and heavy water while properties forFSPLIT gasses and incompressible fluids must be specified by the user. Help screens Creation Modes.are available in both the Node Creation and Path create a model and complete the analysis.They should provide ad If you have problems 4 consult the user's manual. concerns, contact one of the authors.If the manual does not answer your , FSPLITAll FORTRAN. was written in a combination of Microsoft QuickBASIC calculations are in double (64 bit) precision. matrix solutions are performed using the LINPACK solvers. All Exec-The FSPLIT5A.EXE, EXECHK.COM and FSPLIT.H If FSPLIT is to be executed on a Hercules type monitor, the MSHERC.COM program must be run. PRIOR TO EXECUTING THIS CODE, CONTACT ONE OF THE AUTHORS TO ASSURE THAT THIS IS THE LATEST VERSION. FAILURE TO DO S0 MAY INVALIDATE YOUR ANALYSIS. USER'S MANUAL - BWNT-TM-63 (ref 32-1203121-01) FSPLIT 5A FSPLIT SA - FLOW SPLITS by DA FARNSWORTH & JA WEIMER - Light W Check with the authors for the latest version.

SUMMARY

of RESULTS From Input File B: CASE 1 with a Checksum of 0.231685253D+04 CASE 1-4A PAFALLEL CHANNEL FLOW This unrecoverable Only is a Type 8 Case. los. The Total Pressure Drop is Specified. ses are considered.  ! t The system pressure at node 4 ) is 950 psia. There are 4 nodes con (nected by 4 internal flow paths. There is I external flow path. A local convergence criterion of 0;1000D-05 was used with  ! 0.95370-06 met. Overall Convergence was met. Solution required 21 local iterations. I 72 l I

Document Number 32-1203121-01 FSPLIT SA - FLOW SPLITS by DA FARNSWORTH & JA WEIMER - Light Water Option i CASE 1-4A PARALLEL CHANNEL FLOW

                    ----- EXTERNAL PATH INFORMATION -----

Press Drop Ext Node Node Path Flow Node IN Node OUT Path IN OUT Lbm/Sec Lbf/Sq inch Press, psi Press, psi k k k b$bbikkb bk b$bbbbbb bb b$bbbbbb bh b$kbbbbb bk FSPLIT 5A - FLOW SPLITS by DA FARNSWORTH & JA WEIMER - Light Water Option CASE 1-4A PARALLEL CHANNEL FLOW l

                    +---- INTERNAL PATH INFORMATION -----

Connecting Path Flow Press Drop Unrec DELTAP Recov DELTAP Path Node-> Node Lbm/See Lbf/Sq inch Lbf/Sq inch Lbf/Sqinch ! k 1 h b.hkbhb7b bk b.hbbbObb bh b hbbbbbb bh b$bbbbbbb bb l 2 1 3 - 0.152575D+04 0.500000D+02 0.500000D+02 0.000000D+00 l 3 2 4 0.528537D+04 0.000000D+00 0.000000D+00 0.000000D+00 4 3 4 0.152575D+04 0.000000D+00 0.0000000+00 0.000000D+00 FSPLIT SA - FLOW SPLITS by DA FARNSWORTH & JA WEIMER - Light Water Option CASE 1-4A PARALLEL CHANNEL FLOW

                     ----- INTERNAL PATH INFORMATION -----

Connecting Unrecoverable DP Recoverable DP Path Node-> Node Vel Head Friction Total UR Momentum Elevation Total RE ! [ k h bb$bbbb b$bbbb bb$bbbb bbbbb b$bbbb b$bbbb 2 1 3 50.0000 0.0000 50.0000 0.0000 0.0000 0.0000 ! 3 2 4 0.0000 0.0000 0.0000 0.0000 0.0000 0.0000 4 3 4, 0.0000 0.0000 0.0000 0.0000 0.0000 0.0000 FSPLIT 5A - FLOW SPLITS by DA FARNSWORTH & JA WEIMER - Light Water Option CASE 1-4A PARALLEL CHANNEL FLOW l

                     ----- INTERNAL PATH INFORMATION -----

73 I

Document Number 32-1203121-01 Path Node Path Flow Input Heat In Out Velocity Path HTC Conductivity Lbm/Sec Btu /Sec Ft/Sec B/Hr-Ft^2-F B/Hr-Ft-F k k b 2 1 3 b$bhbb4b b4 b$bbbbOb bb bIbhbbbb bb b$bbb 0.152580+04 3 2 4 0.000000+00 0.50609D+02 0.00000D+00 0.39169D+00 4 3 4 0.528540+04 0.00000D+00 0.87658D+02 0.00000D+00 0.39169D 0.152580+04 0.00000D+00 0.50609D+02 0.00000D+00 0.391690 FSPLIT 5A - FLOW SPLITS by DA FARNSWORTH & JA WEIMER - Light Water Op CASE l-4A PARALLEL CHANNEL FLOW

       ~
                           ----- INTERNAL PATH INFORMATION .----

Path Node Vel Head Reynolds In Out Lbf/Sq Inch Friction Density Viscosity Number Factor

............                ...........     ...........                      Lbm/Ft^3      Lbf-Sec/Ft^2 1       I       2                                                                       . . . . . . . . . . . .

2 1 3 0.50000D+02 0.29101D+08 0.00000D+00 0.602950+02 0.6360/D-0 3 2 4 0.16667D+02 0.118800+08 0.000000+00 0.60295D+02 0.63697D-0 4 3 4' 0.500000+02 0.291010+08 0.00000D+00 0.60295D+02 0.636970-05 O.166670+02 0.118800+08 0.00000D+00 0.602950+02 0.636970-05 FSPLIT SA - FLOW SPLITS by DA FARNSWORTH & JA WEIMER - Light Water Option CASE l-4A PARALLEL CHANNEL FLOW

                         ----- INTERNAL PATH INFORMATION -----

Path Node In Out Form Loss Path Temp Flow Area Roughness Deg F Ft^2 M inches Hyd Dia Fric Len Feet Feet 1 1 2 2 1 3 0.1000D+01 0.20000+03 0.1000D+01 0.0000D+00 0.1128D+01 0.0000D 3 2 4 0.3000D+01 0.20000+03 0.5000D+00 0.0000D+00 0.7979D+00 0.0000D 4 3 4 0.00000+00 0.2000D+03 0.1000D+01 0.0000D+00 0.11280+01 0.0000D . 0.0000D+00 0.20000+03 0.5000D+00 0.0000D+00 0.79790+00 0.0000D+ FSPLIT 5A - FLOW SPLITS by DA FARNSWORTH & JA WEIMER - Light Water Option

              .                      CASE 1-4A PARALLEL CHANNEL FLOW                                                  !
                        .---.---- NODE INFORMATION ---------

Path Inlet Rho Outlet Rho Inlet VH Outlet VH Inlet P Outlet P 74

I' . Document Number 32-1203121-01 (Nodes) Lbm/Ft^3 lbm/Ft^3

             ........   ........... ...........      ..../Sq In Lbf/Sg In lbf PSI        FSI 2 1 3 3 2 4                                                              1000.0000 950.0000 4 3 4                                                               950.0000 950.0000 950.0000 950.0000 FSPLIT SA - FLOW SPLITS by DA FARNSWORTH & JA WEIMER - Li CASE 1-4A PARALLEL CHANNEL FLOW Connecting ----- PATH /N0DE DESCRIPTION --.-.

Path Node-> Node Description 1 I 2 LEFT BRANCH 2 1 3 RIGHT BRANCH 3 2 4 4 3 4 DUMMY PATH TO COMMON POINT DUMMY PATH TO COMMON POINT (RIGHT) FSPLIT SA - FLOW SPLITS by DA FA"MSWORTH & JA WEIMER - Light W 4 1 8 CASE 1-4A PARALLEL CHANNEL FLOW LEFT BRANCH I 0.0000000+00 0.112838D+01 0.0000000+002 0.100000D+ RIGHT BRANCH 1-j 0.0000000+00 0.7978850+00 0.000000D+003 0.300000D DUMMY PATH TO COMMON POINT 2 0.000000D+00 0.1128380+01 0.000000D+004 0.0000000+ DUMMY 3 PATH'TO COMMON POINT (RIGHT) , 0.0000000+00 4 0.000000D+00 0.0000000+00 0.200000D+03 0.5000000+00 0.797885D+00 0.000000D+00 4 1 50 i 4 1 950 - 0 12 52 4 12 10 28 68 8 22 28 36 16 22 42 52 12 34 i 75

Document Number 32-1203121-01 CASE 2-4A FSPLIT 5A - FLOW SPLITS by DA FARNSWORTH & JA WEIMER FSPLIT is a PC based thermal hydraulics computer code designed for the generation and solution of steady state ficw networks. The networks can contain any combination of up to 100 nodes and 100 paths. The problem may be specified by imposing up to 10 flow boundary boundary conditions (external paths) or one pressure drop condition. Problems may be defined with either temperature inputs or heat inputs to nodes and paths. FSPLIT uses graphical on screen modeling and parameter specification to greatly simplify model development. In addition to individual path flow rates, FSPLIT solves for momentum, elevation, friction, and form loss pressure drops. The form loss pressure drop calculations can assume a constant lo:s factor or can calculate a flow dependent loss factor. Equivalent heat rates are calculated for temperature input problems and temperatures are calculated for heat input problems. External path input flows may be constant or s head-capacity relationship to simulate a pump.pecified FSPLITas is a designed to accommodate water (H2O), heavy water (020), any incompressible fluid, and gasses. FSPLIT has internal property routines for light water and heavy water while properties for gasses and incompressible fluids must be specified by the user. Help screens are available in both the Node Creation and Path Creation Modes. Ther should provide adequate knowledge to create a model and c.,mplete the analysis. If you have problems consult the user's manual. If the manual does not answer your , concerns, contact one of the authors. FSPLIT was written in a combination of Microsoft QuickBASIC and FORTRAN. All calculations are in double (64 bit) precision. All matrix solutions are performed using the LINPACK solvers. Exec-ution of FSPLIT requires roughly 550 kilobytes of free DOS memory. The FSPLIT5A.EXE, EXECHK.COM and FSPLIT.HLP files must be present. If FSPLIT is to be executed on a Hercules type monitor, the MSHERC.COM program must be run. PRIOR TO EXECUTING THIS CODE, CONTACT ONE OF THE AUTHORS 76

Document Number 32-1203121-01 TO ASSURE THAT THIS IS THE LATEST VERSION. FAILURE TO DO S0 MnY INVALIDATE YOUR ANALYSIS. USER'S MANUAL - BWNT-TM-63 (ref 32-1203121-01) FSPLIT SA ' FSPLIT 5A - FLOW SPLITS by DA FARNSWORTH & JA WEIMER - Light Water Opt Check with tha authors for the latest version.

SUMMARY

of RESULTS From Input File B: CASE 2 with a Checksum of 0.907895253D+04 CASE 2-4A RERUN CASE 1 WITH FLOW INPUT (T0 GET SAME DP This is a Type 7 Case. The Total Flow is Specified. Only unrecoverable losses a e considered. The system pratsure at node 4 ) is 950 psia. There are 4 node; con (nected by 4 internal flow paths. There is I external flow path. A local convergence criterion of 0.10000-05 was used with 0.72640-06 met.  ! Overall Convergence was met. Solution required 26 local iterations. FSPLIT 5A - FLOW SPLITS by DA FARNSWORTH & JA WEIMER - Light Water Option i CASE 2-4A RERUN CASE 1 WITH FLOW INPUT (TO GET SAME DP)

                  ----- EXTERNAL PATH INFORMATION -----

Ext Node Node Path Flow Press Drop Path IN Node IN Node OUT OUT lbm/Sec Lbf/Sq inch Press, psi Press, psi k k k b$bbkkkb hk b bbbbbb bb b 9hbbbb bb bIkhbbbb bk For External Path I the available NPSH is 0.97538D+03 psia (200.0 F suction) FSPLIT SA

               ' FLOW SPLITS by DA FARNSWORTH & JA WEIMER - Light Water Option CASE 2-4A RERUN CASE 1 WITH FLOW INPUT (TO GET SAME DP)
                 ----- INTER'     ?ATH IN"^RMATION -----
                                            '7

i Document Number 32-1203121-01 l Connecting Path Flow Press Drop Unrec DELTAP Recov DELTAP l Path Node-> Node Lbm/Sec Lbf/Sq inch ibf/Sq inch Lbf/Sq inch l 1- 1 2 0.528535D+04 0.499997D+02 0.499997D+02 0.0000000+00 2 1 3 0.152575D+04 0.4999970+02 0.499997D+02 0.000000D+00 3 2 4 0.528535D+04 0.000000D+00 0.000000D+00 0.0000000+00 4 3 4 0.152575D+04 0.000000D+00 0.000000D+00 0.000000D+00 FSPLIT SA - FLOW SPLITS by DA FARNSWORTH & JA WEIMER - Light Water Option CASE 2-4A RERUN CASE 1 WITH' FLOW INPUT (TO GET SAME DP)

                          ----- INTERNAL PATH INFORMATION ---..                                               I l

Connecting Unrecoverable DP Recoverable DP l Path Node-> Node Vel Head Friction Total UR Momentum Elevation Total RE

         .... .... .... ........ ........ ........ ........ ........ ........                                 j 1        1        2    49.9997         0.0000       49.9997    0.0000      0.0000        0.0000 2        1        3    49.9997         0.0000       49.9997 0.0000      0.0000        0.0000   l 3        2        4      0.0000        0.0000        0.0000    0.0000      0.0000        0.0000 4        3        4      0.0000        0.0000        0.0000    0.0000      0.0000        0.0000
                                                                                            .                 l FSPLIT 5A - FLOW SPLITS by DA FARNSWORTH & JA WEIMER - Light Water Option l                    CASE 2 4A RERUN CASE 1 WITH FLOW INPUT (TO GET SAME DP) i
                         ----- INTERNAL PATH INFORMATION -----

Path Node Path Flow Input Heat Velocity In Out Path HTC Conductivity Lbm/Sec Btu /Sec Ft/Sec B/Hr-Ft^2-F B/Hr-Ft-F I 1 2 2 1 3 0.52854D+04 0.000000+00 0.876580+02 0.000000+00 0.391690+00 3 2 4 0.15257D+04 0.00000D+00 0.506090+02 0.000000+00 0.391690+00 4 3 4 0.52854D+04 0.00000D+00 0.876580+02 0.00000D+00 0.391690+00

   .                      0.15257D+04 0.00000D+00 0.50609D+02 0.00000D+00 0.39169D+00 FSPLIT SA - FLOW SPLITS by DA.FARNSWORTH & JA WEIMER - Light Water Option CASE 2-4A RERUN CASC 1 WITH FLOW INPUT (T0 GET SAME DP)
                        ----- INTERNAL PATH INFORMATION -----

Path Node Vel Head Reynolds Friction Density In Out Lbf/Sq Inch Viscosity Number Factor Lbm/Ft^3 lbf-Sec/Ft^2 78

Document Number 32-1203121-01 I 1 2 2 0.500000+02 0.29101D+08 0.00000D+00 0.60295D+02 0.636970-05 1 3 3' 0.16667D+02 0.118 BOD +08 0.000000+00 0.60295D+02 0.63697D-05 2 4 4 3 4 0.500000+02 0.29101D+08 0.00000D+00 0.60295D+02 0.63697D-05 0.16667D+02 0.11880D+08 0.00000D+00 0.60295D+02 0.63697D-05 FSPLIT SA - FLOW SPLITS by DA FARNSWORTH & JA WEIMER - Light Water Option CASE 2-4A RERUN CASE 1 WITH FLOW INPUT (T0 GE1 SAME DP)

                            ----- INTERNAL PATH INFORMATION -----

Path Node Form Loss Path Temp Flow Area Roughness Hyd Dia Fric Len In Out Deg F Ft^2 M inches Feet Feet i 1 1 2 0.1000D+01 0.20000+03 0.10000+01 0.0000D+00 0.1128D+01 0.0000D+00 2 1 3 0.3000D+01 0.2000D+03 0.5000D+00 0.0000D+00 0.7979D+00 0.00000+00 3 2 4 4 0.0000D+00 0.2000D+03 0.1000D+01 0.0000D+00 0.11280+010.0000D+00 3 4 0.0000D+00 0.2000D+03 0.5000D+00 0.00000+00 0.7979D+00 0.00000+00 j I FSPLIT 5A - FLOW SPLITS by DA FARNSWORTH & JA WEIMER - Light Water Option CASE 2-4A RERUN CASE 1 WITH FLOW INPUT (T0 GET SAME DP)

                           --------- N00E INFORMATION ---------

Path Inlet Rho Outlet Rho Inlet VH Outlet VH Inlet P Outlet P (Nodes) Lbm/Ft^3 lbm/Ft^3 lbf/Sg In Lbf/Sq In PSI PSI 1 1 1 2 999.9997 950.0000 2 1 3 999.9997 950.0000 3 2 4 950.0000 950.0000 4 3 4 950.0000 950.0000 FSPLIT SA - FLOW SPLITS by DA FARNSWORTH & JA WEIMER - Light Water Option CASE 2-4A RERUN CASE 1 WITH FLOW INPUT (T0 GET SAME DP)

                          ----- PATH / NODE DESCRIPTION -----

Connecting Path Node-> Node . Description , 79 , 1 i t..

        ....       .... ....                               Document Number 32-1203121-01 I        1       2    LEFT BRANCH 2         1       3    RIGHT BRANCH                                           !

3 2 4 4 3 4 DUMMY PATH TO COMMON POINT i DUMMY PATH TO COMMON POINT (RIGHT) FSPLIT 5A - FLOW SPLITS by DA FARNSWORTH & JA WEIMER - 4 1 LEFT BRANCH 1 7 CASE 2-4A RERUN CASE 1 WITH FLOW INPUT (T0 G 0.000000D+00 0.1128380+01 0.000000D+002 0.100000 RIGHT BRANCH  : 1 0.0000000+00 0.797885D+00 0.000000D+003 0.300000 DUMMY PATH TO COMMON POINT 2 0.000000D+00 0.112838D+01 0.0000000+004 0.000000 DUMMY 3 PATH TO COMMON POINT (RIGHT) ' 0.000000D+00 4 1 0.797885D+00 0.000000D+004 0.000000 1 4 6811.1 1 12 950 0 52 4 12 10 28 68 8 22 28 . 36 16 22 42 52 i 12 34 333333333333333333333333333333333333333333 33333333333BJ33333333333333 CASE 3-4A FSPLIT SA - FLOW SPLITS by DA FARNSWORTH i JA WEIMER for the Ieneration and solution of steady state flow n The nctworks 100 paths. can contain The problem ma any combination of up to 100 nodes and flow boundary conditions (y be specified by imposing up to 10 boundary condition. external paths) or one pressure drop Problems may be defined with either temperature inputs or heat inputs to nodes and paths. FSPLIT greatly simplify model da elopment.uses graphical on screen mod 80

Document Number 32-1203121-01

                                                                                      )

i in addition to individual path flow rates, FSPLIT solves for i momentum, elevation, friction, and form loss pressure drops. i The form loss pressure drop calculations can assume a constant l loss factor or can calculate a flow dependent loss factor. Equivalent heat rates are calculated for temperature input _ prelems and temperatures are calculated for heat input problems. External path input flows may be constant or s head-capacity relationship to simulate a pump.pecified FSPLIT is as a designed to accommodate incompressible fluid, and gaeses. water (H2O), heavy water (D20), any routines for light water and heavy water while properties forFSPL gasses and incompressible fluids must be specified by the user. i Help screens Creation Modes. are available in both the Node Creation and Path create a model and complete the analysis.They should provide If you have problems consult the user's manual. concerns, contact one of the authors.If the manual does not answer your , FSPLITAllwas FORTRAN. written calculations arein a combination in double of Microsoft Quick (64 bit) precision. matrix solutions are performed using the LINPACK solvers. All Exec-The FSPLIT5A.EXE, EXECHX.COM and FSPL If FSPLIT is to be executed on a Hercules type monitor, the MSHERC.COM program must be run. PRIOR TO EXECUTING THIS CODE, CONTACT ONE OF THE AUTHO TO ASSURE THAT THIS IS THE LATEST VERSI0s. FAILURE TO 00 S0 MAY INVALIDATE YOUR P!ALYSIS. USER'S MANUAL FSPLIT-SA BWNT-TM-63 (ref OE-1203121-01) ' i FSPLIT SA - FLOW SPLITS by DA FARNSWORTH & JA WEIMER \ Check with the authors for the latest version.

