TAK 2000
Getting Started Guide
(Demonstration Version)
TAK 2000 Getting Started Guide
Copyright © 2000 K & K Associates, All Rights Reserved
10141 Nelson St., Westminster, CO 80021
This document is being furnished by K & K Associates for information purposes only to users of the TAK 2000 software
product and is furnished on an “AS IS” basis, that is, without any warranties, whatsoever, expressed or implied.
TAK 2000 and PCBUILD are trademarks of K & K Associates.
Information in this document is subject to change without notice and does not represent a commitment on the part of K & K
Associates. The software described in this document is furnished under a license agreement. The software may be used only
in accordance with the terms of that license agreement. It is against the law to copy or use the software except as specifically
allowed in the license.
No part of this document may be reproduced or retransmitted in any form or by any means, whether electronically or
mechanically, including, but not limited to the way of: photocopying, recording, or information recording and retrieval systems,
without the express written permission of K & K Associates.
TAK 2000 Getting Started Guide
Initial Publishing, March 13, 2000
Latest Revision: June 14, 2000
Visit our Web site at http://www.tak2000.com
Table of Contents
Installation ............................................................................................................................... 1
Contents of Package (Not Applicable to Demo) ................................................................... 1
Registering TAK 2000 (Not Applicable to Demo) ................................................................ 1
Technical Support (Not Applicable to Demo)....................................................................... 1
System Requirements.............................................................................................................. 2
Installing TAK 2000 ................................................................................................................ 2
To install the TAK 2000 Demo............................................................................................... 2
Starting TAK 2000 .................................................................................................................. 3
The TAK 2000 WorkBench Screen ....................................................................................... 4
Getting Started – Your First Model ...................................................................................... 4
Creating the Model ................................................................................................................ 5
The Model............................................................................................................................ 17
The Plotter ............................................................................................................................. 28
Installation
This section tells you how to install TAK 2000 on your computer. When you finish
installing the software, take a few moments to examine the TAK 2000 information file
called INFOLIST.TXT. This file is located in the main TAK 2000 directory. It contains
important information that was compiled after this booklet went to press.
Contents of Package (Not Applicable to Demo)
TAK 2000 comes with everything you need to install and use the software. The
package includes the following items:
•
TAK 2000 Installation CD
•
TAK 2000 Getting Started Guide (this booklet)
•
Registration Form
•
Technical Support Form
Registering TAK 2000 (Not Applicable to Demo)
There are two ways to register your new software with K & K Associates.
•
Fill out the registration form included with your software and drop it in the postal
mail
•
Register online after installation. Simply click on the Help menu item and then
click on Register.
Technical Support (Not Applicable to Demo)
If you have a software problem that you cannot solve, refer to the Technical Support
form included with your software.
1
System Requirements
In order to run TAK 2000, you must have a Pentium computer or compatible. In
addition, your system must include the following:
•
A 133 Mhz Pentium or faster CPU
•
Microsoft Windows NT 4.0, Windows 9, or Windows 98. TAK 2000 is not
compatible with earlier versions of Windows.
•
A hard drive with at least 20 megabytes of free space. This is the space required
to install TAK 2000.
•
32 megabytes of RAM
•
A mouse
•
A 800 x 600 (16 bit color) display adapter and color monitor. A 17 inch or larger
monitor is recommended.
•
A CD-ROM drive
Installing TAK 2000
Before installing TAK 2000 you must have WIN 95/98/NT installed on your system.
In addition, if you want to take advantage of TAK 2000 advanced user logic feature
you must install Compaq Visual Fortran v5.0 or higher on your computer BEFORE
you install TAK 2000.
To install the TAK 2000 Demo
1. Shut down ALL other applications. Failure to do so may preclude some modules
from being installed.
2
2. Double click on the Tak2000Demo.Exe file with the Windows Explorer. You may
also start the installation by selecting RUN from the START menu. The Run
dialog box will appear as shown below.
3. Type in C:\Tak2000Demo.exe ( You may need to change the drive and
directory to where the Tak2000Demo.exe file is located on your system). Click on
the OK button.
4. Follow the installation prompts that appear.
5. Once installation has been completed a folder call “TAK 2000 Demo” is created
under C:\Program Files (if you did not override the installation default location).
This contains all of the TAK 2000 program files. You will have two folders
installed directly under “TAK 2000 Demo” folder. The first is the “Documents”
folder. It contains a .PDF version of the user manual and other text files. The
other is the “Samples” folder. This folder contains several sample models that
demonstrate various features of TAK 2000. These models are ready to run.
Starting TAK 2000
This section tells you how to start TAK 2000 and gives you a quick look at the TAK
2000 WorkBench user interface. For detailed information on a specific screen
component or feature, refer to the online Help.
To start the TAK 2000 Demo, click on the START button and then on PROGRAMS.
Find the TAK 2000 DEMO menu item and click on it.