SUMMARY

of RESULTS From Input File B: CASE 3 with a Checksum of 0.9288280760+04 CASE 3-4A PARALLEL CHANNEL FLOW This is a Type 7 C m . Only unrecoverable losses ~are considered.The Total Flow is Specified. The system pressure at node 4 ) is 950 psia. There are 4 nodes con (nected by 5 internal flow paths. There is I external flow path. 81  ! l

Document Number 32-1203121-01 A local convergence criterion of 0.10000-05 was used with 0.7132D-0 . Overall Convergence was met. Solution required 26 local iterations. I FSPLIT SA - FLOW SPLITS by DA FARNSWORTH & JA WEIMER - L CASE 3-4A PARALLEL CHANNEL FLOW

                      ----- EXTERNAL PATH INFORMATION -----

Ext Node Node Path Flow Press Drop Path IN OUT Lbm/Sec Lbf/Sq inch Node IN Node OUT

  .... .... ....         ...........                     Press, pst     Press, psi 1    4       1                                                   ...........

0.68146D+04 0.20988D+02 0.95000D+03 0.970990+03 { For External Path I the available NPSH is 0.95925D+03 psia (200.0 F suction) i'

                                                                                                  \

FSPLIT 5A - FLOW SPLITS by DA FARNSWORTH & JA WEIMER - Lig { CASE 3-4A PARALLEL CHANNEL FLOW

                    ----- INTERNAL PATH INFORMATION -----

Connecting Path Flow Path Node-> Node Press Drop Unree DELTAP Lbm/Sec. Lbf/Sq inch Recov DELTAP

     .... .... ....            ............                       Lbf/Sq inch      Lbf/Sq inch 1      1       2                                                         ............

0.3407530+04 0.207798D+02 j 2 1 3 0.3407300+04 0.207798D+02 0.0000000+00 3 2 4 0.2077980+02 0.207798D+02 0.3407300+04 0.2077980+00 0.0000000+00 4 3 4 0.3407300+04 0.207798D+00 0.0000000+00 5 2 3 0.2077980+00 0.207798D+00 0.2980230-06 0.000000D+00 0.000000D+00 0.000000D+00 0.0000000+00 1 FSPLIT SA - FLOW SPLITS by DA FARNSWORTH & JA WEIMER - Ligh CASE 3-4A PARALLEL CHANNEL FLOW

                  ----- IN TERNAL PATH INFORMATION -----

Connecting Unrecoverable DP Recoverable DP Path Node-> Node Vel Head Friction Total UR Momentum Ele k [ h hb$hhbb b[bbbb hbIhhbb 2 1 3 20.7798 b$bbbb b$bbbb "b[bbbb 0.0000 20.7798 . 0.0000 0.0000 0.0000 82

                                                                                                  )

i 4

Document Number 32-1203121-01 3 2 4 0.2078 0.0000 0.2078 0.0000 0.0000 0.0000 4 3 4 0.2078 0.0000 0.2078 0.0000 0.0000 0.0000 5 2 3 0.0000 0.0000 0.0000 0.0000 0.0000 0.0000 FSPLIT 5A - FLOW SPLITS by DA FARNSWORTH & JA WEIMER - Light Water Option CASE.3-4A PARALLEL CHANNEL FLOW

                     ----- INTERNAL PATH INFORMATION -----

Path Node Path Flow Input Heat Velocity Pa'th HTC Conductivity In Out Lbm/Sec Btu /Sec Ft/Sec B/Hr-Ft^2-F B/Hr-Ft-F I 1 2 0.34073D+04 0.000000+00 0.565100+02 0.000000+00 0.391690+00 2 1 3 0.340730+04. 0.000000+00 0.56510D+02 0.00000D+00 0.391690+00 3 2 4 0.34073D+04 0.000000+00 0.56510D+02 0.00000D+00 0.3916)D+00 4 3 4 0.34073D+04 0.00000D+00 0.565100+02 0.00000D+00 0.391690+00 5 2 3 0.298020-06 0.000000+00 0.49427D-08 0.00000D+00 0.391690+00 FSPLIT 5A - FLOW SPLITS by DA FARNSWORTH & JA WEIMER - Light Water Option CASE 3-4A PARALLEL CHANNEL FLOW

                    ----- INTERNAL PATH INFORMATION -----

Path Node Vel Head Reynolds Friction Density Viscosity In Out Lbf/Sq Inch Number Factor Lbm/Ft^3 Lbf-Sec/Ft^2

  'l      1     2     0.207800+02 0.18760D+08 0.000000+00 0.60295D+02 0.636970-05 2      1-    3     0.207800+02 0.18754D+08 0.00000D+00 0.60295D+0? 0.636970-05 3      2     4     0.20780D+02 0.18760D+08 0.000000+00 0.6029FSt02 0.636970-05 4      3    14 0.207800+02 0.18754D+08 0.000000+00 0.60295D+02 0.636970-05 5      2-    3     0.15897D-18 0.16403D-02 0.000000+00 0.60295D+02 0.636970-05 FSPLIT SA - FLOW SPLITS by DA FARNSWORTH & JA WEIMER - Light Water Option CASE 3-4A PARALLEL CHANNEL FLOW
                   ----- INTERNAL PATH INFORMATION -----

Path Node Form Loss Path Temp Flow Area Roughness- Hyd Dia Fric Len In Out Deg F Ft^2 M inches Feet Feet

 ,                                              83

1 Document Number 32-1203121-01 1 2 2 1 3 3 4 2 3 4 4 0.1000D+01 0.2000D+03 0.1000D+0 0.1000D-01 0.2000D+03 0.10000+01 0.0000D+00 0.11 5 2 3 0.10000-01 0.2000D+03 0.10000+01 0.0000D+00 0.112 0.00000+00 0.2000D+03 0.1000D+01 0.0000D+00 0.11 FSPLIT SA - FLOW SPLITS.by DA FARNSWORTH & JA WEIMER - Ligh CASE 3-4A PARALLEL CHANNEL FLOW

                          --------- NODE INFORMATION ---------

Path Inlet Rho Outlet Rho Inlet VH Outlet VH (Nodes) Lbm/Ft^3 Lbm/Ft^3 Inlet P Outlet P

       ........     ........... ........... ..../Sq lbf In Lbf/Sq In PSI        PSI I 1 2                                                             ........   ........

2 l' 3 970.9876 950.2078 3 2 4 970.9876 950.2078 4 3 4 950.2078 950.0000 5 2 3 950.2078 950.0000 950.2078 950.2078 FSPLIT SA - FLOW SPLITS by DA FARNSWORTH & JA WEIMER - Light W CASE 3-4A PARALLEL CHANNEL FLOW

                        ----- PATH /N0DE DESCRIPTION -----

Connecting Path Node-> Node Description 1 1 2 LEFT BRANCH t 2' 1 3 RIGHT BRANCH 3 2 4 I 4 3 DUMMY PATH TO COMMON POINT 4 5 2 3 DUMMY PATH TO COMMON POINT (RIGHT) CROSS CONNECT  !

                                                                                              }

FSPLIT SA - FLOW SPLITS by DA FARNSWORTH'& JA WEIMER . Light W 5 1 7 CASE 3-4A PARALLEL CHANNEL FLOW LEFT BRANCH 1 0.0000000+00 0.1128380+01 0.000000D+002 0.100000D+01 RIGHT BRANCH 1 3 0.1000000+01 0.1000000+01 0.200000D+03 0.1000000+01 . 84

r Document Number 32-1203121-01 0.0000000+00 0.112800D+01 0.000000D+00 DUMMY PATH TO COMMON POINT 2 0.0000000+00 0.112838D+01 0.000000D+004 0.10000 DUMMY 3 PATH. TO COMMON POINT (RIGHT) 0.000000D+00 0.1128000+01 0.000000D+004 0.10000 CROSS CONNECT 2 0.000000D+00 0.112800D+01 0.0000000+003 0.00000 4 1 4 6814.6 1 12 950 0 52 12 4 28 68 10 8 22 28 36 16 22 42 52 12 34 E5553333E33F5353333333333333333333333 3333333333333333333333333333553335 CASE 4-4A FSPLIT SA - FLOW SPLITS by DA FARNSWORTH & JA WEINER for the generation and solution of steady state flow The networks 100 paths. can contain The problem ma any combination of up to 100 nodes and flow boundary conditions (y be specified by imposing up to 10 boundary condition. Problena may be defined with eitherex.ernal paths) temperature inputs or heat inputs to nodes and paths.FSPLIT greatly simplify model development.uses graphical on screen mode In addition to individual path flow rates, FSPLIT solves for

 .            momentum, elevation, friction, and form loss pressur loss factor or can calculate a flow dependent loss factor.

Equivalent heat rates are calculated for temperature input problems and temperatures are calculated for heat input problems. External path input flows may be constant or s head-capacity relationship to simulate a pump.pecified as a FSPLIT is designed incompressibleto accommodate water (H2O), heavy water (020), any fluid, and gasses. routines for light water and heavy water while properties forFSPLIT gasses and incompressible fluids must be specified by the user. 85

l Document Number 32-1203121-01 Help screens Creation Modes. are available in both the Node Creation and Path create a model and complete the analysis.They should provide If you have problems consult the user's manual. concerns, contact one of the authors.If the manual does not answer your , FSPLIT FORTRAN. All was written calculations are inin a combination double of Microsoft Q (64 bit) precision. matrix solutions are performed using the LINPACK solvers. All Exec-The FSPLIT5A.EXE, EXECHK.COM and FSP . If FSPLIT isprogram MSHERC.COM to bemust executed be run. on a Hercules type monitor, the . PRIOR TO EXECUTING THIS CODE, CONTACT ONE OF THE AU TO ASSURE THAT THIS IS TllE LATEST VERSION. FAILURE TO DO S0 MAY INVALIDATE YOUR ANALYSI USER'S MANUAL - BWNT-TM-63 (ref 32-1203121-01 FSPLIT SA FSPLIT SA - FLOW SPLITS by DA FARNSWORTHon& JA WEIMl Check with the authors for the latest version. (

SUMMARY

of RESULTS { l From Input File B: CASE 4 with a Checksum of ' RECOVERABLE / FRICTION LOSS BENCHMARK 0.55 This is a Type 1 Case. Tha Total Flow is Specified. \ Both recoverable and unrecoverable losses are considered. The system pressure ( at node 1 ) is 1000 psia. There are 4 nodes connected by 3 internal flow paths. There is I external flow path. A local convergence criterion of 0.10000-05 was used with 0.95000-06 met . Overall Convergence was met. Solution required 25 local iterations.

 ~

FSPLIT SA - FLOW SPLITS by DA FARNSWORTH & JA WEIMER RECOVERABLE / FRICTION LOSS BENCHMARK

               ----- EXTERNAL PATH INFORMATION -----

85  !

I Ext Node Node Document Number 32-1203121-01 Path Flow Press Drop Path IN OUT Lbm/See Lbf/Sq inch Node IN Node OUT Press, psi Press, psi k k k j b$kbbbbb bk b$kb7b9b bh b$hhhh[b bh b5[b For External Path 1 the available NPSH is 0.949470+03 psia (200.0 F suction) FSPLIT SA - FLOW SPLITS by DA FARNSWORTH & JA WEIMER RECOVERABLE / FRICTION LOSS BENCHMARK

                         ----- INTERNAL PATH INFORMATION ---.-

Connecting Path Flow Path Node-> Node Press Drop Unrec DELTAP Lbm/Sec Lbf/Sq inch Recov OELTAP Lbf/Sq inch Lbf/Sq inch k [ k b.[bbbbbb+bkb((bkhbbb+bb 2 2 3 0.100000C-04 b bb7bhhb b[ b$kbbhbhbbh 3 3 4 0.139974D+02 0.148859D+02 0.100000D+04 0.729269D+01 - 888424D+00 0.169763D+01 0.559505D+01 FSPLIT SA - FLOW SPLITS by DA FARNSWORTH & JA WEIMER - L - REC 0VERABLE/ FRICTION LOSS BENCHMARK

                       ----- INTERNAL PATH INFORMATION -----

Connecting Unrecoverable DP Path Node-> Node Vel Head Friction Total Recoverable DP 1

                        . ... . . . . . . . . . . . . . . . . . ... . . .UR........
                                                                         . Momentum      El evat i on Total RE 1         2        6.9422                                                          ........

2 1.7364 8.6786 5.2066 2 3 6.9422 7.9437 5.6137 10.8203 3 .3 4 14.8859 -5.2066 4.3182

                                  -0.0000          1.6976         1.6976                             -0.8884 0.0540     5.5410       5.5951 FSPLIT 5A - FLOW SPLITS by DA FARNSWORTH & JA WEIMER - Ligh RECOVERABLE / FRICTION LOSS BENCHMARK
                     ----- INTERNAL PATH INFORMATION -----

Path Node Path Flow Input Heat In Out Velocity Path HTC Conductivity Lbm/Sec Btu /Sec Ft/Sec B/Hr-Ft^2-F B/Hr- Ft-F I ........... ........... 1 2 ........... 2 2 3 0.100000+04 0.000000+00 0.16082D+02 0.00000D+00 0 . 0.10000D+04 0.000000+00 0.321630+02 0.00000D+00 0.3 87

3 3 Document Number 32-1203121-01

                          .4
                                   'O.10000D+04 0.996980+05 0.16293D+02 0.00000D FSPLIT 5A - FLOW SPLITS by DA FARNSWORTH & JA WEIMER - L RECOVERABLE / FRICTION LOSS BENCHMARK
                                  ----- INTERNAL PATH INFORMATION -----

Path Node Vel Head Reynolds Friction In Out Lbf/Sq Inch Number Density Viscosity

      ............                 ...........                             Factor-            Lbm/Ft^3 Lbf-Sec/Ft^2 I       1       2                                                                                    ...........

2 2 3 0.17355D+01 0.246390+07 0.112890-01 0.62182D+0 3 3 4 0.69422D+01 0.34845D+07 0.11412D-01 0.62182D+0 0.17583D+01 0.38891D+07 0.108940-01 0.61378D+0 FSPLIT 5A - FLOW SPLITS by DA FARNSWORTH & JA WEIMER - Lig RECOVERABLE / FRICTION LOSS BENCHMARK

                                ----- INTERNAL PATH INFORMATION -----                                                       '

Path Node In Out Form Loss Path Temp Flow Area Roughness Deg F Ft^2 Hyd Dia Fric Len M inches Feet Feet 1 1- 2

     -2       2      3 3      3       4 0.10000+01 0.1000D+03 0.5000D+00 0.00000+00 0.1500D+03 0.10000+01 0.50000+03 0.112                                                  j FSPLIT 5A - FLOW SPLITS by DA FARNSWORTH & JA WEIMER - Ligh RECOVERABLE / FRICTION LOSS BENCHMARK
                           .......-- NODE INFORMATION ---------

Path. Inlet Rho Outlet Rho Inlet VH Outlet VH (Nodes) tbnVFt^3 Lbm/Ft^3 lbf/Sq In Lbf/Sg In Inlet P Outlet P PSI PSI 5 i b 0.62182D+02 2 2 3 33 4 b$bhibhb bh b$bhkbhb 0.62182D+02 0.69422D+01bh b$khhhhb b5 b 0.17355D+01 980.5011 966.5037 0.62182D+02 0.603050+02 0.17355D+01 0.17896D+01 966.5037- 959.2110 88 I 1

i Document Number 32-1203121-01 FSPLIT 5A - FLOW SPLITS by DA FARNSWORTH & JA WEIMER - Light Water Option RECOVERABLE / FRICTION LOSS BENCHMARK

                          ........- NODE INFORMATION ---------

Path Node' Elev In Elev Out Area In Area Out Temp In Temp Out In Out Feet Feet Ft^2 Ft^2 Deg F Deg F 1 1 2 0.0000D+00 0.1300D+02 0.1000D+01 0.5000D+00 0.1000D+03 0.1000D+03 2 2 3 0.1300D+02 0.2300D+02 0.50000+00 0.1000D+01 0.1000D+03 0.1000D+03 3 3 4 0.2300D+02 0.36000+02 0.1000D+01 0.10000+01 0.1000D+03 0.2000D+03 a FSPLIT 5A - FLOW SPLITS by DA FARNSWORTH & JA WEIMER - Light Water Option REC 0VERABLE/ FRICTION LOSS BENCHMARK

                          ----- PATH / NODE DESCRIPTION -----

Connecting Path Node-> Node Description I 1 2 FIRST LEG 2 2 3 SECOND LEG . 3 3 4 THIRD LEG FSPLIT 5A - FLOW SPLITS by DA FARNSWORTH & JA WEIMER - Light Water Option 3 1 1 REC 0VERABLE/ FRICTION LOSS BENCHMARK FIRST LEG 1 2 0.4000000+01 0.4000000+01 0.1000000+03 0.1000000+01 j 0.500000D+03 0.1128380+01 0.100000D+03 0.000000D+00 0.1300000+02 0.1000000+01 0.5000000+00 0.1000000+03 0.100000D+03 SECOND LEG 2 3 0.100000D+01 0.1000000+01 0.100000D+03 0.500000D+00 0.500000D+03 0.797885D+'O 0.8000000+02 0.1300000+02 0.2300000+02 0.500000D+00 0.100000D+01 0.1000000+03 0.1000000+03 . THIRD LEG i 3 4 0.0000000+00 0.000000D+00 0.1500000+03 0.1000000+01 0.500000D+03 0.112838D+01 0.1000000+03 0.2300000+02 0.360000D+02 0.1000000+01 0.100000D+01 0.1000000+03 0.2000000+03 4 1 1000 4 1 1000 0 1 8 28 18 6 32 36 16 26 58 44 14 46 82 52 12 66

      .                                               89 J

Document Number 32-1203121-01 33 33 33335 5 3 333 3 3 3 3333333333333 3 33 3 333 E 3333D3333332533333333335EE33333333 CASE SA-4A FSPLIT SA - FLOW SPLITS by DA FARNSWORTH & JA WEIMER for the generation and solution of steady state flo The networks can contain any combination of up to 100 nodes and 100 paths. The problem may be specified by imposing up to 10 boundary condition. flow boundary conditions (external paths) or on Problems may be defined with either temperature inputs or heat inputs to nodes and paths. FSPLIT greatly simplify model development.uses graphical on screen mo In addition to individual path flow rates, FSPLIT solves for momentum, elevation, friction, and form loss pressure drops. The form loss pressure drop calculations can assume a constant loss factor or can calculate a flow dependent loss factor. Equivalent heat rates are calculated for temperature input probisms and temperatures are calculated for heat input problems. External path input flows may be constant or s head-capacity relationship to simulate a pump.pecified as a FSPLIT is designed incompressibleto accommodate fluid, and gasses.water (H20), heavy water (D20), any routines for light water and heavy water while properties forFSPLIT gasses and incompressible fluids must be specified by the user. i Help Creationscreens Modes. are available in both the Node Creation and Path i create a model and complete the analysis.They should provide ade If you have problems 4 consult the user's manual. concerns, contact une of the authors.If the manual does not answer your , i FSPLIT was written in a combination of Microsoft QuickBAS{ FORTRAN. matrix solutions are performed using theExec- LINPACK All solve The FSPLIT5A.EXE, EXECHK.COM and FSPLIT. If FSPLIT is to be executed on a Hercules type monitor, the MSHERC.COM program must be run. 90 j

                                                                                          )

l Document Number 32-1203121-01 1 I PRIOR TO EXECUTING THIS CODE, CONTACT ONE OF THE AUTHORS TO ASSURE THAT THIS IS THE LATEST VERSION. FAILURE TO DO S0 MAY INVALIDATE YOUR ANALYSIS. USER'S MANUAL - BWNT-TM-63 (ref 32-1203121-01) FSPLIT SA FSPLIT SA - FLOW SPLITS by DA FARNSWORTH & JA WEINER - Light Water Option Check with the authors for the latest version. ,

SUMMARY

of RESULTS From Input File B: CASE 5A with a Checksum of 0.140605194D+04 CASE 5-4A -100K HEAD CAPACITY CHECK This is a Type 3 Case. The Total Flow is Specified, i Both recoverable and unrecoverable losses are considered. l The system pressure ( at node 4 ) is 100 psia, l l There are 4 nodes connected by 3 internal flow paths. There is 1 external flow path. A local convergence criterion of 0.1000D-05 was used with 0.8848D-06 met. External Pump head capacity was imposed. A head capacity criterion of 0.10000-03 was used with 0.7546D-04 met. Overall Convergence was met. Solutien required 4 head capacity iterations. FSPLIT SA - FLOW SPLITS by DA FARNSWORTH & JA WEIMEP. - Light Water Option CASE 5-4A -100K HEAD CAPACITY CHECK

                      ----- EXTEo"it PATH INFORMATION -----                               1 I

Ext Node Node Fath F.a Press Drop Node IN Node OUT Path IN OUT Lbm/Sec Lbf/Sg inch Press, psi Press, psi 1 4 1 0.14741D+03 0.376270+01 0.10000D+03 0.103760+03 1 For External Path 1 the avail.able NPSH is 0.99675D+02 psia ( 70.0 F suction) TABULAR PUMP The multiplier forHEAD pump/ ICAPACITY was 1 DATA WAS INPUT FROM FILE B: CASE 5.PMP. 91

Document Number 32-1203121-01 FSPLIT 5A - FLOW SPLITS by DA FARNSWORTHon& JA WEIM CASE 5-4A -100K HEAD CAPACITY CHECK

                             ----- INTERNAL PATH INFORMATION -----

Connecting Path Flow Path Node-> Node Press Drop Unrec DELTAP

                   .... ....                Lbm/Sec       Lbf/Sq inch       Lbf/Sq inch    Recov DELTAP 1
                                       ............      ............                        Lbf/Sq inch 1         2                                          ............

2 0.1474090+03 0.654765D-07 ............ 2 3 0.6547650-07 3 0.147409D+03 0.376270D+01 0.000000D+00 3 4 0.376270D+01 0.1474090+03 0.0000000+00 0.000000D+00 0.0000000+00 0.000000D+00 FSPLIT SA - FLOW SPLITS by OA FARNSWORTHon& JA WEIM CASE 5-4A -100K HEAD CAPACITY CHECK

                          ----- INTERNAL PATH INFORMATION -----

Connecting Unrecoverable DP Path Node-> Node Vel Head Friction Recoverable DP Total UR Momentum ElevationTotal RE k [ h bbbbb bbbbb 2 2 3 3.7627 b$bbbb b$bbbb b[bbbb 3 3 4 0.0000 3.7627 0.0000 b$bbbb 0.0000 0.0000 0.0000 0.0000 0.0000 0.0000 0.0000 0.0000 FSPLIT 5A - FLOW SPLITS by DA FARNSWORTH & JA WEIMER CASE 5-4A -100K HEAD CAPACITY CHECK

                        ----- INTERNAL PATH INFORMATION -----

Path Node Path Flow Input Heat Velocity In Out Lbm/Sec Path HTC Conductivity Btu /Sec Ft/Sec B/Hr-Ft^2-F B/Hr-Ft-F k k h 2 3 2 3 3 4 h [khkkb bh b$bbbbbb bb b$hh66hb bk b 0.14741D+03 0.000000+00 0.23652D+01 0.00000D+ 0.14741D+03 0.000000+00 0.236520+01 0.00000D+ FSPLIT SA - FLOW SPLITS by DA FARNSWORTH & JA WEIMER  ! CASE 5-4A -100K HEAD CAPACITY CHECK

                     ----- INTERNAL PATH INFORMATION -----

92 O

                                                                                                           )

i Path Node Vel Head Document Number 32-1203121-01 Reynolds In Out Lbf/Sq Inch Friction Density Number Factor Viscosity Lbm/Ft^3 Lbf-Sec/Ft^2 k k h 2 3 2 3 3 4 b$bhbbhb5bk b$hbbhbb bb b$bbbbb 0.37627D-01 0.253760+06 0.000000+00 0.623 0.376270-01 0.25376D+06 0.000000+00 0.623

       'FSPLIT SA - FLOW SPLITS by DA FARNSWORTHn & JA WEI j

CASE 5-4A -100K HEAD CAPACITY CHECK

                                 ----- INTERNAL PATH INFORMATION -----

Path Node In Out Form Loss Path Temp Flow Area Roughness Deg F Hyd Dia Ft^2 M inches Fric Len Feet Feet 1 1 2 2 2 3 3 3 4 0.1000D+03 0.70000+02 0.1000D 0.0000D+00 0.70000+02 0.1000D+01 0.00000+0 FSPLIT SA - FLOW SPLITS by DA FARNSWORTHon& JA WEIM CASE 5-4A -100K HEAD CAPACITY CHECK

                               ......--- NODE INFORMATION ---------

Path Inlet Rho Outlet Rho (Nodes) lbm/Ft^3 Inlet VH Outlet VH lbm/Ft^3

                        ........... ........... ..../Sq In Lbf/Sq In lbf                         Inlet P Outlet P
                                                                                     ~ PSI         PSI 1 V 2                                                  ....... ...........   ........   ........

2 2 3 0.62323D+02 0.62323D+02 0.376270-01 0.37627D-01 103.7627 103.7627 3 3 4 0.62323D+02 0.62323D+02 0.376270-01103.7627 0.37627D-01 100.0000 0.62323D+02 0.623230+02 0.376270-01100.0000 0.376270-01 100.0000

 .                                                                                                         l FSPLIT SA - FLOW SPLITS by DA FARNSWORTH & JA WEIMER j

CASE 5-4A -100K HEAD CAPACITY CHECK i

                            -------.. NODE INFORMATION ---------

Path Node Elev In ) Elev Out Area In In Out Feet Feet Area Out Temp In Temp Out i Ft^2 Ft^2 Deg F Deg F 93^

                                                                                       \

l Document Number 32-1203121-01 I 1 1 2 2 2 3 0.0000D+00 0.00000+00 0.1000D+01 0.10000+01 0.70000+02 1 3 3 4 0.0000D+00 0.0000D+00 0.10000+01 0.1000D+01 0.7000D+02 0.00000+00 0.00000+00 0.10000+01 0.10000+01 0.70000+02 0.7 j FSPLIT 5A - FLOW SPLITS by DA FARNSWORTH & JA WEINER - Light Water Optio CASE 5-4A -100K HEAD CAPACITY CHECK

                        ----- PATH / NODE DESCRIPTION -----

Connecting Path Node-> Node Description I 1 2 PUMP DISCHARGE 2 2 3 l RESISTANCE PATH

3. 3 4 PUMP SUCTION i

l FSPLIT 5A - FLOW SPLITS by DA FARNSWORTH & JA WEIMER - Light Water Opti 3 1 3 CASE 5-4A -100K HEAD CAPACITY CHECK PUMP DISCHARGE 1 2 0.000000D+00 0.000000D+00 0.700000D+02 0.1000000+01 0.000000D+00 0.1128380+01 0.0000000+00 0.000000D+00 0.000000D+00 0.1000000+01 0.100000D+01 0.700000D+ RESISTANCE PATH l 2 1 3 0.100000D+03 0.100000D+03 0.7000000+02 0.1000000+01 0.000000D+00 0.1128380+01 0.0000000+00 0.000000D+00 0.000000D+00 0.100000D+01 0.100000D+01 0.700000D+ PUMP SUCTION 3 4 0.000000D+00 0.0000000+00 0.700000D+02 0.1000000+01 0.000000D+00 0.1128380+01 0.0000000+00 0.000000D+00 4 1 0.000000D+00 0.100000D+01 0.100000D+01 0.700000D+02 162.6668039959744 4 1 100 0 4 8 52 12 6 28 52 12 22 28 36 16 22 8 36 16  ; 6 - ' l 3 3333 3 3 333338533335 33333535 335tSE58 33385 5353333 333 3333 33 333333333333333333 333 CASE 5B-4A 94

i Document Number 32-1203121-01 FSPLIT SA - FLOW SPLITS by DA FARNSWORTH & JA WEIMER ( for the generation and solution of steady state fl ' The networks 100 paths. can contain The problem ma any combination of up to 100 nodes and flow boundary conditions (y be specified by imposing up to 10 boundary condition. Problems may be defined with eitherexternal pa temperature inputs or heat inputs to nodes and paths.FSPLIT greatly simplify model development.uses graphical on screen mJi In addition to momentum, individual elevation path flow rates, FSPLIT solves for friction and form loss pressure drops. The form loss pressu,re drop ca,lculations can assume a constant loss factor or can calculate a flow dependent loss factor. Equivalent heat rates are calculated for temperature input problems and temperatures are calcula+ed for heat input problems. External path input flows may be connant or s head-capacity relationship to simulate a pump.pecifiedFSPLITas is a designed incompressibleto accommodate fluid, and gasses.water (H20), heavy water (D20), any routines for light water and heavy water while properties forFSPLI I gasses and incompressible fluids must be specified by the user. Creation Modes. Help screens are available in both the Node Creation and create a model and complete the analysis.They should provide ad If you have problas consult the user's manual. concerns, contact one of the authors.If the manual does not answer your , FSPLIT was written in a combination of Microsoft QuickBA FORTPAN. matrix solutions are performed using the LINPACK All Exec-solve The FSPLITSA.EXE, EXECHK.COM and FSPLIT If FSPLI-T is to be executed on a Hercules type monitor, the MShERC.COM program must be run. PRIOR TO EXECUTING THIS CODE, CONTACT ONE OF THE AUTHORS TO ASSURE THAT THIS IS THE LATEST VERSION. FAILURE TO 00 S0 MAY INVALIDATE YOUR ANALYSIS. USER'S MANUAL - SA FSPLIT BWNT-TM-63 (ref 32-1203121-01) l k 95 1

1 l Document Number 32-1203121-01 FSPLIT SA - FLOW SPLITS by DA FARNSWORTH & n JA WEIM Check with the authors for the latest version.

SUMMARY

of RESULTS From Input File B: CASE 5B with a Checksum of 0.1466051940+04 CASE 5-4A-80K HEAD CAPACITY CHECK This is a Type 3 Case. The Total Flow is Specified. Both recoverable and unrecoverable losses are considered. The system pressure at node 4 ) is 100 psia. There are 4 nodes con (nected by 3 internal flow paths. There is 1 external flow path. A local convergence criterion of 0.10000-05 was used with 0.74050-06 met External Pump head capacity was imposed. . A head capacity criterion of 0.10000-03 was used with 0.70690-04 met . Overall Convergence was met. Solution required 3 head capacity iterations. FSPLIT SA - FLOW SPLITS by DA FARNSWORTH &onJA WEIMER CASE 5-4A-80K HEAD CAPACITY CHECK

               ----- EXTERNAL PATH INFORMATION -----

Ext Node Node Path Flow Path IN OUT Press Drop Node IN Lbm/Sec Lbf/Sq inch Node OUT Press, psi Press, psi k k k b$khhh4b bh b hhb48b bk b$5bbbbb bh b kb For External Path 1 the available NPSH is 0.996830+02 psia ( 70 0 F suction) TABULAR The multiplier for PUMP pump I HEAD was 1 / CAPACITY DATA WAS INPUT FRO FSPLIT SA - FLOW SPLITS by DA FARNSWORTH & JA WEIMER - L CASE 5-4A-80K HEAD CAPACITY CHECK

             ----- INTERNAL PATH INFORMATION -----

Connecting Path Flow Path Node-> Node Press Drop Unrec DELTAP Lbm/Sec Lbf/Sq inch Recov DELTAP

.... .... ....         ............                     Lbf/Sq inch       Lbf/Sq inch 96                                                      I

l Document Number 32-1203121-01 - 1 1 2 0.163536D+03 0.6716520-07 0.6716520-07 0.000000D+00 i 2 2 3 0.163536D+03 0.370485D+01 0.370485D+01 0.000000D+00 I 3 3 4 0.1635360+03 0.0000000100 0.000000D+00 0.000000D+00 FSPLIT 5A - FLOW SPLITS by DA FARNSWORTH & JA WEIMER - Light Water Option CASE 5-4A-80K HEAD CAPACITY CHECX

                       ----- INTERNAL PATH-INFORMATION -----                                            I Connecting              Unrecoverable DP                  Recoverable DP Path Node-> Node Vel Head Friction Total UR Momentum Elevation Total RE k         [       2      bbbbb      bbbbb        b$bbbb    b$bbbb     b$bbbb          b$bbbb 2         2       3      3.7048     0.0000       3.7048    0.0000     0.0000          0.0000 3         3       4      0.0000     0.0000       0.0000    0.0000     0.0000          0.0000 FSPLIT SA - FLOW SPLITS by DA FARNSWORTH & JA WEIMER - Light Water Option CASE 5-4A-80K HEAD CAPACITY CHECK
                      ----- INTERNAL PATH INFORMATION -----

Path Node Path Flow Input Heat Velocity Path HTC Conductivity In Out Lbm/Sec Btu /Sec Ft/Sec 8/Hr-Ft^2-F B/Hr-Ft-F

    ............ ........... ........... ........... ...........                    ...........          I 1     1     2     0.16354D+03 0.000000+00 0.26240D+01 0.000000+00 0.34749D+00                      l 2     2     3     0.16354D+03 0.000000+00 0.26240D+01 0.00000D+00 0.34749D+00 l

3 3 4 0.16354D+03 0.00000D+00 0.26240D+01 0.000000+00 0.347490+00 FSPLIT SA - FLOW SPLITS by DA FARNSWORTH & JA WEIMER - Light Water Option  ! CASE 5-4A-80K HEAD CAPACITY CHECK

                      ----- INTERNAL PATH INFORMATION -----

Path Node Vel Head Rey'nolds Friction Density Viscosity In Out Lbf/Sq Inch Number Factor Lbm/Ft^3 Lbf-Sec/ft^2

   ............ ........... ........... ........... ........... ...........                              l 1     1     2    0.463110-01 0.28152D+06 0.00000D+00 0.623230+02 0.20373D-04 2     2     3    0.463110-01 0.28152D+06 0.000000+00 0.62323D+02 0.20373D-04 3     3     4    0.463110-01 0.28152D+06 0.00000D+00 0.62323D+02 0.20373D-04
   '                                                97

Document Number 32-1203121-01 FSPLIT SA - FLOW SPLITS by DA FARNSWORTH & JA WEIMER - CASE 5-4A-80K HEAD CAPACITY CHECK

                       ----- INTERNAL PATH INFORMATION -----

Path Node In Out Form Loss Path Temp Flow Area Roughntss Deg F Hyd Dia Fric Len Ft^2 M inches Feet Feet 1 1 2 2 2 3 3 3 4 0.8000D+02 0.7000D+02 0.10000+0 0.0000D+00 0.7000D+02 0.10000+01 0.00000+00 0.1 I FSPLIT SA - FLOW SPLITS by DA FARNSWORTH & JA WEIMER - Li CASE 5-4A-80K HEAD CAPACITY CHECK

                      --------- N0DE INFORMATION ---------

Path Inlet Rho Outlet Rho Inlet VH Outlet VH (Nodes) Lbm/Ft^3 Lbm/Ft^3 Inlet P Outlet P

 ........        ........... ........... ..../Sg lbf In    Lbf/Sg In       PSI        PSI I     1 2                                     ....... .......,...  ........   ........

2 2 3 0.62323D+02 0.62323D+02 0.463110-01 0.463110-01 103.7048 103.7048 3 3 4 0.62323D+02 0.62323D+02 0.463110-01 0.46311D-01 103.7048 100.0000 0.62323D+02 0.62323D+02 0.463110-01 0.463110-01 100.0000 100.0000 {

                                                                                              \
'FSPLIT 5A - FLOW SPLITS by DA FARNSWORTH & JA WEIMER - Lighti                               I CASE 5-4A-80K HEAD CAPACITY CHECK
                    --------- NODE INFORMATION ---------

-Path Node Elev In Clev Out Area In Area Out In Out Feet Feet Ft^2 Temp In Temp Out Ft^2 Deg F Deg F 1 1 2 2 2 3 3 3 4 0.00000+00 0.00000+00 0.10000+01 0. 0.0000D+00 0.00000+00 0.10000+01 0.10000+01 0.7000D+ FSPLIT SA - FLOW SPLITS by DA FARNSWORTH & JA WEIMER - Light W CASE 5-4A-80K HEAD CAPACITY CHECK 98

1 1

                                                                                }

{ Document Number 32-1203121-01

                   ----- PATH / NODE DESCRIPTION -----

Connecting Path Node-> Node Description 1 .~ 2 PUMP DISCHARGE 2 2 3 RESISTANCE PATH 3 3 4 PUMP SUCTION FSPLIT SA - FLOW SPLITS by DA FARNSWORTH & JA WEIMER - Light Water Opti 3 1 3 CASE 5-4A-80K HEAD CAPACITY CHECK PUMP DISCHARGE 1 2 0.000000D+00 0.0000000+00 0.7000000+02 0.100000D+01 0.0000000+00 0.1128380+01 0.0000000+00 0.0000000+00 0.000000D+00 0.1000000+01 0.100000D+01 0.7000000+ RESISTANCE PATH 2-3 0.800000D+02 0.8000000+02 0.700000D+02 0.100000D+01 0.000000D+00 0.1128380+01 0.000000D+00 0.0000000+00 0.000000D+00 0.1000000+01 0.1000000+01 0.700000D+ PUMP SUCTION 3 4 0.000000D+00 0.0000000+00 0.7000000+02 0.100000D+01 0.000000D+00 0.1128380+01 0.000000D+00 0.000000D+00 4 1 0.000000D+00 0.100000D+01 0.1000000+01 0.700000D+ 162.6668039959744 4 1 100 0 4 8 52 12 6 28 52 12 22

28. 36 16 8 22 36 16 6

............................................................................ i 1 i CASE SC-4A FSPLIT SA - FLOW SPLITS by DA FARNSWORTH.& JA WEIMER FSPLIT is a PC based thermal hydraulics computer code designed for the generation and solution of steady state flow networks. The networks can contain any combination of up to 100 nodes and 100 paths. The problem may be specified by imposing up to 10 flov boundary boundary conditions (external paths) or one pressure drop condition. Problems may be defined with either 99

Document Number 32-1203121-01 temperature inputs or heat inputs to nodes and paths. FSPLIT uses graphical on screen modeling and parameter specification to greatly simplify model development. In addition to individual path flow rates, FSPLIT solves for momentum, elevation, friction, and form loss pressure drops. The form loss pressure drop calculations can assume a constant loss factor or can calculate a flow dependent loss factor. Equivalent heat rates are calculated for temperature input problems and temperatures are calculated for heat input problems. External path input flows may be constant or specified as a head-capacity relationship to simulate a pump. FSPLIT is designed to accommodate water (H2O), heavy water (D20), any incompressible fluid, and gasses. FSPLIT has internkl property routines for light water and heavy water while properties for gasses and incompressible fluids must be specified by the user. Help screens are available in both the Node Creation and Path Creation Modes. They should provide adequate knowledge to create a model and complete the analysis. If you have problems, consult the user's manual. If the manual does not answer your concerns, contact one of the authors. FSPLIT was written in a combination of Microsoft QuickBASIC and FORTRAN. All calculations are in double (64 bit) precision. All matrix solutions are performed using the LINPACK solvers. Exec-ution of FSPLIT requires roughly 550 kilobytes of free DOS memory. The FSPLIT5A.EXE, EXECHK.COM and FSPLIT.HLP files must be present. If FSPLIT is to be executed on a-Hercules type monitor, the MSHERC.COM program must be run. PRIOR TO EXECUTING THIS CODE, CONTACT ONE OF THE AUTHORS TO ASSURE THAT THIS IS THE LATEST VERSION. FAILURE TO DO S0 MAY INVALIDATE YOUR ANALYSIS. USER'S MANUAL - BWNT-TM-63 (ref 32-1203121-01) FSPLIT 5A FSPLIT SA - FLOW SPLITS by DA FARNSWORTH & JA WEIMER - Light Water Option Check with the authors for the latest version.

SUMMARY

of RESULTS From Input File B:CASESC with a Checksum of 0.1410012010+04 CASE 5-4A-60K HEAD CAPACITY CHECK This is a Type 3 Case. The Total Flow is Specified. Both recoverable and unrecoverable losses are considered.