3
The TAK 2000 WorkBench Screen
The startup title screen appears for a few moments followed by the TAK 2000
WorkBench screen. TAK 2000 WorkBench has a Windows Explorer type of
interface. This feature is very useful when dealing with the numerous files created
during the course of a thermal analysis.
Tool Bar
Title Bar
File Pane
DirectoryPanel
Getting Started – Your First Model
The best way to become familiar with TAK 2000 is to actually use it. The following
example will take you through, step by step, the process of building a thermal model,
running the model through the solver and interpreting the results. You will gain
experience on model building techniques, the different analysis options and
4
exposure to some of the post processing tools available to help you understand the
results of your analysis.
In this example, we will be building a simple model of a square aluminum bar. On
one end, the bar is fixed into a wall. We will assume the wall acts as an infinite heat
sink and its temperature will be constant. On the opposite end, the bar is being
heated at a constant rate for 1 hour. The objective of this analysis is to determine
the temperature distribution along the length of the bar as it is being heat and also
as it cools down.
To keep this first model simple we will assume the sides of the bar are perfectly
insulated. With this assumption we can ignore convection and radiation to the
surrounding environment. We will only be modeling the conduction down the length
of the bar.
Creating the Model
Before we can begin constructing the mathematical model we must first must collect
all the pertinent data about the item and process being model.
The bar is made out of 6061 aluminum and its dimensions are 2.5 cm x 2.5 cm x 25
cm. The thermophysical properties of this type of aluminum are,
k = 1.712 Watts/cm-°°C (thermal conductivity)
Cp = 0.962 Joule/Gram-°°C ( heat capacity )
ρ = 2.718 grams/cm3 ( density )
We will assume that initially the bar and the wall is at 20°C (room temperature). One
end of the bar is being heated at a rate of 100 watts
In order to develop a math model (thermal network) of the bar and use numerical
techniques to solve for temperatures and heat transfer rates, it is necessary to
subdivide the thermal system into a number of finite sub-volumes called nodes.
In this example, we are only interested in a rough distribution of temperature along
the length so we will only divide the bar into 5 discrete nodes. Each node will be 5
5
cm long. To identify a node we need to assign it an integer number. We are free to
assign it any number from 1 to 99999. We will arbitrarily assign the numbers 1
through 5 to these five nodes. These five nodes are called diffusion nodes because
they have a finite volume or mass associated with them.
We will also place a node at the face of the bar that is being heated. Again, we will
arbitrarily assign this node the number 10. This node will be mass-less and will
represent the surface of the end of the bar. Mass-less nodes are usually called
“arithmetic” nodes. Arithmetic nodes have zero capacitance and physically they are
an unreal quantity. However, are useful in representing surface temperature,
bondline temperatures and interfaces.
The face of the other end of the bar will be fixed at a constant temperature by using
what is called a boundary node. A boundary node can be considered an infinite sink
and is used to represent constant temperature sources/sinks within a thermal
network. We arbitrarily assign this node the number 20. The one-dimensional
model that we’ve conceptually created is shown schematically in Figure 1.
Note that in TAK 2000, each node can be assigned an alphanumeric “name” in
addition to its integer node number. This is use full in annotating the printed and
plotted results.
20C
20
1
2
3
4
100W
5
10
Figure 1 Model Schematic
Now that we have determine the composition of the mathematical network we can
begin computing the thermal mass of each node and the thermal conductance’s
between them.
6
As mentioned above each of the bar nodes are diffusion nodes and have a thermal
mass. The thermal mass of each node is computed from the following equation.
MCp =V • ρ • Cp
= (2.5cm) • (2.5cm) • (5cm) • (2.71 grams/cm3) • (0.962 Joule/Gram-°°C)
= 81.47 J/°°C
Where,
V – Node volume
ρ – Density
Cp – Specific Heat
Node 10 is an arithmetic node that represent the surface so this node has no mass.
To indicate an arithmetic node input a zero or negative value for the thermal mass.
The most common convention is to input a value of –1.0.
Node 20 is a boundary node. Mass is meaningless for this type of node. Although,
not used, you still must enter a dummy value for the thermal mass. The most
common convention is to enter a value of 1.
The conductance between each of the bar nodes will be computed from the simple
conduction equation shown below.
G =ka/L
= (1.712 w/cm-°°C) • (2.5cm) • (2.5cm) / (5cm)
= 0.856 w/°°C
Where,
K-
thermal conductivity
a - cross sectional area of the flow path between each node
L - distance between the node centers.
Now we need to calculate the conductance’s to the ends of the bar. The
conductance from node 1 and node 20 is the same as that between node 5 and
7
node 10. Since this conductance is from the node center to the end of the bar the
distance is 2.5 cm instead of 5 cm.
G = (1.712 w/cm-°°C) • (2.5cm) • (2.5cm) / (2.5cm)
= 4.28 w/°°C
We are now ready to create the mathematical model that can be read and solved by
the TAK 2000 analyzer (solver).
TAK 2000 math models are text files that can be generated and edited with any text
editor. The best way to create a new model is to use TAK 2000’s Model Wizard
function. This function will generate the framework of the model. Once the
framework is built, you add the model’s nodes, conductors, sources, etc with the
editor.