   ,                                   100

Document Number 32-1203121-01 The system pressure ( at node 4 ) is 100 psia'. There are 4 nodes connected by 3 internal flow paths. LThere is 1 external flow path. A local convergence criterion of 0.10000-05 was used with 0.8851D-06 met. External Pump head capacity was imposed. A head capacity criterion of 0.10000-03 was used with 0.37790-04 met. Overall Convergence was met. Solution required 7 head capacity iterations. FSPLIT SA - FLOW SPLITS by DA FARNSWORTH & JA WEIMER - Light Water Option CASE 5-4A-60K HEAD CAPACITY CHECK

                    ----- EXTERNAL PATH INFORMATION -----

Ext Node Node Path Flow Press Drop Node IN Node OUT Path IN OUT Lbm/Sec Lbf/Sq inch Press, psi Press, psi k k [ b$kbhbhb bh b$hhhb7b hk b$[bbbbb bh b$kbbhbb bh For External Path 1 the available NPSH is 0.99696D+02 psia ( 70.0 F suction) TABULAR PUMP HEAD / CAPACITY DATA WAS INPUT FROM FILE B: CASE 5.PMP. The multiplier for pump I was 1 FSPLIT SA - FLOW SPLITS by DA FARNSWORTH & JA WEIMER - Light Water Option CASE 5-4A-60K HEAD CAPACITY' CHECK l I

                   ----- INTERNAL PATH INFORMATION ------                                   '

1 Connecting Path Flow Press Drop Unrec DELTAP Recov DELTAP Path Node->Nede Lbm/Sec Lbf/Sq inch Lbf/Sq inch Lbf/Sq inch j i 1 h b.1hhbkhb+b5 b lbkhlbb-06 h kbhh[hb bh b bbbbbb bb 2 2 3 0.185022D+03 0.355674D+01 0.355674D+01 0.000000D+00 ; 3 3 4 0.185022D+03 0.000000D+00 0.00000JD+00 0.0000000+00 l

 . FSPLIT SA - FLOW SPLITS by DA FARNSWORTH & JA WEIMER - Light Water Option CASE 5-4A-60K HEAD CAPACITY CHECK                            i l

101

Document Number 32-1203121-01

                         ----- INTERNAL PATH INFORMATION -----

Connecting Unrecoverable DP Recoverable DP Path ' Node-> Node Vel Head Friction Total UR Momentum Elevation Total RE [ k h b$bbbb b$bbbb b$bbbb b$bbbb b$bbbb b$bbbb - 2 2 3 3.5567 0.0000 3.5567 0.0000 0.0000 0.0000 3 3 4 0.0000 0.0000 0.0000 0.0000 0.0000 0.0000 FSPLIT5A-FLOWSPLITSbyDAFARNSWdRTH&JAWEIMER-LightWaterOption CASE 5-4A-60K llEAD CAPACITY CHECK 4

                         ----- INTERNAL PATH INFORMATION -----

Path Node Path Flow Input Heat Velocity Path HTC Conductivity In Out, Lbm/Sec Btu /Sec Ft/Sec B/Hr-Ft^2-F B/Hr-Ft-F k k b b$kbhbhb bb b$bbbbbb bb b$hbbbhb bk b$bbbbbb bb b$bkhkbb 2 2 3 0.18502D+03 0.000000+00 0.296870+01 0.000000+00 0.347490+00 3 3 4 0.18502D+03 0.00000D+00 0.29687D+01 0.00000D+00 0.347490+00 FSPLIT 5A - FLOW SPLITS by DA FARNSWORTH & JA WE1MER - Light Water Option

                                                                                                   )

CASE 5-4A-60K HEAD CAPACITY CHECK I

                        ----- INTERNAL PATH INFORMATION -----

Path ^ Node Vel Head Reynolds Friction Density Viscosity In" Out Lbf/Sq Inch Number Factor Lbm/Ft^3 ~Lbf-Sec/Ft^2 k k h b$hbbhbb bk b hkbh[b bb b bbbbbb bb b bhhhhb bh b$hbbhhb bk 2 2 3 0.592790-01 0.31851D+06 0.00000D+00 0.62323D+02 0.20373D-04 3 3 4 0.59279D-01 0.31851D+06 0.000000+00 0.62323D+02 0.203730-04 FSPLIT 5A - FLOW SPLITS by CA FARNSWORTH & JA WEIMER - Light Water Option CASE 5-4A-60K HEAD CAPACITY CHECK

                        ----- INTERNAL PATH INFORMATION -----

Path Node . Form Loss Path Temp Flow Area Roughness Hyd Dia Fric Len ) 102

In Out Document Number 32-1203121-01 Deg F Ft^2 M inches Feet 1 1 2 Feet 2 2 3 3 3 -4 0.6000D+02 0.70000+02 0.1000 0.0000D+00 0.70000+02 0.10000+01 0.0000D FSPLIT SA - FLOW SPLITS by DA FARNSWORTH p on & JA WEI CASE 5-4A-60K HEAD CAPACITY CHECK

                               ------ -- NODE INFORMATION ---------

Path Inlet Rho Outlet Rho (Nodes) .Lbm/Ft^3 Inlet VH Outlet VH

      ........                                     Lbm/ft^3                           Lbf/Sq In        Inlet P Outlet P
                        . . . . . . . . . . . . . . . . . . . . . . . .l b f/ S q I n                    PSI 11       2                                                        ......... ...........     ........

PSI 2 2 3 0.62323D+02 0.623230+02 0.592790-01 103.55670.59279D-01 103.5567 3 3 4 0.62323D+02 0.62323D+02 0.592790-01 103.55670.592790-01 100.0000 0.623230+02 0.62323D+02 0.592790-01 0.592790-01 100.0000 100.0000 FSPLIT SA - FLOW SPLITS by DA FARNSWORTH on

                                                                                                                         & JA WEIM CASE 5-4A-60K HEAD CAPACITY CHECK
                            --------- NODE Ih70RMATION ---------

P'th a Node Elev In Elev Out In Out Feet Area In Area Out Temp In Feet Ft^2 Temp Out Ft^2 Deg F 1 1. 2 Deg F-2 2 3 3 3 4 0.00000+00 0.0000D+00 0.1000D 0.00000+00 0.0000D+00 0.10000+01 0.1000D+0 FSPLIT 5A - FLOW SPLITS by DA FARNSWORTH on

                                                                                                                      & JA WEIM CASE 5-4A-60K HEAD CAPACITY CHECX                                                     .

Connecting ----- PATH /N00E DESCRIPTION ----- Path Node-> Node Description I 1 2 2 PUMP DISCHARGE 2 3 3 RESISTANCE PATH 3 4 PUMP SUCTION 103

Document Nuraber 32-1203121-01 FSPLIT SA - FLOW SPLITS by DA FARNSWORTH & JA WEIME 3 1 3 CASE

  ' PUMP DISCHARGE       5-4A-60K HEAD CAPACITY CHECK 1

0.0000000+00 0.112838D+01 0.000000D+002 0.00000 0.0000000+00 RESISTANCE PATH 2 0.0000000+00 0.100000D+01 0.1000 . 0.0000000+00 0.1128380+01 0.000000D+003 0.60000 0.0000000+00 PUMP SUCTION 3 0.000000D+00 0.100000D+01 0.1000 . 0.0000000+00 0.1128380+01 0.0000000+004 0.000000 0.0000000+00 0.000000D+00 4 4 1 1 146.6268653109858 0.1000000+01 0.10000

                                                                             .   +

8 100 0 52 4 28 12 6 52 12 28 36 22 16 8 36 22 16 6 CASE SD-4A FSPLIT 5A - FLOW SPLITS by DA FARNSWORTH & JA WEIMER FSPLIT is a PC based thermal hydraulics computer code designed for the generation and solution of steady state flow networks. Thepaths. 100 networks The can contain problem ma any combination of up to 100 nodes and flow boundary conditions (y be specified by imposing up to 10 boundary condition. external paths) or one pressure drop Problems may be defined with either temperature inputs or heat inputs to nodes and paths. FSPLIT greatly simplify model development.uses graphical on screen mod In addition to individual path flow rates, FSPLIT solves for momentum, elevation, friction, and form loss pressure loss factor or can calculate a flow dependent loss factor. Equivalent heat rates are calculated for temperature input External path input flows may be constant 'or specifie 104

Document Number 32-1203121-01 head-capacity relationship to simulate a pump. FSPLIT is

       -designed to accommodate water (R M), heavy water (D20), any incompressible fluid, and gasse. . FSPLIT has internal property routines for light water and heavy water while properties for gasses and incompressible fluids must be specified by the user.

Help screens are available in both the Node Creation and Path Creation Modes. They should provide adequate knowledge to create 4 model and complete the analysis. If yuu have problems consuh the user's manual. If the manual does not answer your , concern:, contact orie of the authors.

                                         ~

FSPLIT was written in a combination of Microsoft QuickBASIC and FORTRAN. All calculations are in double (64 bit) precision. All matrix solutions are performed using the LINPACK solvers. Exec-ution of FSPLIT requires roughly 550 kilobytes of free DOS memory. The FSPLITSA.EXE, EXECHK.COM and FSPLIT.HLP files must be present. If FSPLIT is to be executed on a Hercules type monitor, the MSHERC.COM program must be run. PRIOR TO EXECUTING THIS CODE, CONTACT ONE OF THE AUTHORS TO ASSURE THAT THIS IS THE LATEST VERSION. FAILURE TO DO S0 MAY INVALIDATE YOUR ANALYSIS. USER'S MANUAL - BWNT-TM-63 (ref 32-1203I21-01) FSPLIT SA . FSPLIT SA - FLOW SPLITS by DA FARNSWORTH & JA WEIMER - Light Water Option Check with the authors for the latest version.

SUMMARY

of RESULTS From Input File B: CASE 5D with a Checksum of 0.140742203D+04 CASE 5-4A-40K HEAD CAPACITY CHECK This is a Type 3 Case. The Total Flow is Specified. Both recoverable and unrecoverable losses are considered. The system pressure { at node 4 ) is 100 psia. There are 4 nodes connected by 3 internal flow paths. There is I external flow path. { A local convergence criterion of 0.10000-05 was used with 0.8299D.06 met. External Pump head capacity was imposed. i A head capacity criterion of 0.10000-03 was used with 0.8295D-04 met. Overall Convergence was met. Solution required 8 head capacity itrettions. 105 b

F i I Document Number 32-1203121-01 FSPLIT 5A - FLOW SPLITS by DA FARNSWORTH & JA WEIMER - Light Wat CASE 5-4A-40K HEAD CAPACITY CHECK

                          ----- EXTERNAL PATH INFORMATION -----

Ext Node Node Path Flow Press Drop Path IN OUT Node IN Node OUT Lbm/Sec Lbf/Sq inch Press, psi Press, psi 1 4 1 0.21752D+03 0.32773D+01 0.100000+03 0.103280+03 For External Path I the available NPSH is 0.997190+02 psia ( 70.0 F suction) TABULAR The multiplierPUMP for pump HEAD I was 1/ CAPACITY DATA WAS IM:UT FROM FILE FSPLIT 5A FLOW SPLITS by DA FARNSWORTH & JA WEIMER - Light Water Option CASE 5-4A-40K HEAD CAPACITY CHECK

            .           ----- INTERNAL PATH INFORMATION -----

Connecting Path Flow Press Drop Path Node-> Node Unrec DELTAP Recov DELTAP Lbm/Sec Lbf/Sq inch Lbf/Sq inch Lbf/Sq (nch 1 1 2 0.Z17520D+03 2

                                                      .130646D-06       .1306460-06 2        3     0.2175200+03                                        0.000000D+00 3                                        0.3277260+01      0.3277260+01 3        4      0.2175200+03                                        0.0000000+00 0.000000D+00      0.000000D+00    0.0000000+00  s FSPLIT 5A - FLOW SPLITS by DA FARNSWORTH & JA WEIMER - Light Water O CASE 5-4A-40K HEAD CAPACITY CHECK
                      -.--- INTERNAL PATH INFORMATION -----

Connecting Unrecoverable DP Recoverable DP Path Node-> Node Vel Head Friction Total UR Momentum Elevation jT k k 2 bbbbb bbbbb 2 2 3 b bbbb b bbbb b bbbb $b bbbb 3.2773 0.0000 3.2773 0.0000 3 3 4 0.0000 0.0000 0.0000 0.0000 0.0000 0.0000 0.0000 0.0000 FSPLIT SA - FLOW SPLITS by DA FARNSWORTH & JA WEIMER - Light Water Optio 106 I I

l l Document Number 32-1203121-01 CASE 5-4A-40K HEAD CAPACITY CHECK

                            ----- INTERNAL PATH INFORMATION -----

Path Node Path Flow Input Heat Velt ity Path HTC Conductivity

            .In Out              Lbm/Sec         8tu/Sec        Ft/Sec
                                                                                                   }

B/Hr-Ft^2-F B/Hr-Ft-F i i b l 2 2 3 b$bih5bbhbb b$bbbbbb bb b$b4bbhb bi b$bbbbbb 3 3 4 0.21752D+03 0.00000D+00 0.34902D+01 0.00000D+00 0.347490+00 0.21752D+03 0.00000D+00 0.34902D+01 0.00000D+00 0.347490+00 FSPLIT SA - FLOW SPLITS by DA FARNSWORTH & JA WEIMER - Light Water Optica CASE 5-4A-40K HEAD CAPACITY CHECK

                           ----- INTERNAL PATH INFORMATION -----

Path Node Vel Head Reynolds In Out Lbf/Sq Inch Friction Density Viscosity l i Number Factor Lbm/Ft^3 lbf-Sec/Ft^2 ) I 1 2 0.81932D-01 2 2 3 0.37445D+06 0.00000D+00 0.62323D+02 0.203730-04 3 3 4 0.819320-01 0.37445D+06 0.000000+00 0.62323D+02 0.20373D-04 0.81932D-01 0.37445D+06 0.00000D+00 0.62323D+02 0.203730-04 FSPLIT SA - FLOW SPLITS by DA FARNSWORTH & JA WEIMER - Light Water Option CASE 5-4A-40K HEAD CAPACITY CHECK

                         ----- INTERNAL PATH INFORMATION -----

Path Node Form Loss Path Temp Flow Area Roughness Hyd Dia Fric Len In Out Deg F Ft^2 M inches Feet Feet 1 1 2 2 2 0.0000D+00 0.7000D+02 0.10000+01 0.00000+00 0.1128D+01 0.0000D+00 3 3 3 4 0.40000+02 0.7000D+02 0.1000D+01 0.0000D+00 0.11280+01 0.0000D+00 0.0000D+00 0.70000+02 0.1000D+01 0.0000D+00 0.11280+01 0.00000+00 FSPLIT SA - FLOW SPLITS.by DA FARNSWORTH & 'A WEIMER - Light Water Opti,n CASE 5-4A-40K HEAD CAPACITY CHECK

                        ......... NODE INFORMATION ---------

l 107 j 1

Document Number 32-1203121-01 Path Inlet Rho Outlet Rho Inlet VH Outlet VH Inlet P Outlet P (Nodes) Lbm/Ft^3 Lbm/Ft^3 Lbf/Sq In Lbf

    ........        ...........     ........... ........... ..../Sq In PSI PSI I 1 230.62323D+02 0.62323D+02 0.81932D-01 0.81932D-01 22                                                                      103.2773 103.2773 3 3 4          0.62323D+02       0.62323D+02    0.81932D-01   0.81932D-01 103.2773 100.0000 0.623230+02 0.623230+02 0.81932D-01 0.81932D-01          100.0000 100.0000 FSPLIT 5A'. FLOW SPLITS by DA FARNSWORTH & JA WEIMER - Light Water Option CASE 5-4A-40K HEAD CAPACITY CHECK
                        --------- N0DE INFORMATION ---------

Path flode Elev In Elev Out Area In Area Out Temp In In Out Feet feet Ft^2 Ft^2 Temp Out Deg F Deg F 1 1 2 2 2 3 0.0000D+00 0.0000D+00 0.10000,01 0.10000+01 0.7000D+02 0.7000D+02 3 3 4 0.00000+00 0.00000+00 0:1000D+01 0.10000+01 0.70000+02 0.7000 0.0000D+00 0.0000D+00 0.10000+01 0.1000D+01 0.7000D+02 0.700

                                    ~

FSPLIT 5A - FLOW SPLITS by DA FARNSWORTH & JA WEIMER - Light Water Option CASE 5-4A-40K HEAD CAPACITY CHECK

                       ----- PATH /N0DE DESCRIPTION -----

Connecting Path Node-> Node Description I 1 2 PUMP DISCHARGE 2 2 3 RESISTANCE PATH 3 3 4 PUMP SUCTION FSPLIT 5A - FLOW SPLITS by DA FARNSWORTH & JA WEIMER - Light Water Option 3 1 3 CASE 5-4A-40K HEAD CAPACITY CHECK PUMP DISCHARGE

  • 1 2 0.0000000+00 0.000000D+00 0.7000000+02 0.100000D+01 0.000000D+00 0.1128380+01 0.000000D+00 0.0000000+00 RESISTANCE PATH 0.0000000+00 0.100000D+01 0.100000D+01 0.7000000+02 0.700 2

0.000000D+003 0.400000D+02 0.1128380+01 0.4000000+02 0.;300000+000.7000000+02 0.1000000+01 0.000000D+00 0.000000D+00 0.100000D+01 0.1000000+01 0.700000D+02 0.700 108

PUMP SUCTION Document Number 32-1203121-01 3 0.000000D+00 0.1128380+01 0.0000000+004 0.0000 0.000000D+00 0.0000000+00 4 4 1 1 184.0368941579841 100 0.1000000+01 0.100

                                                                                  +

8 52 0 4 28 12 6 52 12 28 36 22 8 16 22 36 16 6 CASE 6-4A FSPLIT SA - FLOW SPLITS by DA FARNSWORTH & JA WEIMER FSPLIT is a PC cased thermal hydraulics computer code designed for the generadon and solution of steady state flow networks The networks 100 paths. can contain The problem ma any combination of up to 100 nodes and flow boundary conditions (y be specified by imposing up to 10 . boundary condition. Problems may be defined with eitherexternel path temperature inputs or heat inputs to nodes and paths. FSPLU greatly simplify model development.uses graphical on screen m In additionelevation momentum, to individual path flow rates, FSPliT solves for friction, and form loss pressure drops. The form loss pressu,re drop calculations can assume a constant loss factor or can calculate a flow dependent loss factor. Equivalent heat rates are calculated for temperature input problems and temperatures are calculated for heat input problems. External path input flows may be constant or s head-capacity relationship to simulate a pump.pecified as a FSPLIT is designed incompressible to accommodate fluid, and gasses. water (H2O), heavy water (D20), any routines' for light water and heavy water while properties forFSPLIT gasse.s and incompressible fluids must be specified by the user. Help screens Creation Modes. are available in both the Node Creation and Pd h create a model and complete the analysis.They should provide adeq If you have problems consult the user's manual. concerns, contact one of the authors.If the manual does not answer your , 109

Document Number 32-1203121-01 FSPLIT was written in a combination of Microsoft QuickBA FORTRAN. matrix solutions are performed using the Exec- LINPACK All solv The FSPLIT5A.EXE, EXECHK.COM and FSPLIT If FSPLIT is to be executed on_a Hercules type monitor, the MSHERC.COM program must be run. PRIOR TO EXECUTING THIS CODE, CONTACT ONE OF THE AUTHORS TO ASSURE THAT THIS IS THE LATEST VERSION. FAILURE TO DO S0 MAY INVALIDATE YOUR ANALYSIS. USER'S MANUAL - BWNT-TM-63 (ref 32-1203121-01) FSPLIT SA FSPLIT SA - FLOW SPLITS by DA FARNSWORTH & JA WEIMER - L Check with the authors for the latest version.

SUMMARY

of RESULTS From Input File B: CASE 6 with a Checksum of 0.4677366700+04 CASE 6-4A HEAT INPUT This is a Type l' Case. The Total Flow is Specified. Both recoverable and unrecoverable losses are considered. The system pressure at node I ) is 1000 psia. There are 6 nodes con (nected by 5 internal flow paths. There is 1- external flow path. A local convergence criterion of 0.10000-05 External heat addition was imposed. was used with 0.4883D-06 met. ' A heat addition criterion of 0.1000D-03 was used with 0.4883 Overall Convergence was met. Solution required 3 heat addition iterations. FSPLIT SA - FLOW SPLITS by DA FARNSWORTH & JA WEIMER - Light CASE 6-4A HEAT INPUT

                ----- EXTERNAL PATH INFORMATION -----

Ext Node Node Path Flow Press Orop Path IN OUT Node IN Node OUT Lbm/Sec Lt f/Sq inch Press, psi Ext Path .... .... .... ........... ........... Press, psi Tout, F 110 '

Document Number 32-1203121-01 1 6 1 0.20000D+04 0.35685D402 0.96432D+03 0.10000D+04 0 For Extarnal Path I the available NPSH is 0.970860+03 psia ( 70.0 F suction) FSPLIT Sa FLOW SPLITS by DA FARNSWORTH & JA WEIMER - Light Water Optio CASE 6-4A HEAT INPUT

                         ----- INTERNAL PATH INFORMATION -----

Connecting Path Flow Press Drop Path Node-> Node Unree DELTAP Recov DELTAP Lbm/Sec Lbf/Sq inch Lbf/Sq inch Lbf/Sq inch k [ h b bbbbbbb+bkb 5hhkhbb bhb bbbh[hb b5 2 2 3 0.200000D+04 0.1201770+02 b$khkbiib bk 3 3 4 0.714277D+01 0.4874880+01 0.2000000+04 0.758510D+01 4 4 5 0.7585100+01 0.000000D+00 0.2000000+04 0.2267900+01 5 6 5 0.714278D+01 .4874880+01

                                   .200000D+04       .256707D+01      .6907180+01    0.434011D+01 FSPLIT SA - FLOW SPLITS by DA FARNSWORTH & JA WEIMER - Light Wate CASE 6-4A HEAT INPUT
                       ----- INTERNAL PATH INFORMATION -----

Connecting Unrecoverable DP Recoverable DP Path Node-> Node Vel Head Friction Total UR Momentum Eleva k k h b$bbhh b$bbbb 2 2 3 bbbhh b$bbbb k$hkhk k$hkbi 7.1428 0.0000 7.1428 3 3 4 0.6779 4.1970 4.8749 7.5851 0.0000 7.5851 4 4 0.0000 0.0000 0.0000 5 7.1428 0.0000 7.1428 5 6 -0.6779 -4.1970 4.8749 5 -6.9072 0.0000 -6.9072 0.0000 1.1401 4.3401 FSPLIT SA - FLOW SPLITS by DA FARNSWORTH & JA WEINER - Light Water O CASE'6-4A HEAT INPUT

                     ----- INTERNAL PATH INFORMATION -----

Path Node Path Flow Input Heat In Out Velocity Path HTC Conductivity Lbm/sec Btu /Sec Ft/Sec B/Hr-Ft^2-F 8/Hr-Ft-F k i h b$hbbbbb b4 b$bbbbbb bb b$hhbbhb bh b$bbbb 111

Document Number 32-1203121-01 2 2 3

       -3     3    4     0.20000D+04 0.50000D+06 0.33093D+02 0.000000+00 0.390870+00 4     4    5 0.20000D+04 0.000000+00 0.35142D+02 0.00000D+00 0.39634D+00 0.20000D+04 5     6    5
                                         .50000D+06 0.33093D+02 0.00000D+00 0.390870+00
                           .20000D+04 0.00000D+00         .32001D+02 0.00000D+00 0.34942D+00 FSPLIT SA - FLOW SPLITS by DA FARNSWORTH & JA WEIMER - Light Water Option CASE 6-4A HEAT INPUT
                       ----- INTERNAL PATH INFORMATION -----

Path Node Vel Head Reynolds Friction Density In Out Lbf/Sq Inch Viscosity Number Factor Lbm/Ft^3 lbf-Sec/Ft^2 i k b b$b9bhhb bk b$bk5bbb bh b$bbbbbb bb b$bhk9hb bh b 2 2 3 3 3 4 0.71428D+01 0.106330+08 0.000000+00 0.604360+02 0.659650-05 4 4 5 0.75851D+01 0.19541D+08 0.000000+00 0.56912D+02 0.35895D-05 5 6 5 0.714280+01 0.106330+08 0.000000+00 0.604360+02 0.659650-05

                          .69072D+01 0.34500D+07 0.000000+00 0.624980+02 0.203310-04 FSPLIT SA - FLOW SPLITS by DA FARNSWORTH & JA WE1MER - Light Water Option CASE 6-4A HEAT INPUT
                      ----- INTERNAL PATH INFORMATION -----

Path Node Form Loss Path Heat Flow Area Roughness Hyd Dia In Out Fric Len Btu /Sec Ft^2 M inches Feet _ Feet 1 1 2 2 2 0.1000D+01 0.0000D+00 0.1000D+01 0.00000+00 0.11280+01 0.0000D+00 3 3 3 0.1000D+01 0.5000D+06 0.10000+01 0.00000+00 0.11280+01 0.00000+00 4 0.1000D+01 0.00000+00 0.10000+01 0.00000+00 0.11280+01 0.0000D+00 4 4 5 0.10000+01 .5000D+06 0.1000D+01 0.0000D+00 0.11280+01 0.00000+00 5 6 5 0.1000D+01 0.0000D+00 0.1000D+01 0.0000D+00 0.11280+01 0.0000D+00 FSPLIT SA - FLOW SPLITS by DA FARNSWORTH & JA WEIMER - Light Water Option CASE 6-4A HEAT INPUT

                     --------- NODE INFORMATION ---------

Path Inlet Rho

                                                       ~

Outlet Rho Inlet VH Outlet VH Inlet P Outlet P (Nodes) Lbm/Ft^3 lbm/Ft^3 lbf/Sg In Lbf/Sq In PSI PSI 112 4

Document Number 32-1203121-01 I 1 2 0.624980+02 0.624980+02 0.69072D+01 0.69072D+01 1000.00c) 988.7527 2 2 3 0.62498D+02 0.56912D+02 0.69072D+01 0.75851D+01 988.7527 976.7351 3 3 4 0.56912D+02 0.56912D+02 0.75851D+01 0.75851D+01 976.7351 969.1500 4 4 5 0.56912D+02 0.62498D+02 0.75851D+01 0.69072D+01 969.1500 966.8821 5 6 5 0.624980+02 0.624980+02 0.69072D+01 0.69072D+01 964.3150 966.8821 FSPLIT SA - FLOW SPLITS by DA FARNSWORTH & JA WEIMER - Light Water Option CASE 6-4A HEAT INPUT

                       --------- NODE INFORMATION ---------

Path Node Elev In Elev Out Area In Area Out Temp In Temp Out In Out Feet Feet Ft^2 Ft^2 Deg F Deg F 1 1 2 0.0000D+00 0.1000D+02 0.1000D+01 0.1000D+01 0.7000D+02 0.7000D+02 2 -2 3 0.1000D+02 0.20000+02 0.1000D+01 0.1000D+01 0.7000D+02 0.31890+03 3 3 4 0.20000+02 0.2000D+02 0/1000D+01 0.1000D+01 0.3189D+03 0.3189D+03 4 4 5 0.2000D+02 0.10000+02 0.1000D+01 0.1000D+01 0.31890+03 0.7000D+02 5 6 5 0.0000D+00 0.1000D+02 0.10000+01 0.1000D+01 0.7000D+02 0.70000+02 FSPLIT SA - FLOW SPLITS by DA FARNSWORTH & JA WEIMER - Light Water Option CASE 6-4A HEAT INPUT

                       ----- PATH / NODE DESCRIPTION -----

Connecting Path Node-> Node Description I 1 2 PATH I 2 2 3 HEAT IN PATH 3 3 4 PATH 3 4 4 5 HEAT OUT PATH 5 6 5 PATH 5 . FSPLIT SA - FLOW SPLITS by DA FARNSWORTH & JA WEIMER - Light Water Option 5 1 -1 CASE 6-4A HEAT INPUT PATH 1 1 2 0.100000D+01 0.100000D+01 0.699999D+02 0.100000D+01 0.000000D+00 0.1128380+01 0.0000000+00 0.0000000+00 0.000000D+00 0.100000D+02 0.1000000+01 0.100000D+01 113

i I Document Number 32-1203121-01  ; HEAT IN PATH 2 0.000000D+00 0.1128380+01 0.0000000+00 0.5000 0.100000D+02 PATH 3 0.200000D+02 0.1000000+01 0.100000D+01 3 0.000000D+00 0.112838D+01 0.000000D+00 0.0000 0.2000000+02 HEAT OUT PATH 0.2000000+02 0.1000000+01 0.1000000+01 4 5 0.100000D+01 0.100000D+01 0.194431D+03 0.1000000+01 0.0000000+00 0.1128380+01 0.000000D+00 .5000000+06 0.200000D+02 PATH 5 0.2000000+02 0.100000D+01 0.100000D+01 6 5 0.000000D+00 0.1000000+01 0.6999990+02 0.100000D+01 0.0000000+00 0.1128380+01 0.0000000+00 0.0000000+00 0.000000D+00 6 1 0.100000D+02 0.100000D+01 0.1000000+01 6 1 2000 0.7000000+02 1000 0 18 -12 1 1 28 14 18 20 20 14 18 44 14 14 i 42 44 14 34 42 4 24 34 42 -12 28 34 5E3532333333333333E553355E333535533333335233333553333 33333333333333333333333E3 CASE 7-4A i FSPLIT 5A - FLOW SPLITS by DA FARNSWORTH & JA WEIMER FSPLIT is a PC based thermal hydraulics computer code designed for the' generation and solution of steady state flow networks. The networks can contain any combinati3n of up to 100 nodes and 100 paths. . The problem may be specified by imposing up to 10 , boundary condition. flow boundary conditions (external paths) or one pressurl Problems may be defined with either ~  ! temperature inputs or heat inputs to nodes and paths. FSPLIT , uses graphical on screen modeling and parameter specification to greatly simplify model development.  ! In addition to individual path flow rates, FSPLIT solves for momentum, elevation, friction, and form loss pressure drops.  ! The form loss pressure drop calculations can assume a constant l 114 l

r Document Number 32-1203121-01 loss factor or can calculate a flow dependent loss factor. Equivalent heat rates are calculated for temperature input problems and temperatures are calculated for heat input problems. External path input flows may be constant or s head-capacity relationship to simulate a pump.pecified as a FSPLIT is designed to accommodate incompressible fluid, and gasses. water (H2O), heavy water (D20), any routines for light water and heavy water while properties forFSPLI gasses and incompressible fluids must be specified by the user. Creation Modes. Help screens are available in both the Node Creation an create a model and complete the analysis.They should provide { ad consult the user's manual. If you have problems concerns, contact one of the authors.If the manual does not answer your , FSPLITAllwas FORTRAN. writtenare calculations ininadouble combination of Microsoft QuickBA (64 bit) precision. matrix solutions are performed using the LINPACK solvers. Exec- All 1 The FSPLITSA.EXE, EXECHK.COM and FSPLITl i If FSPLIT is to be executed on a Hercules type monitor, the MSHERC.COM program must be run. PRIOR TO EXECUTING THIS CODE, CONTACT ONE OF THE AUTHORS TO ASSURE THAT THIS IS THE LATEST VERSION. FAILURE TO D0 S0 MAY INVALIDATE YOUR ANALYSIS. USER'S MANUAL - BWNT-TM-63 (ref 32-1203121-01) FSPLIT 5A FSPLIT SA - FLOW SPLITS by DA FARNSWORTH & JA WEIMER - Lig Check with the authors for the latest version.

SUMMARY

of RESULTS From Input File B: CASE 7 with a Checksum of 0.218064190D+04 CASE 7-4A NATURAL CIRCULATION This is a Type 2 Case. . Both recoverable and unrecoverable losses are considered.The T The system pressure at node 1 ) is 1000 psia. There are 6 nodes con (nected by 5 internal flow paths. There is 1 external flow path. A local convergence External heat addition was criterion imposed. of 0.1000D-05 was used with 0.2923D-08 met A heat addition criterion of 0.1000D-03 was usr.a with met. 0.76450-04 115

~ Document Number 32-1203121-01 Overall Convergence was met. Solution required 17 heat addition iterations. FSPLIT 5A - FLOW SPLITS by DA FARNSWORTH & JA WEIMER - Lig CASE 7-4A NATURAL CIRCULATION

                       ----- EXTERNAL PATH INFORMATION -----

Ext Node Node Path Flow Press Drop Path IN OUT Lbm/Sec Lbf/Sq inch Node IN Node OUT Ext Path Press, psi Press, psi Tout, F k 6 1 0.f0596b Ob b.bObbbb 00 b.100bbb Ok b FSPLIT SA - FLOW SPLITS by DA FARNSWORTH & JA WEIMER - Ligh CASE 7-4A NATURAL CIRCULATION

                     ----- INTERNAL PATH INFORMATION -----

Connecting Path Flow Press Drop Unrec DELTAP Path Node-> Node Lbm/Sec Lbf/Sq inch Recov DELTAP - Lbf/Sq inch Lbf/Sq inch [ [ h b.[bkhhdb 05 b k5h((hb b[ 2 2 3 0.1059560+03 b [b5b[bb$b5 b$khbibbb+bi 3 3 4 0.4261770+01 0.1983590-01 0.'1059560+03 0.198341D-01 0.4241930+01 4 4 5 0.105956D+03 0.1983410-01 0.000000D+00 5 5 6

                                                     .4282070+01      0.1955990-01 0.1059560+03         .432073D+01                          .430163D+01 0.193855D-01        .434011D+01 FSPLIT SA - FLOW SPLITS by DA FARNSWORTH & JA WEIMER - Light W CASE 7-4A NATURAL CIRCULATION J---

INTERNAL PATH INFORMATION ----- Connecting Unrecoverable DP Recoverable DP Path Node-> Node Vel Head Friction

    ....   .... ....          ........    ........      ........           Total     UR Momentum Ele 1      I       2         0.0196      0.0000 2     2 0.0196      0.0004       4.3012 3         0.0198      0.0000        0.0198                                4.3016 3     3        4         0.0198                               -0.0000      4.2419        4.2419 0.0000        0.0198     -0.0000 4     4        5         0.0196      0.0000                                0.0000      -0.0000 5     5 0.0196     -0.0004     -4.3012 6         0.0194      0.0000        0.0194                              -4.3016
                                                                     -0.0000     -4.3401      -4.3401 116

Document Number 32-1203121-01 FSPLIT SA'- FLOW SPLITS by DA FARNSWORTH & JA WEIMER - L CASE 7-4A NATURAL CIRCULATION

                                   ----- INTERNAL PATH INFORMATION -----

Path Node Path Flow Input Heat Velocity In Out Lbm/Sec 8tu/Sec Path HTC Conductivity

          ............               ...........                      Ft/Sec     B/Hr-Ft^2-F B/Hr-Ft-F I        1. 2 2        2     3       0.10596D+03 -0.100000+05 0.17107D+01 0.00000D+0 3        3     4       0.10596D+03 0.00000D+00 0.173460+01 0.00000D+00 4~

5 4 5 0.105960+03 0.00000D+00 '0.17346D+01 0.000000+00 0.105960+03 5 6 .10000D+05 0.17107D+01 0.00000D+00 0.37034D+ 0.10596D+03 0.00000D+00 0.16954D+01 0.00000D+00 FSPLIT 5A - FLOW SPLITS by DA FARNSWORTH & JA WEIMER - Ligh CASE 7-4A NATURAL CIRCULATION

                                 ----- INTERNAL PATH INFORMATION -----

Path Node . Vel Head Re3 olds - Friction Density In Out Lbf/Sq Inch Number Factor Viscosity Lbm/Ft^3 Lbf-Sec/Ft^2 i i b 2 2 3 b$kbbbib bk b$bikbbb bb b$bbbbbb bb b$b 3 3 4 0.19835D-01 0.461090+06 0.00000D+00 0.'61084D+02 0 4 4 5 0.198350-01 0.461090+06 0.00000D+00 0.61084D+02 0. 5 5 6 0.195610-01 0.31105D+06 0.00000D+00 0.61937D+02 0.1 0.19386D-01 0.18277D+06 0.00000D+00 0.62498D+02 0.2 FSPLIT.5/L - FLOW SPLITS by DA FARNSWORTH & JA WEIMER - Licht W CASE 7-4A NATURAL CIRCULATION

                               ----- INTERNAL ~ PATH INFORMATION -----

' ' Path Node . In Out Form Loss Path Heat Flow Area Roughness Hyd Dia Btu /Sec Ft'2 M inches fric Len Feet Feet 1 1 2' 2 2 0.10000+01.0.10000+05 0.1000D+01 0.0000D+00 0.11280+0 3 3 3 4 0.1000D+01 0.0000D+00 0.1000D+01 0.0000D+00 0.1128D+01 4 4 5 0.1000D+01 0.00000+00 0.1000D+01 0.00000+00 0.11280+01 0.1000D+01

                                             .1000D+05 0.10000+01 0.0000D+00 0.1128D+01 0.0000D+

117 e

Docuraent Number 32-1203121-01 5 5 6 0.1000D+01 0.00000+00 0.1000D+01 0.00000+00 0.1128D+01. 0.0000 FSPLIT 5A - FLOW SPLITS by DA FARNSWORTH & JA WEIMER - Light Water Option CASE 7-4A NATURAL CIRCULATION

                             --------- NODE INFORMATION -- ------

Path Inlet Rho Outlet Rho Inlet VH Outlet VH Inlet P Outlet P (Nodes) Lbm/Ft^3 Lbm/Ft^3 lbf/Sq In Lbf/Sq In PSI PSI k k h b$6h4hbb bh b$hkbbkb bh b$khhbbb5b[ b$5hbbhb 2 2 3 0.61084D+02 0.61084D+02 0.198350-01 0.19835D-01 995.6788 991.41

3. 3 4 0.F'084D+02 0.61084D+02 0.198350-01 0.198350-01 991.4170 991.397 4 4 5 0.6tv84D+02 0.62498D+02 0.19835D-01 0.193860-01 991.3972 995.6793 5 5 6 0.62498D+02 0.624980+02 0.193860-01 0.193860-01 995.6793 1000.000 FSPLIT SA - FLOW SPLITS by DA FARNSWORTH & JA WEIMER - Light Water Option CASE 7-4A NATURAL CIRCULATION
                            --------- NODE INFORMATION ---------

Path Node Elev In Elev Out Area In Area Out Temp In Temp Out In Out Feet Feet Ft^2 Ft^2 Deg F Deg F 1 1 2 2 32 0.0000D+00 0.1000D+02 0.1000D+01 0.10000+01 0.70000+02 0.1649D+03 3 3 4 0.10000+02 0.20000+02 0.1000D+01 0.10000+01 0.16490+03 0.16490+03 4 4 5 0.20000+02 0.2000D+02 0.1000D+01 0.10000+01 0.1649D+03 0.1649D+03 5 5 6 0.2000D+02 0.1000D+02 0.1000D+01 0.1000D+01 0.1649D+03 0.7000D+02 0.10000+02 0.0000D+00 0.1000D+01 0.1000D+01 0.7000D+02 0.7000D+02 i FSPLIT SA - FLOW SPLITS by DA FARNSWORTH & JA WEIMER - Light Water Option CASE 7-4A NATURAL CIRCULATION

                          ----- PATH / NODE D$5CRIPTION -----

Connecting I Path Node-> Node Description I 1 2 PATH I 2- 2 3 HEAT IN PATH 3 3 4 PATH 3 4 4 5 HEAT OUT PATH 118

                                                                                               )

Document Number 32-1203121-01 5 5 6 PATH 5 FSPLIT 5A - FLOW SPLITS by DA FARNSWORTH & JA WEIMER - Light Water 5 1 -2 CASE 7-4A NATURAL CIRCULATION PATH 1 1 2 0.1000000+01 0.100000D+01 0.7000000+02 0.1000000+01 0.000000D+00 0.112838D+01 0.0000000+00 0.100000D+05 0.0000000+00 HEAT IN PATH 0.1000000+02 0.1000000+01 0.100000D+01 2 3 0.100000D+01 0.1000000+01 0.7000000+02 0.100000D+01 0.0000000+00 0.1128380+01 0.000000D+00 0.000000D+00 0.100000D+02 PATH 3 0.200000D+02 0.1000000+01 0.1000000+01 3

             ~4 0.1000000+01 0.1000000+01 0.700000D+02 0.1000000+01 0.0000000+00 0.112838D+01 0.000000D+00 0.0000000+00 0.2000000+02 0.2000000+02 0.1000000+01 0.100000D+01 HEAT OUT PATH 4

5 0.100000D+01 0.100000D+01 0.7000000+02 0.1000000+01 0.0000000+00 0.112838D+01 0.0000000+00 .1000000+05 0.200000D+02 PATH 5 0.100000D+02 0.100000D+01 0.100000D+01 5 6 0.1000000+01 0.100000D+01 0.700000D+02 0.100000D+01 0.000000D+00 0.1128380+01 0.0000000+00 0.000000D+00 0.100000D+02 6 1 0.000000D+00 0.100000D+01 0.1000000+01 6 0 0.7000000+02 1 1000 18 0 1

                     -12            28             14 18                  20           20             14 18                  44           14             14 42                 44            14            34 42                  4             24            34 42                -12             28            34 CASE 8-4A FSPLIT 5A - FLOW SPLITS by DA FARNSWORTH & JA WEIMER FSPLIT is a PC based thermal hydraulics computer code designed for the generation and solution of steady state flow networks.

119 l

Document Number 32-1203121-01 The networks can contain any combination of up to 100 nodes and 100 paths. The problem may be specified by imposing up to 10

      ' flow boundary conditions (external paths) or one pressure drop boundary condition. Problems may be defined with either temperature inputs or heat inputs to nodes and paths. FSPLIT uses graphical on screen modeling and parameter specification to greatly simplify model development.