To start the Model Wizard, click on the FILE menu item and then click on MODEL
WIZARD. This will bring up a tabbed dialog box as shown below.
Before we begin using the Model Wizard, please keep in mind that any selections
you make using the Wizard can be changed by simply editing the model that has
been created. For example, if you were to select a transient solution type in the
Wizard you could still run a steady state solution by simply changing the solution
8
option in the OPTIONS DATA block of the model. You would not rerun the Wizard
and create a new model of the same system.
Model Name
The default name of a new model is “Model.tak”. We will want to change this name
to something more descriptive, so we will call it “Bar.tak”.
Model Units
We have the option of working in either SI or English units. We are working this
sample in SI so we do not need to change the selected units.
Solution Type
We are interested in calculating the temperature distribution down the length of the
bar as a function of time. Therefore, we would like the transient solver to be used.
We have the option of three different transient solvers. For this exercise we will
select the forward differencing solver.
Data Modules
Once a model framework is created with the Model Wizard, the model data such as
nodes, conductors, etc is inserted by the user with the editor. In this sample
exercise, besides defining nodes and conductors we will be defining a source that
represents the heat input into the end of the bar. Therefore, we will select SOURCE
DATA.
We will not be doing any advanced modeling in which psuedo Fortran code is
involved so we will leave the INCLUDE USER LOGIC check box unchecked. The
screen should look as follows.
9
We are now finished with the GENERAL model data so click on the OPTIONS DATA
tab. The following screen should appear.
10
The OPTIONS DATA tab allows you to select various input/output features. The only
default I/O option is the impressed heating rate option QPRINT. This option specifies
that the impressed heating rate for all nodes will be printed at each output interval.
We will leave this option selected because we want to verify in the output file that the
heat rates were applied correctly.
We are also interested in plotting the impressed heat rate as a function of time. To
do this we select the PLOTQ option. This option will cause TAK 2000 to generate a
heat rate plot file that can be read by TAK 2000’s X/Y plotter. After selecting the
PLOTQ feature the screen should now look like this.
We are now finished with the OPTIONS DATA so click on the CONTROL DATA tab.
This tab allow us to specifies values for the various model control parameters. This
tab is shown below.
11
ABSZRO
The ABSZRO program constant is the temperature of Absolute Zero in units
consistent with the rest of the model. For example if we modeling in Celsius we
would enter –273.15. If we were modeling in Kelvin, ABSZRO would be 0.0. If SI
units are specified on the GENERAL tab, the Model Wizard defaults to units of
Celsius so the default value in ABSZRO is –273.15. Since we are modeling in units
of Celsius we don’t need to change this value.
SIGMA
The SIGMA parameter is the Stefan-Boltzmann constant. The default value is 5.67E08 W/m2-K4 if SI units are selected on the GENERAL tab. This value is only used if
you include radiation conductors in your model. Since we are not modeling radiation
in our bar model, we can ignore this value.
IFC
12
The IFC parameter allows you to print out your temperature in units other than what
was used to built the model. For example, in our model we are using degrees
Celsius. If however, we would like to print the temperature results in some other
units, such as degrees Fahrenheit, we would set this value to 1. The default is 0
which indicate no conversion is to be done. We don’t want to do any conversion so
we’ll leave this at 0.
The steady state parameters are grayed out because we have indicated that we are
going to use one of the transient network solvers. For your convenience these
parameters will still be written to the output file with the default values but, they will
be “commented” out so TAK 2000 will ignore them. This will allow you to easily set
up your model to run a steady state solution later if you would like.
Note: During a transient solution all steady state control parameters are ignored if
defined in the model. Likewise, during a steady state solution all transient
parameters are ignored. Therefore, its perfectly fine to have both sets of parameters
defined in your model in the Control Data block.
TIMEO
The TIMEO parameter indicates the initial or “start” time of the simulation. Usually
this will be set to zero. However, if you were restarting a solution to continue it out
for a longer period of time, you would set this value to the stop time from the
previous analysis. In this example, we want our initial time to be zero.
IMPORTANT:
The units for this and all other parameters must be consistent
throughout the model. Therefore, since we selected units of J/gram
for thermal mass of nodes then the time units must be in seconds.
TIMEND
TIMEND is the "end" or stop time of a transient simulation (solution). This parameter
does not have a default value. It must be specified for a transient solution.
13
In this example we will be applying the heating to the end of the bar for 1 hour and
then we want to see how much the bar cools down for an additional hour. Therefore
the total simulation time will be 2 hours..
We need to be careful that the units of time that we use here is dimensionally
consistent with the rest of the model. For example, the product of the heat rate and
time units must match the units of energy that we used when calculating the thermal
masses.
The units of specific heat that we used is J/kg-°C. Therefore, the energy unit that
we are using is Joules. Since we want to model the heat source in units of Watts
(Joules/second) all parameters having to do with time must be specified in units of
seconds.
Since we want to simulate 2 hours of time we will enter 7200. seconds for the
simulation time.
HINT:
If you would like your units of time to be in hours instead of seconds,
convert your units of specific heat from J/kg-°°C to (W-Hr)/Kg-°°C.
PLOT
The PLOT parameter indicates the frequency that temperature and heat rate data is
written to the transient temperature and heat rate plot files. The transient
temperature file will have a .PLT extension and the heat rate file will have a .PLQ
extension. In this case we want data every 120 seconds so enter 120. for this
parameter. Note that a transient heat rate file will only be created if the PLOTQ
option is set to ON in the OPTIONS DATA block.
14
SCALE
The SCALE parameter is a conversion factor that is applied to the time units written
to the temperature and heat rate files (the .PLT and .PLQ files). This allows the user
to change the time units in these plot files.
For example, in this example the model time units is seconds. If we want the output
time units in the plot file to be minutes we would enter 0.01667 which is the
conversion factor from seconds to minutes. If we want hours we would enter
2.778E-04. Since we would like units of minutes we will enter 0.01667 for this value.
TRCRIT
The TRCRIT parameter is the iterative temperature convergence criteria used during
a transient solution routine. When using an implicit routine such as FWDBCK or
FWDWRD, it is used as the convergence criteria for both the diffusion and arithmetic
nodes, since both are solved iteratively at each time step.
In the explicit routine FWDWRD, it is applicable only to the arithmetic nodes
(diffusion nodes are not solved iteratively).
When the maximum change in temperature of all arithmetic nodes (and diffusion, if
applicable), is less than the value of TRCRIT, the solution is considered converged
and the solution continues on to the next computational time. If the convergence
criteria is not met within a certain number of loops, defined by the NLOOPT
parameter, a warning message is printed and the solution continues onto the next
computational time. The default value of TRCRIT is 0.01. This is usually sufficient
form most models so that’s what we will leave it at for this model.
15
NLOOPT
The NLOOPT parameter is the maximum number of iterations to perform at each
calculation interval, during a transient solution. If an implicit solution routine
(FWDBCK or BCKWRD) is being used, it applies to both the diffusion and arithmetic
nodes because both type of nodes are receiving an iterative solution. If the explicit
solution routine FWDWRD is being used, NLOOPT only applies to the arithmetic
nodes. The diffusion node temperatures are not iterated.
During each time step, the diffusion node temperatures are solved first. After the
temperatures for the diffusion nodes are solved, the temperature of the arithmetic
nodes are calculated with a "steady state" technique. If the number of iterations
performed exceeds NLOOPT, a warning message is issued and the run continues.
The default value of NLOOPT is 50. This is usually sufficient and that is what we will
leave it at for this model.
OUTPUT
The OUTPUT parameter specifies the interval at which data are written to the output
file (.OUT) during a transient solution. Like TIMEO and TIMEND, the units of
OUTPUT must be consistent with the rest of the model. Therefore, the units will be
in seconds. Since we would like the temperature printed to the output file every 20
minutes (simulation time) we will enter 1200. Note: OUTPUT has no default and
must be specified for a transient solution.
16
That’s it! Your Control Tab should look like the following.
You are now ready to actually create a model file. To do this simply click on the
CREATE MODEL button. Once you do this TAK 2000 will create the model file
called BAR.TAK. After creating the model the Wizard will automatically launch the
editor .
The Model
The model framework that TAK 2000 created should look like the following.
C-------------------------------------------------------------------------C TAK 2000 Version 1.0, Model Template
C-------------------------------------------------------------------------HEADER OPTIONS DATA
UID
SOLRTN
Logic
= SI
$ Model Units
= FORWARD$ Solution routine
= Off
$ User logic
QPrint
GPrint
= On
= Off
$ Print impressed heat rates
$ Print conductor values
17
CPrint
CSGDump
NCVPrint
HPrint
LPrint
=
=
=
=
=
Off
Off
Off
Off
Off
$
$
$
$
$
Print
Print
Print
Print
Print
QDump
PlotQ
LodTmp
AutoDamp
=
=
=
=
Off
On
Off
On
$
$
$
$
Print network map
Create .PLQ plot file of heat rates (SS Only)
Load initial temperatures from .INI file
Automatic solution damping function (SS Only)
Dictionaries = Off
nodal thermal masses
noal CSG values
NCV conductor status
heater status
thermal louver status
$ Print actual/relative data dictionaries
HEADER CONTROL DATA
C
-------------------
ABSZRO
SIGMA
IFC
= -273.15
= 5.67E-09
= 0
Program Control Constants ------------------$ Absolute Zero
$ Stefan-Boltzmann Constant
$ Output Units Flag
C Steady State constants
C
C
C
SSCRIT
NLOOPS
DAMP
= 0.001
= 500
= 1.0
$ Steady state convergence criteria
$ Max number steady state iterations
$ Steady state damping factor
C Transient constants
TIMEO
TIMEND
PLOT
SCALE
TRCRIT
NLOOPT
OUTPUT
=
=
=
=
=
=
=
0.0
7200.