In addition to individual path flow rates, FSPLIT solves for momentum, elevation, friction, and form loss pressure drops. The form loss pressure drop calculations can assume a constant loss factor or can calculats a' flow dependent loss factor. Equivalent heat rates are calculated for temperature input problems and temperatures are calculated for heat input problems. External path input flows may be constant or specified as a head-capacity relationship to simulate a pump. FSPLIT is designed to accommodate water (H20), heavy water (D20), any incompressible fluid, and gasses. FSPLIT has internal property routines for light water and heavy water while properties for gasses and incompressible fluids must be specified by the user. Help screens are available in both the Node Creation and Path Creation Modes. They should provide adequate knowledge to create a model and complete the analysis. If you have problems, consult the user's manual. If the manual does not answer your concerns, contact one of the authors. FSPLIT was written in a combination of Microsoft QuickBASIC and FORTRAN. All calculations are in double (64 bit) precision. All matrix solutions are performed using the LINPACK solvers. Exec-ution of FSPLIT requires roughly 550 kilobytes of free DOS memory. The FSPLITSA.EXE, EXECHK.COM and FSPLIT.HLP files must be present. If FSPLIT is to be executed on a Hercules type monitor, the MSHERC.COM program must be run. PRIOR TO EXECUTING THIS CODE, CONTACT ONE OF THE AUTHORS TO ASSURE THAT THIS IS THE LATEST VERSION. FAILURE TO DO S0 MAY INVALIDATE YOUR ANALYSIS. USER'S MANUAL - BWNT-TM-63 (ref 32-1203121-01) FSPLIT 5A FSPLIT 5A - FLOW SPLITS by DA FARNSWORTH & JA WEIMER - Light Water Option Check with the authors for the latest version.

SUMMARY

of RESULTS From Input File B: CASE 8 with a Checksum of 0.670375385D+05 CASE 8-4A TEE & Y TEST (IDELCHIK)

 .                                        120 l

l

Document (umber 32-1203121-01 This is a Type 3 Cue. The Total Flow is Specified Both recoverable and unrecoverable losses are consiSered. The system pressure ( at node 11 ) is 200 psia. There are 12 nodes connected by 13 internal flow paths. There is 1 external flow path. A local convergence criterion of 0.1000D-05 was used with 0.22980-13 met. Overall convergence was met. Solution required 58 local iterations. FSPLIT SA - FLOW SPLITS by DA FARNSWORTH & JA WEIMER - Light Water Option CASE 8-4A TEE & Y TEST (IDELCHIK)

                      ------ EXTERNAL PATH INFORMATION -----                                       ]

Ext Node Node Path Flow Press Drop Node IN Node OUT Path IN OUT Lbm/Sec Lbf/Sq inch Press, psi ' Press, psi k kk [b b$hbbbbb b4 b$bbbbhb bb b$bbbbbb bb b$bb6bhb bb For External Path 1 the available NPSH is 0.284460+03 psia ( 70.0 F suction) FSPLIT 5A - FLOW SPLITS by DA FARNSWORTH & JA WEIMER - Light Water Option CASE 8-4A TEE & Y TEST (IDELCHIK)

                       ----- INTERNAL PATH INFORMATION -----

Connecting Path Flow Press Drop Unrec DELTAP Recov DELTAP Path Node-> Node Lbm/Sec Lbf/Sq inch Lbf/Sq if:h Lbf/Sq inch

                                    ~         ~
         ~~i~ "i~ "5             5.~2fb5bfb+b4 b[b5kibib+b5        b$555bb4b+b5 [555khib+bb 2       1        2    0.425493D+04     0.722770D+02     0.1257600+03        .5348280+02 3       3        5    0.2745070+04     0.258093D+03     0.106875D+03      0.1512180+03

. 4 2 4 0.425493D+04 0.0000000+00 0.000000D+00 0.000000D+00 5 4' 8 0.267471D+04 0.436353D+02 0.254382D+02 0.181971D+02 6- 8 7 0.267471D+04 0.0000000+00 0.0000000+00 0.0000000+00 7 7 5 0.267471D+04 0.237600D+03 0.8374170+02 0.1538580+03 8 4 6 0.1580210+04 .2209180+02 0.492571D+01 - 2701750+02

 .         9       5        9    0.5419790+04     0.568434D-13     0.568434D-13      0.0000000+00 10       6      10     0.158021D+04     0.335834D+03     0.255334D+03      0.805003D+02 11       9      10     0.5419790+04     0.325069D+02     0.1510800+03        .118573D+03 ,

12 12 1 0.7000000+04- .1136870-12 .113687D-12 0.000000D+00 121 l

Document Number 32-1203121-01 13 10 11 0.700000D+04 0.000000D+00 0.0000000+00 0.0000000+00 FSPLIT 5A - FLOW SPLITS by DA FARNSWORTH'& JA WEIMER - Light Water Option CASE 8-4A TEE & Y TEST (IDELCHIK)

                       ----- INTERNAL PATH INFORMATION -----

Connecting Unrecoverable DP Recoverable DP Path Node-> Node Val Head Friction Total UR Momentum Elevation Total RE I 1 3 128.0642 0.0000 128.0642 -32.6451 2 1. 0.0000 -32.6451 2 125.7598 0.0000 125.7598 -53.4828 0.0000 -53.4828 3 3 5 106.8755 0.0000 106.8755 151.2179 0.0000 151.2179 4 2 4 0.0000 0.0000 0.0000 0.0000 0.0000 0.0000 ( 5 4 8 25.4382 0.0000 25.4382 18.1971 0.0000 18.1971 I 6 8 7 0.0000 0.0000 0.0000 -0.0000 0.0000 -0.0000 7 7 5 83.7417 0.0000 83.7417 153.8584 0.0000 153.8584 8 4 6 4.9257 0.0000 4.9257 -27.0175 0.0000 -27.0175 j 9 5 9 -0.0000 0.0000 -0.0000 0.0000 0.0000 0.0000 10 6 10 255.3339 0.0000 255.3339 80.5003 0.0000 80.5003 11 9 10 151.0797 0.0000 151.0797 -118.5727 0.0000 -118.5727 4 12 12 - 1 -0.0000 0.0000 -0.0000 0.0000- 0.0000 0.0000 13 10 11 0.0000 0.0000 0.0000 0.0000 0.0000 0.0000 FSPLIT SA - FLOW SPLITS by DA FARNSWORTH & JA hEIMER - Light Water Option CASE 8-4A TEE & Y TEST (IDELCHIK)

                      ----- INTERNAL PATH INFORMATION -----

Path Node Path Flow Input Heat Velocity Path HTC Conductivity In Out Lbm/Sec Btu /Sec Ft/Sec B/Hr-Ft^2-F B/Hr-Ft-F [ [ b b$bhkb[b hk b$bbbbbb bb b$kkbbbD bb b$bbbbbb bb b$bkhh , 2 1 2 0.425490+04 0.000000+00 0.682500+02 0.00000D+00 0.34771D+00 l 3 3 5 0.27451D+04 0.00000D+00 0.88064D+02 0.00000D+00 0.347710+00 4 2 4 0.425490+04 0.000000+00 0.682500+02 0.00000D+00 0.34771D+00 5 =4 8 0.267470+04 0.00000D+00 0.858070+02 0.000000+00 0.34771D+00 6 8 7 0.26747D+04 0.00000D+00 0.85807D+02 0.00000D+00 0.34771D+00 7 7 5 0.26747D+04 0.00000D+00 0.85807D+02 0.000000+00 0.34771D+00 8 4 6 0.15802D+04 0.000000+00 0.253470+02 0.00000D+00 0.34771D+00 9 5 9 0.54198D+04 0.000000+00 0.17387D+03 0.00000D+00 0.347710+00 10 6 10 0.15802D+04 0.00000D+00 0.253470+02 0.00000D+00 0.34771D+00  ! 11 9 10 0.541980+04 0.000000+00 0'.86935D+02 0.000000+00 0.34771D+00 12 12 1 0.70000D+04 0.000000+00 0.112280+03 0.000000+00 0.34771D+00 l 122 O

Document Number 32-1203121-01 i 13 10 11-0.700000+04 0.000000+00 0.112280+03 0.000000+00 0.34771D+0 FSPLIT SA - FLOW SPLITS by DA FARNSWORTH & JA WEIMER - Light Water Option CASE 8-4A TEE & Y TEST'(IDELCHIK)

                           ------INTERNAL PATH INFORMATION --.--

Path Node Vel Head Reynolds In Out Lbf/Sq Inch Friction Density Viscosity Number Factor

    ............            ...........                                Lbm/Ft^3     lbf-Sec/Ft^2 I      1     3 2      1      2'     0.13044D+02 0.33423D+07 0.000000+00 0.62343D+02 0.20368D-0 3      3      5      O.31340D+02 0.73264D+07 0.00000D+00 0.62343D+02 0.203680-0 4      2     4       0.521780+02 0.66845D+07 0.000000+00 0.62343D+02 0.20368D-04 5      4     8       0.31340D+02 0.73264D+07 0.000000+00 0.62343D+02 0.20368D-0 6      8     7 0.495370+02 0.92110D+07 0.00000D+00 0.62343D+02 0,203680-04 7      7     5       0.495370+02 0.921100+07 0.000000+00 0.62343D+02 0.20368D-04 8      4     6       0.495370+02 0.65132D+07 0.00000D+00 0.62343D+02 0.20368D-0 9      5     9      0.432260+01 0.272090+07 0.000000+00 0.62343D+02 0.203680-04 10      6 10          0.20340D+03 0.131980+08 0.000000+00 0.62343D+02 0.203680-04 11       9 10          0.432260+01 0.272090+07 0.00000D+00 0.62343D+02 0.203680-04 12 12           1      0.50849D+02 0.65989D+07 0.000000+00 0.62343D+02 0.20368D-04 13 10 11               0.84823D+02 0.120530+08 0.000000+00 0.62343D+02 0.203680-04 l         .

0.84823D+02 0.12053D+08 0.00000D+00 0.62343D+02 0.203680-04( FSPLIT SA - FLOW SPLITS by DA FARNSWORTH & JA WEIMER - Light Water Option CASE 8-4A TEE & Y TEST (IDELCHIK)

                        ----- INTERNAL PATH INFORMATION -----

Path Node In Out Form Loss Path Temp Flow Area Roughness Deg F- Ft^2 M inches Hyd Dia Fric Len

                         ,                                                     Feet       Feet 1     1      3 2      1      2    0.9818D+01 0.7000D+02 0.10000+01 0.00000+00 0.79790+00 0.0000D+00 3      3           0.4013D+01 J.7000D+02 0.1000D+01 0.0000D+00 0.1128D+01 0.00000+0 4      2'     5 'O.20480+01 0.7000D+02 0.50000+00 0.0000D+00 0.79790+00 0 4

5 4 8 0.00000+00 0.70000+02 0.10000+01 0.0000D+00 0.11280+01 0.0000D+00 6 8 7 0.51350+00 0.7000D+02 0.5000D+00 0.0000D+00 0.11280+01 0.0000D+00 . 7 7 5 0.0000D+00 0.7000D+02 0.50000+00 0.0000D+00 0.11280+01 0.0000D+00 8' 4 6 0.1690D+01 0.7000D+02 0.5000D+00 0.00000+00 0.79790+00 0.0000D+00 9 5 9 0.1140D+01 0.7000D+02 0.1000D+01 0.0000D+00 0.11280+01 0.0000D+0 10' 6 10 0.0000D+00 0.70000+02 0.50000+00 0.0000D+00 0.79790+00 0.0000D+00 0.59070+02 0.70000+02 0.10000+01 0.0000D+00 0.11780+01 0.00000+00 123

1 Document Number 32-1203121-01 11 9 10 0.2971D+01 0.7000D+02 0.1000D+01 0.0000D+00 0.7979D+00 0.0000D+00 12 12 1 0.0000D+00 0.70000+02 0.1000D+01 0.00000+00 0.11280+01 0.0000D+00 13 10 11 0.0000D+00 0.70000+02 0.1000D+01 0.00000+00 0.11280+01 0.0000D+00 FSPLIT SA - FLOW SPLITS by DA FARNSWORTH & JA WEIMER - Light Water Option CASE 8-4A TEE & Y TEST (IDELCHIK)

                      --------- NODE INFORMATION ---------

Path Inlet Rho Outlet Rho Inlet VH Outlet VH Inlet P Outlet P (Nodes) Lbm/Ft^3 . Lbm/Ft^3 Lbf/Sq In Lbf/Sq In PSI PSI I 1 3 0.62343D+02 0.62343D+02 0.84823D+02 0.521780+02 586.0194 490.6003 2 'l 2 0.62343D+02 0.62343D+02-0.84823D+02 0.31340D+02 586.0194 513.7424 3 3 5 0.62343D+02 0.62343D+02 0.521780+02 0.203400+03 490.6003 232.5069 4 2 4 0.62343D+02 0.62343D+02 0.31340D+02 0.31340D+02 513.7424 513.7424 5 4-8 0.62343D+02 0.62343D+02 0.31340D+02 0.49537D+02 513.7424 470.1071 6 8 7 0.62343D+02 0.62343D+02 0.49537D+02 0.49537D+02 470.1071 470.1071 7 7 5 0.62343D+02 0.62343D+02 0.49537D+02 0.20340D+03 470.1071 232.5069 8 4 6 0.62343D+02 0.62343D+02 0.31340D+02 0.43226D+01 513.7424 535.8342 9 5 9 0.62343D+02 0.62343D+02 0.20340D+03 0.203400+03 232.5069 232.5069 10 6 10 0.62343D+02 0.62343D+02 0.432260+01 0.84823D+02 535.8342 200.0000 11 9 10 0.62343D+02 0.62343D+02 0.203400+03 0.84823D+02 232.5069 200.0000 12 12 1 0.62343D+02 0.62343D+02 0.84823D+02 0.84823D+02 586.0194 586.0194

   - 13 10 11 '0.62343D+02 0.62343D+02 0.84823D+02 0.84823D+02 200.0000 200.0000 FSPLIT SA - FLOW SPLITS by DA FARNSWORTH & JA WEIMER - Light Water Option CASE 8 4A TEE & Y TEST (IDELCHIK)
                      --.------ N0DE INFORMATION ---------                                             i Path      Node ' Elev In            Elev Out        Area In      Area Out   Temp In   Temp Out In Out         Feet            Feet           Ft^2           Ft^2     Deg F      Deg F  .

I 1 3 0.0000D+00 0.0000D+00 0.10000+01 0.50000+00 0.7000D+02 0.7000D+02 2 1 2 0.0000D+00 0.0000D+00 0.1000D+01 0.1000D+01 0.70000+02 0.7000D+02 3 3 5 0.0000D+00 0.0000D+00 0.50000+00 0.5000D+00 0.70000+02 0.70000+02 4 2 4 0.0000D+00 0.0000D+00 0.1000D+01 0.1000D+01 0.7000D+02 0.70000+02 5 -4 8 0.0000D+00 0.00000+00 0.10000+01 0.50000+00 0.70000+02 0.7000D+02

 ~

6 8 7 0.0000D+00 0.0000D+00 0.5000D+00 0.5000D+00 0.7000D+02 0.70000+02 7 7 5 0.0000D+00 0.0000D+00 0.5000D+00 0.5000D+00 0.7000D+02 0.7000D+02 8 4 6 0.0000D+00 0.0000D+00 0.1000D+01 0.1000D+01 0.70000+02 0.70000+02 9 5 9 0.00000+00 0.00000+00 0.5000D+00 0.5000D+00 0.70000+02 0.70000+02 124

Document Number 32-1203121-01 10 6 10 11 9 10 0.0000D+00 0.00000+00 0.10000+01 0.1000D+01 0.7000D+02 0 12 12 :1 0.0000D+00 0.00000+00 0.5000D+00 0.1000D+01 0.7000D+02 0 13 10 11 0.00000+00 0.00000+00 0.10000+010.10000+01 0.70000+02 0.700 0.0000D+00 0.0000D+00 0.1000D+01 0.10000+010.7000D+02 0. FSPLIT SA - FLOW SPLITS by DA FARNSWORTH & JA WEIMER - Light Water Optio CASE 8-4A TEE & Y TEST (IDELCHIK)

                            ----- PATH / NODE DESCRIPTION -----

Connecting Path Node-> Node Description

        "~~ "~~              ~

l l 3 RIGHT PATH #1 2 1 2 LEFT PATH #2 3 3 5 TOP OF CONVERGING TEE #3 4 2 4 5 4 8 TOP OF DIVERGING TEE (LEFT #4) 6 8 7 MIDDLE LEG#6OF DIVERGING TEE-(CROSS CONCT #5) CROSS CONNECT 7 7 5 8 4 6 MIDDLE LEG 0F CONVERGING TEE (CROSS CONCT #7) 9 5 9 DIVERGING TOP OF TEE (LEFT # 8) 10 6 10 CONVERGING LEFT #10 TOP OF TEE (RIGHT # 9)~ 11 9 10 RIGHT #11 12 12 1 PATH 12 13 10 11 PATH.13

       .FSPLIT SA - FLOW SPLITS by DA FARNSWORTH & JA WEIMER - Light Water Option 13 1 3 CASE 8-4A TEE & Y. TEST (IDELCHIK)

RIGHT PATH #1 1 1 3 0.300000D+01 12 12 0.3000000+01 0.700000D+02 .100000D+01 11 0.000000D+00 0.100000D+01 0.100000D+00 0.100000D+01 0.2000000+00 0.101000D+01 0.300000D+00 0.103000D+01 - 0.400000D+00.0.105000D+01~ 0.500000D+00 0.1080000+01

  • 0.600000D+00 0.1110000+01
    .0.700000D+00 0.1150000+01 0.800000D+00 0.119000D+01
   -0.900000D+00 0.1240000+01 0.100000D+01 0.130000D+01 0.000000D+00 0.797885D+00 0.0000000+00 0.0000000+00 LEFT      PATH #2       0.000000D+00 0.100000D+01 0.500000D+00 0.7000000+02 0.700 125

Document Number 32-1203121-01 2_1 2 0.1000000+01 12 12 0.1000000+01 11 0.700000D+02 .100000D+01 0.0000000+00 0.1000000+01 0.100000D+00 0.1000000+01 0.;!000000+00 0.1010000+01 0.300000D+00 0.1030000+01 0.4000000+00 0.105000D+01 0.500000D+00 0.1080000+01 0.600000D+00 0.111000D+01 0.7000000+00 0.115000D+01 0.8000000+00 0.1190000+01 0.500000D+00 0.124000D+01 . 0.1000000+01 0.130000D+01 0.0000000+00 0.112838D+01 0.0000000+00 0.000000D+00 irP OF CONVERGING TEE #3 3~ 0.0000000+00 0.100000D+01 0.100000D+01 0 7 5 0.000000D+00 9 9 0.000000D+00 11 0.7000000+02 .5000000+00 0.000000D+00 0.000000D+00 0.100000D+00 0.160000D+00 0.200000D+00 0.2700000+00 0.300000D+00 0.3800000+00 0.400000D+00 0.460000D+00 0.5000000+00 0.530000D+00 0.6000000+00 0.5700000+00 . 0.700000D+00 0.5900000+00 0.800000D+00 0.6000000+00 0.900000D+00 0.590000D+00 0.1000000+01 0.5500000+00 0.000000D+00 0.7978850+00 0.000000D+00 TOP 2 GF DIVERGING TEE (LEFT #4)0.000000D+00 0.00 4.0.0000000+00 0.000000D+00 0.7000000+02 0.1000000+01 0.0000000+00 0.112838D+01 0.000000D+00

   ' MIDDLE LEG OF DIVERGING TEE (CROSS CONC 4

5 8 0.000000D+00 4 4 0.000000D+00 12 0.7000000+02 .500000D+00 0.0000000+00 0.550000D+00 0.5000000-01 0.550000D+00 0.1000000+00 0.555500D+00 0.200000D+00 0.577500D+00 0.300000D+00 0.610500D+00 0.400000D+00 0.6545000+00 0.5000000+00 0.715000D+00 0.6000000+00 0.7865000+00 0.700000D+00 0.874500D+00 0.800000D+00 0.973500D+00 0.100000D+01 0.121000D+01 0.1300000+01 0.220500D+01

 +

126 9

I Document Number 32-1203121-01 0.000000D+00 0.1128380+01 0.0000000+00 0.0000000+00 0.0000000+00 0.100000D+01 0.500000D+00 0.700000D+02 0 CRDSS CONNECT #6 8 17 0.000000D+00 0.000000D+00 0.7000000+02 0.5000000+00 0.000000D+00 0.112838D+01 0.0000000+00 0.0000000+00 MIDDLE' LEG 0.0000000+00 0.5000000+00 0.500000D+00 0.700000D+02 0.70 7 OF CONVERGING TEE (CROSS CONCT #7)- 7 -5 0.000000D+00 9 0.0000000+00 0.7000000+02 .5000000+00 9 11 0.000000D+00 .550000D+00 ' O.100000D+00 .335500D+00 0.200000D+00 .165000D+00

    '    0.300000D+00 .6050000-01 O.400000D+00 0.242000D+00 0.500000D+00 0.423500D+00 0.6000000+00 0.572000D+00 0.700000D+00 0.715000D+00 0.800000D+00 0.858000D+00 0.9000000+00 0.9900000+00 0.100000D+01 0.110000D+01 0.0000000+00 0.7978850+00 0.0000000+00 0.000000D+00 0.0000000+00 0.5000000+00 0.5000000+00 0.7000000+02 0.700000 DIVERGING TOP OF TEE (LEFT # 8) 4 6 0.000000D+00 0.000000D+00 0.7000000+02     .100000D+01              I 8                 '4       4         9 0.000000D+00 0.400000D+00 0.100000D+00 0.3200000+00 0.200000D+00 0.2600000+00 0.300000D+00 0.200000D+00 0.400000D+00 0.140000D+00 0.500000D+00 0.100000D+00
      '0.600000D+00 0.6000000-01 0.800000D+00 0.2000000-01 0.100000D+01 0.000000D+00-0.000000D+00 0.112838D+01'O.000000D+00 0.0000000+00 0.0000000+00 0.100000D+01 0.100000D+01 0.7000000+02 0.700000D CONVERGING TOP 0F TEE (RIGHT # 9) 5 9 0.000000D+00 0.0000000+00 0.7000000+02 0.5000000+00 0.0000000+00 0.7978850+00.0.000000D+00 0.000000D+00 LEFT    #10            0.000000D+00 0.500000D+00 0.500000D+00 0.700000D+02 0.700000D 6

10: 10 0.300000D+02 0.300000D+02 0.7000000+02 "

                                                                .100000D+01 13      13        'll.