120.
0.01667
0.01
50
1200.
$
$
$
$
$
$
$
Transient start time
Transient stop time
Plot file output interval
Plot file time scale factor
Transient convergence criteria
Max number of iterations per step
Transient output interval
C
----------------------- User Defined Constants ---------------------
C
*** Put your own user constants here ***
HEADER NODE DATA
C *** This is where you define your nodes ***
C Example Format:
Node, Temp, MCp
$ Node Name
HEADER CONDUCTOR DATA
C *** This is where you define your conductors ***
C
Example Format:
Cond #,
N1,
N2,
G
HEADER SOURCE DATA
18
C *** This is where you define your sources (heating) ***
C Example Format:
Node #,
Q
END OF DATA
Now that you have the model framework built you can begin entering the node,
conductor and source data. The format for a basic node is shown below.
Node #, Temperature, MCp
The definition of a node consists of a node number, temperature and thermal mass,
each separated with a comma. Each value must be separated by a comma.
Optionally a name can be defined for each node. You can do this by entering a
alphanumeric string after the thermal mass. However, you must precede the name
with the TAK 2000 comment character “$”. This name is used in the output (printed
and plot files) for easy identification of the nodes.
After entering the node data your NODE DATA block should like similar to the figure
below. Notice that the thermal mass for the diffusion nodes has been entered as
mathematical expression instead of single number. In TAK 2000 any floating point
quantity can be entered in this manner.
HEADER NODE DATA
1,
20.0,
2.5*2.5*5.0*2.713*.962
$ Bar Node 1
2,
20.0,
2.5*2.5*5.0*2.713*.962
$ Bar Node 2
3,
20.0,
2.5*2.5*5.0*2.713*.962
$ Bar Node 3
4,
20.0,
2.5*2.5*5.0*2.713*.962
$ Bar Node 4
5,
20.0,
$ Bar Node 5
10,
20.0,
2.5*2.5*5.0*2.713*.962
19
-1.0
$ Bar Surface
-20,
20.0,
1.0
$ Wall Boundary Node
Before we continue on in our example, a few comments should be made about data
types. The TAK 2000 preprocessor is data type sensitive and expects node,
conductor and array numbers to be entered as integer. Temperatures, thermal
masses, heat rates must be entered as floating point numbers or expressions.
Once we have completed the node data we can enter the conductor data. The
format for a basic constant conductor is shown below.
Cond #, Node I, Node J, Value
It consists of a conductor number, Node “I”, Node “J” and conductor value, each
separated with a comma.
HEADER CONDUCTOR DATA
1, 20,
1, 1.712*2.5*2.5/2.5
2,
1,
2, 1.712*2.5*2.5/5.0
3,
2,
3, 1.712*2.5*2.5/5.0
4,
3,
4, 1.712*2.5*2.5/5.0
5,
4,
5, 1.712*2.5*2.5/5.0
6, 10,
5, 1.712*2.5*2.5/2.5
The last bit of data that we need to enter is the heat rate that we are going to apply
to the end of the bar (Node 10). Since we want to apply this heat rate only for the
first hour of the two transient run we will use the input function “STF” that applies a
heat rate as a step function.
The format for STF input function is shown below.
20
STF Node #, Time(on), Time(off), Period, Heat rate
It consists of the node where the heat will be applied, the on time, the off time, the
period (for cyclic heating), and finally the heat rate, all separated by a comma. A
more complete description of the STF source function can be found in the user
manual.
The following is what you would enter in the source data block. Notice that we must
convert the time off into seconds to be consistent with the rest of the model.
HEADER SOURCE DATA
STF
10,
0.0,
3600.,
7200., 100.
After entering the heat source data save the model by clicking on the SAVE menu
item under FILE. Exit the editor by clicking on EXIT under FILE. The model has
now been saved in a text file called BAR.TAK
Congratulations! You have completed the construction of your first
mathematical thermal model.
You should now be back at the TAK 2000 WorkBench environment. Now that the
mathematical model has been constructed we can run it through the solver. Once
the solver is done, we can then use TAK 2000’s built in X/Y plotter to generate a plot
of the temperatures as a function of time.
To run the solver (analyzer) simply press the ANALYZER button on the toolbar or
select ANALYZE from the main menu. An OPEN dialog box will open and will
display the TAK 2000 models in the current directory.
Select the BAR.TAK model and click on the dialog box’s ANALYZE button. The
solver will be launched at this point. If you made any typographic errors when
entering the data the pre-processor will display a screen as shown below.
21
If this occurs don’t be alarmed. Simply click on the YES button and you can view the
output file and determine where you made the error. To quickly find the error simply
do a search for “ERROR”. To do this select EDIT from the main menu, click on FIND
and then enter “ERROR” in the text box and then click on FIND NEXT That should
take you right to the line with your error.
The TAK 2000 syntax checker is very concise when reporting error so you should
have no trouble deducing what the problem it. Once you do, simply exit the editor
and open the BAR.TAK model file and fix your error. After exiting the editor run the
model again by pressing the ANALYZE button.