0.000000D+00 0.200000D+01 0.1000000+00 0.1730000+01 0.200000D+00 0.152000D+01 0.300000D+00 0.137000D+01 0.400000D+00 0.128000D+01 0.5000000+00 0.125000D+01 0.600000D+00 0.128000D+01 127 h

y Document Number 32-1203121-01 0.700000D+00 0.137000D+01 0.8000000+00 0.152000D+01 i 0.9000000+00 0.173000D+01 0.1000000+01 0.200000D+01 0.000000D+00 0.1128380+01 0.000000D+00 0.000000D+00 0.000000D+00 0.100000D+01 0.100000D+01 0.700000D+02 0.7000000+02 RIGHT #11 9 10.0.500000D+00 0.5000000+00 0.700000D+02 .1000000+01 11 13 13 11 0.0000000+00 0.200000D+01 0.100000D+00 0.173000D+01 0.200000D+00 0.1520000+01 j 0.300000D+00 0.137000D+01 ' 0.400000D+00 0.128000D+01 O.500000D+00 0.125000D+01 0.6000000+00 0.1280000+01 0.700000D+00 0.1370000+01 0.800000D+00 0.1520000+01 0.900000D+00 0.1730000+01 0.100000D+01 0.2000000+01 0.0000000+00 0.797885D+00 0.000000D+00 0.0000000+00 0.0000000+00 0.500000D+00 0.1000000+01 0.7000000+02 0.7000000+02 PATH 12-12 1 0.0000000+00 0.000000D+00 0.700000D+02 0.1000000+01

         . 0.000000D+00 0.1128380+01 0.0000000+00 0.000000D+00 0.000000D+00 0.100000D+01 0.1000000+01 0.7000000+02 0'.700000D+02-PATH 13 10   11 0.0000000+00 0.000000D+00 0.7000000+02 0.100000D+01 0.000000D+00 0.1128380+01 0.000000D+00 0.0000000+00 0.000000D+00 0.1000000+01 0.1000000+01 0.700000D+02 0.7000000+02 11             12             7000 12             1              200          0              11 22             52             12            18 22             68            8              18 22          /  36             16            18 32             68            8             26 32             36             16           26 42             68            8             34 32             44             14           26 32-             60            10            26 42             36            16            34
   .       42              52            12            34 62             52'           12            50 2               52            12            2 4

383335333333333333333333333333333333333333333333333553333333333333333333333333 128 I

r Document Nurber 32-1203121-01 CASE 9-4A FSPLIT 5A - FLOW SPLITS by DA FARNSWORTH & JA WEIMER 1 FSPLIT is a PC based thermal hydraulics computer code designed for the generation and solution of steady state flow networks. - The networks can contain any combination of up to 100 nodes and 100 paths. The problem may be specified by imposing up to 10 flow boundary conditions (external paths) or one pressure drop boundary condition. Problems may be defined with either temperature inputs or heat inputs to nodes and paths. FSPLIT uses graphical on screen modeling and parameter specification to greatly simplify model development. In addition to individual path flow rates, FSPLIT solves for momentum, elevation, friction, and form loss pressure drops. The form los.s pressure drop calculations can assume a constant loss factor or can calculate a flow dependent loss factor. Equivalent heat rates are calculated for temperature input problems and temperatures are calculated for heat input problems. External path input flows may be constant or specified as a head-capacity relationship to simulate a pump. FSPLIT is designed to accommodate water (H20), heavy water (D20), any incompressible fluid, and gasses. FSPLIT has internal property routines for light water and heavy water while properties for gasses and incompressible fluids must be specified by the user. l Help screens are available in both the Node Creation and Path Creation Modes. They should provide adequate knowledge to create a model and complete the analysis. If you have problems, consult the user's manual. If the manual does not answer your , concerns, contact one of the authors. FSPLIT was written in a combination of Microsoft QuickBASIC and FORTRAN. All calculations are in double (64 bit) precision. All matrix solutions are performed using the LINPACK solvers. Exec-ution of FSPLIT requires roughly 550 kilobytes of free DOS memory. The FSPLITSA.EXE, EXECHK.COM and FSPLIT.HLP files must be.present. If FSPLIT is to be executed on a Hercules type monitor, the MSHERC.COM program must be run. PRIOR TO EXECUTING THIS CODE, CONTACT ONE OF THE AUTHORS TO ASSURE THAT THIS IS THE LATEST VERSION. FAILURE TO DO S0 MAY INVALIDATE YOUR ANALYSIS. USER'S MANVAL - BWNT-TM-63 (ref 32-1203121-01) FSPLIT SA 129

h Document Number 32-1203121-01 FSPLIT 5A - FLOW SPLITS by DA FARNSWORTH & JA WEIMER - Gas Table Option Check with the authors for the latest version.

SUMMARY

of RESULTS

               'From Input File B: CASE 9 with a Checksum of 0.501597885D+04 CASE 9-4A GAS (AIR) FLOW This is a Type 1 Case. The Total Flow is Specified.

Both recoverable and unrecoverable losses are considered.

                                                    ~

The system pressure ( at node 6 ) is 15 psia. There are 6 nodes connected by 5 internal flow paths. There' is.1 external flow path. The maximum Mach number was 0.385 in path 5 - Cp/Cv was 1.4 A local convergence criterion of 0.10000-05 was used with.0.98600-06 met. Variable density-pressure was imposed. A density-pressure criterion of 0.10000-03 was used with 0.6902D-05 met. Overall Convergence was met. Solution required 6 density-pressure iterations. FSPLIT SA - FLOW SPLITS by DA FARNSWORTH & JA WEIMER - Gas Table Option CASE 9-4A GAS (AIR) FLOW

                        ----- EXTERNAL PATH INFORMATION -----

Ext Node Node Path Flow Press Drop Nede IN Node OUT Path IN OUT Lbm/Sec Lbf/Sq inch Press, psi Press, psi h b [ b$hbbbbb bh b$kkkhhb bh b$[hbbbb bh b$hbkkhb bh FSPLIT SA - FLOW SPLITS by DA FARNSWORTH & JA WEIMER - Gas Table Option

                         .                                                                         3 CASE 9-4A GAS (AIR) FLOW
                        ----- INTERNAL PATH INFORMATION -----

Connecting Path Flow Press Drop Unrec DELTAP Recov DELTAP f Path Node-> Node Lbm/Sec Lbf/Sq inch Lbf/Sq inch Lbf/Sq inch h h h 0.700000b+0k b [bbbhhb bl bhkkbhbbbk hhkbkkbbb l 3 0.700000D+02 0.209886D+01 0.196224D+01 0.136623D+00 j 2 2 3 3 4 0.700000D+02 0.258476D+01 0.235437D+01 0.230396D+00 130

4 Document Number 32-1203121-01 4 5 5 0.7000000+02 0.328726D+01 5 6 0.700000D+02 0.2957380+01 0.329883D+00 0.458452D+01 0.395184D+01 0.632689D+00  : FSPLIT 5A - FLOW SPLITS by DA FARNSWORTH & JA WEIMER - Gas Tab CASE 9-4A GAS (AIR) FLOW

                        ----- INTERNAL PATH INFORMATION -----

Connecting Unrecoverable DP Recoverable DP Path 5 Node-> k Node Vel Head Friction Total UR Momentum E h 5$hkbh b[h[h 2 2 3 h [kb5 5b$hh[b b$bbbb $b$hhib 1.7883 0.1740 1.9622 3 3 4 2.1446 0.1366 0.0000 0.1366 0.2097 2.3544 0.2304 4 4 5 2.6926 0.2648 0.0000 0.2304 5 5 2.9574 0.3299 0.0000 6 3.5963 0.3555 3.9518 0.3299 ' 0.6327 0.0000 0.6327 FSPLIT SA - FLOW SPLITS by DA FARNSWORTH & JA WEINER - Gas Table CASE 9-4A GAS (AIR) FLOW ,

                      ----- INTERNAL PATH INFORMATION -----

Path Node Path Flow Input Heat In Out Velocity Path HTC Conductivity Lbm/Sec Btu /Sec Ft/Sec B/Hr-Ft^2-F B/Hr-Ft-F k k b b$hbbbbb bh 2 3 2 3 $hhhhkb b4 b$hhhhbb bh b kk[hhb bh b$ 0.70000D+02 0.60581D+03 3 4 0.23672D+03 0.39795D+02 0.137500-01 4 4 5 0.70000D+02 0.909280+03 0.283890+03 0.40572D+02 0.148 5 5 6 0.70000D+02 0.91022D+03 0.35642D+03 0.41465D+02 0.161 0.70000D+02 0.91211D+03 0.47605D+03 0.422210+02 0.173 FSPLIT 5A - FLOW SPLITS by DA FARNSWORTH & JA WEINER - Gas Table Opt - CASE 9-4A GAS (AIR) FLOW

                    ----- INTERNAL PATH INFORMATION -----

Path Node Vel Head Reynolds In Out Lbf/Sq Inch Friction Density Viscosity Number Factor Lbm/Ft^3 Lbf-Sec/Ft^2 h k b 2 2 3 b$bhkkhb bb b$khkbhb bh b$kbkkkb bk b$[hhhh 0.89413D+00 0.486790+07 0.103500-01 0.147860+00 0.35661D 0

   .                                             131

Document Number 32-1203121-01 3 3 4 4 4~ 5 0.10723D+01 0.45434D+07 0.10403D-01 0.12329D+00 0.3820 5 5 6 0.13463D+01 0.42192D+07 0.10463D-01 0.981980-01 0.41144D 0.17981D+01 0.395360+07 0.105170-01 0.735220-01 0.439080 FSPLIT SA - FLOW SPLITS by DA FARNSWORTH & JA WEIMER - Gas Table Option CASE 9-4A GAS (AIR) FLOW

                    ----- INTERNAL PATH INFORMATION -----

Path Node In Out Form Loss Path Temp Flow Area Roughness Deg F Ft^2 Hyd Dia Fric Len M inches Feet Feet 1 1 2 2 2 3 0.2000D+01 0.1070D+03 0.2000D+01 0.5000D+03 0.1596D+01 0 3 3 4 0.20000+01 0.2600D+02 0.20000+01 0.50000+03 0.1596D+01 0.30 4 4 5 0.2000D+01 0.7100D+02 0.20000+01 0.5000D+03 0.1596D+01 0.3 5 5 6 0.20000+01 0.12500+03 0.2000D+01 0.50000+03 0.15960+01 0.30 0.2000D+01 0.1790D+03 0.20000+01 0.50000+03 0.1596D+01 0.3 FSPLIT SA - FLOW SPLITS by DA FARNSWORTH & JA WEIMER - Gas Table Option CASE 9-4A GAS (AIR) FLOW

                  --------- N0DE INFORMATION ---------

l Path Inlet Rho Outlet Rho Inlet VH Outlet VH Inlet P Outlet P (Nodes) Lbm/Ft^3 lbm/Ft^3 Lbf/Sq In Lbf/Sq In PSI PSI [ [ h b$khbkbb bb b$khbhhb bb b$kkbbbbhb$khkb bk b$bbhhhb hh$bhb4 bb 2 2 3 0.15933D+00 0.13681D+00 0.82972D+00 0.96634D+00 27.5554 25.4565 3 3 4 0.13681D+00 0.110470+00 0.96634D+00 0.11967D+01 25.4565 22.8718 4 4 5 0.11047D+00 0.86598D-01 0.119670+01 0.152660+01 22.8718 19.5845 5 5 6 0.865980-01 0.612240-01 0.15266D+01 0.21593D+01 19.5845 15.0000 FSPLIT 5A - FLOW SPLITS by DA FARNSWORTH & JA WEIMER - Gas Table Option CASE 9-4A GAS (AIR) FLOW

                                                                                           )
                 --------- NODE INFORMATION ---------

Path Node Elev In Elev Out Area In Area Out Temp In Temp Out 132 l

Document Number 32-1203121-01 In Out Feet Feet Ft^2 Ft^2 Deg F Deg F 1 1 2 2 2 3 0.0000D+00 0.00000+00 0.20000+01 0.20000+01 0.2060D+03 0.8000 3 3 4 0.0000D+00 0.0000D+00 0.2000D+01 0.2000D+01 0.80000+01 0.4400 4 4 5 0.0000D+00 0.0000D+00 0.20000+01 0.2000D+01 0.44000+02 0.98000 5 5 6 0.00000+00 0.00000+00 0.20000+01 0.2000D+01 0.98000+02 0.1520D 0.0000D+00 0.00000+00 0.2000D+01 0.2000D+01 0.15200+03 0.20600 FSPLITSA-FLOWSPLITSbyDAFARNSWORTH&JdWEIMER-GasTableOption CASE 9-4A GAS (AIR) FLOW

                            ----- PATH / NODE DESCRIPTION -----

Connecting Path Node-> Node Description 1 1 2 HEAT OUT PATH 2 2 3 HEAT IN PATH 3 3 4 HEAT IN PATH 4 4 5 HEAT IN PATH 5 5 6 HEAT IN PATH FSPLIT SA - FLOW SPLITS by DA FARNSWORTH & JA WEIMER - Gas Table Opti; 5 1 1 CASE 9-4A GAS (APR) FLOW HEAT OUT PATH 1 2 0.200000D+01 0.2000000+01 0.1070000+03 0.2000000+01 0.500000D+03 0.159577D+01 0.3000000+02 0.000000D+00 HEAT IN PATH 0.000000D+00 0.2000000+01 0.200000D+01 0.206000D+03 0.80' 2 3 0.200000D+01 0.200000D+01 0.260000D+02 0.200000D+01 0.500000D+03 0.159577D+01 0.3000000+02 0.0000000+00 HEAT IN PATH 0.0000000+00 0.2000000+01 0.2000000+01 0.800000D+01 0.4400 i I 3 4 0.200000D+01 0.2000000+01 0.710000D+02 0.200000D+01 0.500000D+03 0.159577D+01 0.300000D+02 0.000000D+00 HEAT IN PATH 0.000000D+00 0.200000D+01 0.200000D+01 0.440000D+02 0.9800! 4 5 0.2000000+01 0.200000D+01 0.1250000+03 0.2000000+01  ! 0.5000000+03 0.1595770+01 0.3000000+02

                                                                                                )

' 0.000000D+00 HEAT I:1 PATH 0.0000000+00 0.2000000+01 0.200000D+01 0.980000D+02 0.152000D  ! 5 6 0.2000000+01 0.200000D+01 0.179000D+03 0.200000D+01 0.500000D+03 0.1595770+01 0.300000D+07. 0.000000D+00 6 1 0.0000000+00 70 0.2000000+01 0.200000D+01 0.152000D+03 0.206000D+0 6 1 15 0 6 i 133 l i

m Document Number 32-1203121-01 12 52 12 10 28 52 12 22 48 52 12 38' 48 20 20 38 28 20 20 22 12 20 20 10

   ..n...........................................................................

CASE 10-4A FSPLIT SA - FLOW SPLPS by DA FARNSWORTH & JA WEIMER FSPLIT is a PC based thermal hydraulics computer code designed for the generation and solution of steady state flow networks. The networks can contain any combination of up to 100 nodes and

       "  100 paths. The problem may be specified by imposing up to 10 flow boundary conditions (external paths) or one pressure drop boundary condition. Problems may be defined with either temperature inputs or heat inputs to nodes and paths. FSPLIT uses graphical on screen modeling and parameter specification to        !

greatly simplify model develcpment. In addition to individual path flow rates, FSPLIT solves for momentum, elevation, fricticn, and form loss pressure drops. The form loss pressure drop calculations can assume a constant loss factor or can calculate a flow dependent loss factor. Equivalent heat rates are calculated for temperature input problems and temperatures are calculated for heat input problems. External path input flows may ba constant or specified as a head-capacity relationship to simulate a pump. FSPLIT is designed to accommodate water (H2O), heavy water (020), any incompressible fluid, and gasses. FSPLIT has internal property routines for light water and heavy water while properties for gasses and incompressible fluids must be specified by the user. Help screens are available in both the Node Creation and Path Creation Modes. They should provide adequate knowledge to create a model and complete the analysis. If you have problems, consult the user's manual. If the manual does not answer your concerns, contact one of the authors. FSPLIT was written in a combination of Microsoft QuickBASIC and  ; 1 134

  • j
                                                                                  \

Document Number 32-1203121-01 FORTRAN. All calculations are in double (64 bit) precision. All matrix solutions are performed using the LINPACK solvers. Exec-ution of FSPLIT requires roughly 550 kilobytes of free DOS memory. The FSPLITSA.EXE, EXECHK.COM and FSP1IT.HLP files must be present. If FSPLIT is to be executed on a Hercules type monitor, the MSHERC.COM program must be run. PRIOR TO EXECUTING THIS CODE, CONTACT ONE OF THE AUTHORS TO ASSURE THAT THIS IS THE LATECT VERSION. FAILURE TO DO S0 MAY INVALIDATE YOUR ANALYSIS. USER'S MANUAL - BWNT-TM-63 (ref 32-1203121-01) FSPLIT SA FSPLIT SA - FLOW SPLITS by DA FARNSWORTH & JA WEIMER - Light Water Option Check with the authors for the latest version.

SUMMARY

of RESULTS From Input File B: CASE 10F with a Checksum of 0.592722293D+05 CASE 10-4A HEAT ADDED, PUMPS, TEE'S, UNREC & REC DPS This is a Type 1 Case. The Total Flow is Specified. Both recoverable and unrecoverable losses are considered. The system pressure ( at note 12 ) is 500 psia. There are 12 nodes connected by 13 aternal flow paths. There is I external flow path. A local convergence criterion of 0.10000-05 was used with 0.70550-06 met. Exterjal Pump head capacity war imposed. A head capacity criterion of 0.10000-03 was used with 0.95900-05 met. External heat addition was imposed. A heat addition criterien of 0.1000D-03 was used with 0.15370-06 met. Overall Convergence was met. Solution required I head capacity iterations. q

  ,     Solution required I heat addition iterations.

FSPLIT 5A - FLOW SPLITS by DA FARNSWORTH & JA WEINER - Light Water Option l CASE 10-4A HEAT ADDED, PUMPS, TEE'S, UNREC & REC DPS

                                                                                  \
                 ----- EXTERNAL PATH INFORMATION -----                             I Ext Node Node      Path Flow     Press Drop      Node IN   Node OUT    Ext Path Path IN     OUT        Lbm/Sec   Lbf/Sq inch    Press, psi Press, psi             J Tout, F j 135

Document Number 32-1203121-01 k 55 [h b.kk0h8b 05 bI[hhhhb bh b hhhkkb b5 b bb For External Path I the available NPSH is0.37239D+03 psia ( 70.0 F suction) TA8ULAR The multiplierPUMP for pumpHEAD 1 was 1 / CAPACITY DATA WAS INPUT FROM F FSPLIT SA - FLOW SPLITS by DA FARNSWORTH & JA WEIMER - Light W CASE 10-4A HEAT ADDED, PUMPS, TEE'S, UNREC & REC DPS

                      ----- INTERNAL PATH INFORMATION -----

Connecting Path Flow Press Drop Unrec DELTAP Path Node-> Node Lbm/Sec Lbf/Sq inch Recov DELTAP Lbf/Sq inch Lbf/Sq inch [ [ 5 0.fhh[hhb 05 b.hkhhhbb b1 bhkbbhbbhk 2 1 2 khhbbhbhk 0.225813D+03 .141963D+00 3 3 5 0.195169D+03 0.962083D-01 .2381720+00 4 2 0.929637D+00 0.387922D+00 4 0.225813D+03 0.541715D+00 5 4 8 0.4399280-08 0.439928D-08 0.136298D+03 0.000000D+00 6 8 0.119963D+00 0.7597080-01 7 0.136298D+03 0.439924D-01 7 7 0.7010010-01 0.701001D-01 5 0.136298D+03 0.000000D+00 8 4 0.9064960+00 0.217510D+00 6 0.8951560+02 0.6889860+00 9 5

                                                   .674133D-01    0.136762D-01 9      0.3314670+03     0.3270880-08                         .810894D-01 10        6       10       0.8951560+02 0.3270880-08     0.0000000+00 11        9                                 0.127474D+01     0.955474D+00 10       0.3314670+03                                       0.319261D+00 12      12 0.1107630+00      0.605569D+00 1      0.420983D+03                                         .4948060+00 13      10                                  0.632743D+02      0.632464D+02 11       0.420983D+03                                       0.2787300-01 0.632186D+02      0.632464D+02       .2787310-01 FSPLIT SA - FLOW SPLITS by DA FARNSWORTH & JA WEIMER - Light Wate CASE 10-4A HEAT ADDED, PUMPS, TEE'S, UNREC & REC DPS
                   ----- INTERNAL PATH INFORMATION -----
  • Connecting Unrecoverable DP Recoverable DP k I 1 3 0.0719 0.0000 0.0719 2 -0.0469 0.0000 -0.0469 1 2 0.0962 0.0000 0.0962 3 3 5 0.3879
                                                                 -0.2382      0.0000      -0.2382 0.0000       0.3879    0.5417 4       2         4       -0.0000      0.0000                            0.0000        0.5417 5       4
                                                       -0.0000    0.0000     0.0000        0.0000 8         0.0760     0.0000 6                                                    0.0760    0.0440     0.0000 8         7        0.0701      0.0000                                          0.0440 0.0701    0.0000     0.0000        0.0000  l 136

Document Number 32-1203121-01 7 7 5 0.2175 0.0000 0.2175 0.6890 0.0000 0.6890 8 4 6 0.0137 0.0000 0.0137 -0.0811 0.0000 -0.0811 9 5 9 -0.0000 0.0000 -0.0000 0.0000 0.0000 0.0000 j 10 6 10 0.9555 0.0000 0.9555 e '193 0.0000 0.3193

11. 9 10 0.6056 0.0000 0.6056 1948 0.0000 -0.4948 12 12 1 63.2464 0.0000 63.2464 0.0279 0.0000 0.0279 13 10 11 63.2464 0.0000 63.2464 -0.0279 0.0000 -0.0279 i

FSPLIT 5A - FLOW SPLITS by DA FARNSWORTH & JA WEIMER - Light Water Option CASE 10-4A HEAT ADDED, PUMPS, TEE'S, UNREC & REC DPS I

                     - --- INTERNAL PATH INFORMATION -----

Path Node Path Flow Input Heat Velocity Path HTC Conductivity In Out Lbm/Sec Btu /Sec Ft/Sec B/Hr-Ft^2-F B/Hr-Ft-F i 1 1 3 0.195170+03 0.00000D+00 0.34121D+01 0.000000+00 0.395780+00 2 1 2 0.22581D+03 0.00000D+00 0.394780+01 0.000000+00 0.39578D+00 ] 3 3 5 0.195170+03 0.00000D+00 0.68242D+01 0.00000D+00 0.395780+00 4 2 4 0.22581D+03 0.00000D+00 0.394780+01 0.00000D+00 0.395780+00 1 5 4 8 0.13630D+03 0.000000+00 0.476570+01 0.00000D+00 0.39578D+00 6 8 / 0.13630D+03 0.00000D+00 0.476570+01 0.000000+00 0.39578D+00 - 7 7 5 0.13630D+03 0.00000D+00 0.47657D+01 0.000000+00 0.395780+00 8 4 6 0.895160+02 0.00000D+00 0.15650D401 0.00000D+00 0.395780+00 9 5 9 0.331470+03 0.00000D+00 0.11590D+02 0.00000D+00 0.39578D+00 10 6 10 0.89516D+02 0.000000+00 0.15650D+01 0.00000D+00 0.395780+00 11 9 10 0.331470+03 0.000000+00 0.579490+01 0.000000+00 0.395780+00 12 12 1 0.42098D+03 0.100000+06 0.69605D+01 0.000000+00 0.38864D+00 13 10 11 0.420980+03 .10000D+06 0.696050+01 0.00000D+00 0.38864D+00 FSPLIT SA - FLOW SPLITS by DA FARNSWORTH & JA WEIMER - Light Water Option CASE 10-4A HEAT ADDED, PUMPS, TEE'S, UNREC & REC DPS

                     ----- INTERNAL PATH INFORMATION -----

Path Node. Vel Head Reynolds Friction Density Viscosity In Out Lbf/Sq Inch Number Factor Lbm/Ft^3 Lbf-Sec/Ft^2 [ [ 5 b[f[b6bb b[ b.[5bbbb bh b bbbbOb bb b bhhbbb bh b.5f4b5b bb ] 2 1 2 0.962080-01 0.21123D+07 0.000000+00 0.572000+02 0.374930-05 3 3 5 0.28747D+00 0.258180+07 0.000000+00 0.572000+02 0.37493D-05 4 2 4 0.962080-01 0.21123D+07 0.00000D+00 0.57200D+02 0.374930-05 5 4 8 0.140200+00 0.254990+07 0.00000D+00 0.57200D+02 0.374930-05 6 8 7 0.140200+00 0.25499D+07 0.000000+00 0.57200D+02 0.37493D-05 137 I

Document Number 32-1203121-01 7 7 5 8 4 6 0.14020D+00 0.18030D+07 0.000000+00 0.572000+02 0.3 9 5 9 0.15119D 0.837340+06 0.00000D+00 0.572000+02 0.3 10 6 10 0.82919D+00 0.438490+07 0.000000+00 0.57200D+02 0.3 11 9 10 0.151190-01 0.83734D+06 0.000000+00 0.572000+02 0.37 12 12 1 0.207300+00 0.21924D+07 0.000000+00 0.572000+02 0.37 13 10 11 0.31623D+00 0.215590+07 0.00000D+00 0.604820+02 0.68 0.31623D+00 0.215590+07 0.00000D+00 0.60482D+02 0.6

    'FSPLIT SA - FLOW SPLITS by DA FARNSWORTH & JA WEIMER - Light W CASE 10-4A HEAT ADDED, PUMPS, TEE'S, UNREC & REC DPS
                          .-..- INTERNAL PATH INFORMATION -----

Path Node In Out Form Loss Path Heat Flow Area Roughness Stu/Sec Ft^2 Hyd Dia Fric Len M inches Feet Feet j 1 1 3 2 1 2 0.1000D+01 0.0000D+00 0'.10000+01 0.0000D+00 0.7979D+ 3 3 5 0.10000+01 0.00000+00 0.1000D+01 0.0000D+00 0.11280+0 4 2 4 0.1349D+01 0.0000D+00 0.50000+00 0.0000D+00 0.79790+ 5 4 8 0.00000+00 0.0000D+00 0.10000+01 0.00000+00 0.11280+0 6 8 7 0.54190+00 0.00000+00 0.50000+00 0.00000+00 0.11280+01 7 7 -5 0.5000D+00 0.00000+00 0.5000D+00 0.00000+00 0.11280+0 8 4 6 0.1551D+01 0.0000D+00 0.50000+00 0.0000D+00 0.79790+0

     ~9     5     9 0.9046D+00 0.0000D+00 0.10000+01 0.0000D+00 0.11280+0 10     6 10 0.00000+00 0.0000D+00 0.5000D+00 0.0000D+00 0.79790+00 11     9 10 0.63200+02 0.0000D+00 0.10000+01 0.00000+00 0.11280+01 12 12 1 0.29210+01 0.00000+00 0.10000+01 0.00000+00 0.7979D+00 13 10 11 0.20000+03 0.10000+06 0.1000D+01 0.00000+00 0.11280+01 0.2000D+03 .10000+06 0.1000D+01 0.0000D+00 0.11280+010 FSPLIT SA -' FLOW SPLITS by DA FARNSWORTH & JA WEIMER - Light Wate                                 '

CASE 10-4A HEAT ADDED, PUMPS, TEE'S, UNREC & REC DPS

                       ........- NODE INFORMATION ---------

Path Inlet Rho Outlet Rho Inlet VH Outlet VH Inlet P Outlet P (Nodes) Lbm/Ft^3 Lbm/Ft^3 Lbf/Sq In

 ........        ........... ...........                      Lbf
                                              ........... ..../Sq In          PSI         PSI 11 3                                                                    ........   ........

2 1 2 0.57200D+02 0.572000+02 0.33438D+00 0.297470+00 436.7257 436.7007 3 3. 5 0.57200D+02 0.57200D+02 0.334380+00 0.962080-01 436.7257 436.8676 4 2 4 0.57200D+02 0.572000+02 0.287470+00 0.829190+00 436.7007 435.7711 5 4 8 0.572000+02 0.572000+02 0.962080-01 0.962080-01 436.8676 436 0.572000+02 0.57200D+02 0.962080-01 0.14020D+00 436.8676 436.7477 138 i

Document Number 32-1203121-01 6 8 7 0.57200D+02 0.57200D+02 0.14020D+00 0.140200+00 436.7477 436.6776 7 7 5 0.57200D+02 0.57200D+02 0.14020D+00 0.829190+00 436.6776 435.7711 8 4 6 0.57200D+02 0.57200D+02 0.962080-01 0.151190-01 436.8676 436.9351 9 5 9 0.57200D+02 0.57200D+02 0.82919D+00 0.829190+00 435.7711 435.7711

10. 6 10 0.57200D+02 0.57200D+02 0.151190-01 0.334380+00 436.9351 435.6603 11 9 10 0.572000+02 0.572000+02 0.829190+00 0.334380+00 435.7711 435.6603 12 12 1 0.62401D+02 0.572000+02 0.30651D+00 0.334380+00 500.0000 436.7257 13 10 11 v.57200D+02 0.62401D+02 0.33438D+00 0.30651D+00 435.6603 372.4418 FSPLIT 5A - FLOW SPLITS by DA FARNSWORTH & JA WEIMER - Light Water Option 1

CASE 10-4A HEAT ADDED, PUMPS, TEE'S, UNREC & REC DPS I

                      --------- N0DE INFORMATION ---------

Path Node Elev In Elev Out Area In Area Out Temp In Temp Out In Out Feet Feet Ft^2 Ft^2 Deg F Deg F 1 1 3 0.0000D+00 0.0000D+00 0.10000+01 0.5000D+00 0.3063D+03 0.3063D+03 2 1 2 0.00000+00 0.0000D+00 0.10000+01 0.10000+01 0.3063D+03 0.3063D+03 3 3 5 0.0000D+00 0.0000D+00 0.50000+00 0.50000+00 0.3063D+03 0.3063D+03 4 2 4 0.00000+00 0.00000+00 0.10000+01 0.1000D+01 0.3063D+03 0.3063D+03

     -5    4    8 0.00000+00 0.0000D+00 0.1000D+01 0.50000+00 0.3063D+03 0.3063D+03 6   8    7 0.0000D+00 0.0000D+00 0.5000D+00 0.5000D+00 0.3063D+03 0.3063D+03 7   7    5 0.0000D+00 0.00000+00 0.50000+00 0.5000D+00 0.3063D+03 0.3063D+03 8   4    6 0.00000+00 0.00000+00 0.10000+01 0.10000+01 0.3063D+03 0.3063D+03 9   5    9 0.0000D+00 0.0000D+00 0.5000D+00 0.50000+00 0.3063D+03 0.3063D+03 10    6 10 0.0000D+00 0.00000+00 0.1000D+01 0.10000+01 0.3063D+03 0.3063D+03 11    9 10 0.0000D+00 0.0000D+00 0.50000+00 0.1000D+01 0.3063D+03 0.3063D+03 12- 12     1 0.00000+00 0.0000D+00 0.10000+01 0.1000D+01 0.7000D+02 0.3063D+03 113~ 10 11 0.0000D+00 0.0000D+N 0.1000D+010.1000D+010.3063D+03 0.70000+02 FSPLIT 5A - FLOW SPLITS by DA FARNSWORTH & JA WEIMER - Light Water Option           I CASE 10-4A HEAT ADDED, PUMPS, TEE'S, UNREC & REC DPS                  !
                      ----- PATH / NODE DESCRIPTION -----                               l Connecting                                                                     l Path Node-> Node Description                                                         i
                        ~
   "i"    "5" "3             RIGHT PATH #1                                              !

2 1 2 LEFT PATH #2 3 3 5 TOP OF CONVERGING TEE #3 4 2 4 . TOP OF DIVERGING TEE (LEFT #4) 5 4= 8 MIDDLE LEG OF DIVERGING TEE (CROSS CONCT #5) 6 8 7 CROSS CONNECT #6 7 7 5 MIDDLE LEG OF CONVERGING TEE'(CROSS CONCT #7) 139

Document Number 32-1203121-01 8 .4 6 DIVERGING TOP OF TEE-(LEFT # 8)

        '9        5      9    CONVERGING TOP OF TEE (RIGHT # 9) 10        6     10    LEFT #10 11        9     10    RIGHT #11 12      12'      1    PATH 12~

13 10 11 PATH 13 FSPLIT 5A'- FLOW SPLITS by DA FARNSWORTH & JA WEIMER - Light Water Option

13. 1 -1 CASE 10-4A HEAT ADDED,- PUMPS, TEE'S, UNREC-& REC DPS RIGHT. PATH #1 1 3 0.1000000+01 0.100000D+01 0.306293D+03 0.100000D+01 0.0000000+00 0.797885D+00 0.000000D+00 0.000000D+00 0.000000D+00 0.000000D+00 0.100000D+01 0.5000000+00 LEFT PATH #2 1 2 0.100000D+01 0.1000000+01 0.306293D+03 0.1000000+01 0.000000D+00 0.112838D+01 0.0000000+00 0.000000D+00 0.000000D+00 0.000000D+00 0.1000000+01 0.100000D+01 TOP OF CDNVERGING TEE #3 3 5 0.0000000+00 0.000000D+00 0.306293D+03. .500000D+00 7 9 9 11 0.000000D+00 0.000000D+00.

0.100000D+00 0.1600000+00 - 0.2000000+00 0.270000D+00 0.300000D+00 0.3800000+00 0.400000D+00 0.4600000+00 0.500000D+00 0.530000D+00 0.6000000+00 0.570000D+00 0.7000000+00 0.590000D+00 0.800000D+00 0.6000000+00 0.900000D+00 0.590000D+00 0.100000D+01 0.550000D+00 0.0000000+00 0.7978850+00 0.0000000+00 0.0000000+00 0.000000D+00 0.0000000+00 0.500000D+00 0.5000000+00 l TOP OF DIVERGING TEE (LEFT #4) 2 .4 0.000000D+00 0.000000D+00 0.306293D+03 0.100000D+01 0.0000000+00 0.1128380+01 0.000000D+00 0.000000D+00 0.000000D+00 0.000000D+00 0.100000D+01 0.1000000+01 MIDDLE LEG OF DIVERGING TEE (CROSS CONCT #5) 4 .8 0.000000D+00 0.0000000+00 0.306293D+03 .500000D+00 5 4- 4 12 0.000000D+00 0.550000D+00 0.5000000-01 0.550000D+00  ; I 0.100000D+00 0.5555000+00 0.200000D+00 0.5775000+00 l 0.300000D+00 0.610500D+00, i 0.4000000+00 0.6545000+00 0.500000D+00 0.715000D+00

         .                                      140

1 i l Document Number 32-1203121-01 0.6000000+00 0.786500D+00 0.7000000+00 0.874500D+00 0.800000D+00 0.973500D+00 0.100000D+01 0.121000D+01 0.130000D+01 0.220500D+01 0.000000D+00 0.1128380+01 0.0000000+00 0.0000000+00 0.0000000+00 0.000000D+00 0.1000000+01 0.500000D+00 CROSS CONNECT #6 8 7 0.500000D+00 0.100000D+01 0.306293D+03 0.5000000+00 0.0000000+00 0.1128383+01 0.0000000+00 0.000000D+00 0.0000000+00 0.ft'0000+00 0.5000000+00 0.500000D+00 MIDDLE 7 LEG OF ( JERGING TEE (CROSS CONCT #7) 5 0.0000000+00 0.0000000+00 0.306293D+03 .500000D+00 7 9 9 11 0.000000C+00 .550000D+00 - 0.1000000+00 .335500D+00 0.2000000+00 .1650000+00 ' O.3000000+00 .6050000-01 0.4000000+00 0.242000D+00 0.500000D+00 0.4235000+00 1 0.6000000+00 0.572000D+00  !

        '0.700000D+00 0.715000D+00 0.800000D+00 0.858000D+00 0.9000000+00 0.9900000+00                                                        )

0.100000D+01 0.1100000+01 0.000000D+00 0.7978850+00 0.000000D+00 0.000000D+00 0.000000D+00 0.000000D+00 0.500000D+00 0.500000D+00 DIVERGING TOP OF TEE (LEFT # 8) 4 6 0.0000000+00 0.0000000+00 0.306293D+03 .100000C+01 8 4 4 9 0.0000000+00 0.400000D+00 0.100000D+00 0.3200000+00 0.200000D+00 0.2600000+00 0.3000000+00 0.2000000+00 0.400000D+00 0.1400000+00 0.5000000+00 0.1000000+00 0.600000D+00 0.6000000-01 0.8000000+00 0.2000000-01 0.100000D+01 0.000000D+00 0.000000D+00 0.1128380+01 0.0000000+00 0.0000000+00 0.0000000+00 0.000000D+00 0.1000000+01 0.1000000+01 CONVERGING TOP OF TEE (RIGHT # 9) .. 5 9 0.0000000+00 0.0000000+00 0.306293D+03 0.500000D+00 0.000000D+00 0.797885D+00 0.0000000+00 0.000000D+00 0.000000D+00 0.000000D+00 0.5000000+00 0.5000000+00 LETT #10' 10 0.3000000+02 0.300000D+02 0.306293D+03 .1000000+01

       -10                  13            13              11 0.0000000+00 0.200000D+01 0.100000D+00 0.173000D+01 141
                                                                                       )

i Document Nunser 32-1203121-01 0.200000D+00 0.1520000+01 0.3000000+00 0.1370000+01 0.4000000+00 0.128000D+01 0.5000000+00 0.1250000+01 0.600000D+00 0.1280000+01-

          ' 0.700000D+00 0.137000D+01 0.8000000+00 0.152000D+01 0.900000D+00 0.173000D+01 0.1000000+01 0.200000D+01 0.000000D+00 0.1128380+01 0.0000000+00 0.000000D+00 0.000000D+00 0.000000D+00 0.100000D+01 0.1000000+01 RIGHT #11    ~

9 10 0.5000000+00 0.5000000+00 0.306293D+03 .100000D+01 11 13 13 11 j 0.0000000+00 0.200000D+01 ' O.1000000+00 0.1730000+01 0.2000000+00 0.1520000+01 { 0.3000000+00 0.137000D+01 1' 0.4000010+00 0.128000D+01 O.5000000+00 0.1250000+01 0.6000000+00 0.1280000+01 0.700000D+00 0.1370000+01 0.8000000+00 0.1520000+01

                                                                                       ]

0.9000000+00 0.1730000+01 0.1000000+01 0.2000000+01- - 0.0000000+00 0.797885D+00 0.000000D+00 0.000000D+00 0.000000D+00 0.000000D+00 0.500000D+00 0.1000000+01 PATH 12 12 1 0.200000D+03 0.0000000+00 0.188146D+03 0.100000D+01 0.00L: 00D+00 0.112838D+01 0.000000D+00 0.1000000+06

         - 0.0000000+00 0.0000000+00 0.1000000+01 0.1000000+01 PATH 13 3

10 11 0.2000000+03 0.000000D+00 0.1881460+03 0.100000D+01 0.000000D+00 0.1128380+01 0.0000000+00 .1000000+06 0.000000D+00 0.000000D+00 0.1000000+01 0.100000D+01 11 12 420.9760582891602 0.700000D+02 12- 1 500 0 12 22 52 12 18 22 68 8 18 22 36 16 18 32 68 8 26 j

     .       32             36            16            26                              i 42            ~68           8              34                              l 13 2            44            14            26 32             60            10            26 42             36'           16           -34 42             52            12            34 62             52            12            50 2              52            12            2
               .                                  142 L}}