The solver should take only a few seconds so solve this simple model. After it has
completed the solution you will be asked if you want to view the output file. Click on
the YES button.
As always, a comprehensive output file was generated. This file will have a .OUT
extension so will be named BAR.OUT. This file is divided into two parts: 1.) an echo
of the input to insure that what you thought you were modeling is what TAK 2000
actually modeled. 2.) the temperature of every node for every time cut (if transient)
and a heat balance (if steady state). In addition, output from any other print function
selected in the OPTIONS DATA block will go to this file. You can quickly select an
output file for review by selecting the OUTPUT toolbar button. This will bring up a list
of all output files in the current folder.
In addition to the output file, a final temperatures file is always created at the end of
a solution. This file has the same rootname as you TAK file and a .FIN extension.
This file is a text file and can be used as initial temperatures for a new analysis run
22
or by several of the WorkBench tools. These tools are available from the menu or
toolbar buttons. A right-mouse click will also offer tools that are appropriate to a final
temperature file (file must be selected).
During a transient solution a temperature plot file is generated if the program
constant PLOT is set to a non-zero value in the CONTROL DATA block.
Temperature in the file can by plotted with the XY PLOT tool by selecting PLOT X-Y
toolbar button. Alternatively, the menu, or a right-mouse click will also offer tools that
are appropriate to a plot file.
Your output file should look like the following. It is annotated here to help you
understand the various sections.
------------------------------------------------------------------------------TAK 2000
Thermal Analysis Kit 2000
Release 1.0, DEMO
K & K Associates
*** TO BE USED FOR EVALUATION PURPOSES ONLY ***
Serial Number: 265100
------------------------------------------------------------------------------Input File: C:\PROGRAM FILES\KK ASSOCIATES\TAK 2000 DEMO\SAMPLES\BAR.TAK
C-------------------------------------------------------------------------C TAK 2000 Version 1.0
C-------------------------------------------------------------------------HEADER OPTIONS DATA
UID
SOLRTN
Logic
= SI
$ Model Units
= FWDWRD $ Solution routine
= Off
$ User logic
QPrint
GPrint
CPrint
CSGDump
NCVPrint
HPrint
LPrint
=
=
=
=
=
=
=
On
Off
Off
Off
Off
Off
Off
$
$
$
$
$
$
$
Print
Print
Print
Print
Print
Print
Print
QDump
PlotQ
LodTmp
AutoDamp
=
=
=
=
Off
On
Off
On
$
$
$
$
Print network map
Create .PLQ plot file of heat rates (SS Only)
Load initial temperatures from .INI file
Automatic solution damping function (SS Only)
Dictionaries = Off
impressed heat rates
conductor values
nodal thermal masses
noal CSG values
NCV conductor status
heater status
thermal louver status
Summary of
Program Options
$ Print actual/relative data dictionaries
-------------------------------- Options Summary -------------------------------Option
Default
Current
Description
------------------------------------------------------------------------LOGIC
OFF
OFF
Compile/Link User Logic
AUTODAMP
ON
ON
Automatic Relaxation Factor, S.S. Only
QDUMP
OFF
OFF
Generate Network Heat Flow Map
LODTMP
OFF
OFF
Load Initial Temperatures From .INI File
CSGDUMP
OFF
OFF
Print nodal CSG data
QPRINT
OFF
ON
Print Impressed Heating Rates
23
CPRINT
GPRINT
HPRINT
SUMMARY
LPRINT
PLOTQ
DICTIONARIES
UID
OFF
OFF
ON
ON
ON
OFF
OFF
None
OFF
OFF
OFF
ON
OFF
ON
OFF
SI
Print Thermal Masses
Print Conductor Values
Print Heater Status
Print Data Block Summaries
Print Louver Status
Create Heating Rate Plot File .PLQ
Print Actual/Relative Data Dictionaries
Units, ENG=English, SI=Sys. International
------------------------------------------------------------------------------
HEADER CONTROL DATA
C
-------------------
ABSZRO
SIGMA
IFC
= -273.15
= 5.67E-09
= 0
Program Control Constants ------------------$ Absolute Zero
$ Stefan-Boltzmann Constant
$ Output Units Flag
C Steady State constants
C
C
C
SSCRIT
NLOOPS
DAMP
= 0.001
= 500
= 1.0
$ Steady state convergence criteria
$ Max number steady state iterations
$ Steady state damping factor
C Transient constants
TIMEO
TIMEND
PLOT
SCALE
TRCRIT
NLOOPT
OUTPUT
=
=
=
=
=
=
=
0.0
7200.
120.
0.01667
0.01
50
1200.
$
$
$
$
$
$
$
Transient start time
Transient stop time
Plot file output interval
Plot file time scale factor
Transient convergence criteria
Max number of iterations per step
Transient output interval
C
----------------------- User Defined Constants ---------------------
C
*** Put your own user constants here ***
Summary of
Program control
constants
--------------------- Program Control Constants Summary --------------------Constant
NLOOPS
NLOOPT
TIMEO
TIMEND
OUTPUT
SIGMA
ABSZRO
MAXTEMP
TRCRIT
SSCRIT
DAMP
PLOT
SCALE
DTIMEI
DTIMEL
DTIMEH
NODELIST
IFC
Description
Default
Max # of iterations (S.S.)
Max # of iterations (trans.)
Start or "old" time
Stop or "end" time
Output interval
Stefan-Boltzmann constant
Absolute zero temperature, SI
Max. allowable temperature
Convergence criteria (trans.)
Convergence criteria (S.S)
Relaxation factor (S.S.)
Plot file output frequency
Plot file time scale factor
Transient time-step
Minimum time-step allowed
Maximum time-step allowed
Array # of QDUMP nodes.
Output units:1=C,2=F,3=K,4=R
100
50
0.0
none
none
1.0
-273.0
10000
0.01
0.001
1.0
0.0
1.0
none
1.0E-30
1.0E+30
0
0
Current Value
100
50
0.000
7200.
1200.
5.6700E-09
-273.150
10000
1.0000E-02
1.0000E-03
1.00
120.0
1.6670E-02
0.000
1.0000E-30
1.0000E+30
0
1
------------------------------------------------------------------------------
HEADER NODE DATA
1,
2,
3,
4,
5,
20.0,
20.0,
20.0,
20.0,
20.0,
2.5*2.5*5.0*2.713*.962
2.5*2.5*5.0*2.713*.962
2.5*2.5*5.0*2.713*.962
2.5*2.5*5.0*2.713*.962
2.5*2.5*5.0*2.713*.962
$
$
$
$
$
Bar
Bar
Bar
Bar
Bar
24
Node
Node
Node
Node
Node
1
2
3
4
5
10,
-20,
20.0,
20.0,
-1.0
1.0
$ Bar Surface
$ Wall Boundary Node
-------------------------------- Node Summary -------------------------------Node Type
Option
# Nodes
Description
---------------------------------------------------------------Diffusion
5
Regular Non-Zero Mass
Arithmetic
1
Zero Mass
Boundary
1
Regular Boundary
Summary of Node
Total Number of Nodes =
7 (
16 Maximum )
-----------------------------------------------------------------------------data
HEADER CONDUCTOR DATA
1, 20,
2, 1,
3, 2,
4, 3,
5, 4,
6, 10,
1,
2,
3,
4,
5,
5,
1.712*2.5*2.5/2.5
1.712*2.5*2.5/5.0
1.712*2.5*2.5/5.0
1.712*2.5*2.5/5.0
1.712*2.5*2.5/5.0
1.712*2.5*2.5/2.5
Summary
of
Conductor
----------------------------- Conductor Summary ----------------------------Cond. Type
Option
# Cond.
Description
-------------------------------------------------------Linear
6
Regular
Total Number of Conductors =
6 (
32 Maximum )
------------------------------------------------------------------------------
Summary
of Source
HEADER SOURCE DATA
stf 10, 0., 3600., 7200., 100.
------------------------------ Source Summary -----------------------------Source Type
Option
# Sources
---------------------------------------Step Function
STF
1
Total Number of Non-Constant Source Calls
1 (
16 Maximum )
------------------------------------------------------------------------------
Preprocessor done.
l
||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
St t f th
Processing of input file completed. No errors detected.
||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
-----------------------------Begin Solution: Routine FWDWRD
------------------------------
This is the
computed timestep
<<< INFO >>> Successfully opened temperature plot(.PLT) file
<<< INFO >>> Successfully opened heat rate plot(.PLQ) file
Node
1 has CSGMIN =
Calculated Time Step
=
12.704
12.069
25
Time Step Used
=
12.069
( Calculated )
26
Time =
0.000
: Temperatures (C) :
CSGMIN =
12.70
===============================================================================
T
1=
20.000 T
2=
20.000 T
3=
20.000 T
4=
20.000
T
5=
20.000 T
10=
20.000 T
20=
20.000
Time =
0.000
: Impressed Heating Rates :
===============================================================================
Q
1=
0.00
Q
2=
0.00
Q
3=
0.00
Q
4=
0.00
Q
5=
0.00
Q
10=
100.
Q
20=
0.00
Time =
1200.
: Temperatures (C) :
CSGMIN =
12.70
===============================================================================
Temperatures
and heat
T
1=
42.034 T
2=
86.233 T
3=
130.809 T
4=
175.974
T
5=
221.881 T
10=
245.246 T
20=
20.000
rates are printed for each
node every
Time =
1200.
: Impressed Heating Rates :
===============================================================================
Q
1=
0.00
Q
2=
0.00
Q
3=
0.00
Q
4=
0.00
Q
5=
0.00
Q
10=
100.
Q
20=
0.00
1200 seconds
Time =
2400.
: Temperatures (C) :
CSGMIN =
12.70
===============================================================================
T
1=
43.306 T
2=
89.925 T
3=
136.560 T
4=
183.220
T
5=
229.913 T
10=
253.278 T
20=
20.000
Time =
2400.
: Impressed Heating Rates :
===============================================================================
Q
1=
0.00
Q
2=
0.00
Q
3=
0.00
Q
4=
0.00
Q
5=
0.00
Q
10=
100.
Q
20=
0.00
Time =
3600.
: Temperatures (C) :
CSGMIN =
12.70
===============================================================================
T
1=
43.362 T
2=
90.086 T
3=
136.811 T
4=
183.537
T
5=
230.264 T
10=
253.629 T
20=
20.000
Time =
3600.
: Impressed Heating Rates :
===============================================================================
Q
1=
0.00
Q
2=
0.00
Q
3=
0.00
Q
4=
0.00
Q
5=
0.00
Q
10=
100.
Q
20=
0.00
Time =
4800.
: Temperatures (C) :
CSGMIN =
12.70
===============================================================================
T
1=
21.373 T
2=
23.984 T
3=
26.205 T
4=
27.819
T
5=
28.667 T
10=
28.667 T
20=
20.000
Time =
4800.
: Impressed Heating Rates :
===============================================================================
Q
1=
0.00
Q
2=
0.00
Q
3=
0.00
Q
4=
0.00
Q
5=
0.00
Q
10=
0.00
Q
20=
0.00
Time =
6000.
: Temperatures (C) :
CSGMIN =
12.70
===============================================================================
T
1=
20.060 T
2=
20.174 T
3=
20.271 T
4=
20.342
T
5=
20.379 T
10=
20.379 T
20=
20.000
Time =
6000.
: Impressed Heating Rates :
Temperatures and
===============================================================================
Q
1=
0.00
Q
2=
0.00
Q
3=
0.00
Q
4= rates
0.00 after 7200
Q
5=
0.00
Q
10=
0.00
Q
20=
0.00
seconds
Time =
7200.
: Temperatures (C) :
CSGMIN =
12.70
===============================================================================
T
1=
20.003 T
2=
20.008 T
3=
20.012 T
4=
20.015
T
5=
20.017 T
10=
20.017 T
20=
20.000
Time =
7200.
: Impressed Heating Rates :
===============================================================================
Q
1=
0.00
Q
2=
0.00
Q
3=
0.00
Q
4=
0.00
27
heat
Q
5=
0.00
Q
10=
0.00
Q
20=
0.00
<<<<<<<<<<<<<<<<<<<<<<<<<<<<<< RUN DIAGNOSTICS >>>>>>>>>>>>>>>>>>>>>>>>>>>>>>
The is the
Number of warnings issued:
0
Number of errors detected:
0
solution
Run Start Time:
Run Stop Time:
11-Mar-00
11-Mar-00
summary
16:05:16
16:05:17
Pre-Process Time:
0 Hrs. 0 Min. 0 Sec.
Compile/Link Time:
0 Hrs. 0 Min. 0 Sec.
Solution Time:
0 Hrs. 0 Min. 1 Sec.
Miscellaneous:
0 Hrs. 0 Min. 0 Sec.
------------------------------------------Total Run Time:
0 Hrs. 0 Min. 1 Sec.
<<<<< End Of Job >>>>>
At 1 hour (3600 seconds) the end of the bar that is being heated has almost reached
254°C. However, after removing the heat and allowed to cool the complete bar is
almost back to its initial temperature of 20°C after just 1 more hour.
The Plotter
You can use TAK 2000 ‘s built in X/Y plotter to create a plot of the temperatures as a
function of time. To start the plotter just click on the XY Plot button on the toolbar. An
Open dialog box will appear. Select BAR.PLT and click the OPEN button on the
dialog box.
The plotter will be launch and a NODES SELECTIONS dialog box will appear. Click
on each node to add them to the list of nodes to be plotted. A alternate method to
quickly add groups of nodes is to enter a node range in the ENTER NODES TO
PLOT text box. Since our nodes range from 1 to 20 we will enter “1-20” in this text
box.
After entering the node range in the text box, press the Add button. All the nodes in
the model should be added to the “Selected Node List” box as shown below.
28
Once the nodes that you want plotted are listed in “Selected Node List” click on the
display button to generate the plot as shown below.
29
You’ll notice that the x-axis label indicates units of time in hours. This is the default
label for the x-axis and can easily be changed in the OPTIONS menu. Once you are
finished viewing the plot exit the plotter by clicking on EXIT under the FILE menu
item. Press the YES button when you are asked if you want to exit the plotter.
You should now be back at the TAK 2000 WorkBench. To exit TAK 2000 completely,
click on EXIT under the FILE menu item.
That concludes this simple demonstration. You are encouraged to read the online
help and to try out the other features of TAK 2000 like the MinMax, Compare and
Reformat tools. They are located under the TOOLS menu. Also, read about the
numerous features of the analyzer with the on-line help.
We at K & K Associates thank you for taking the time to evaluate our software. We
believe you won’t find any similar product on the market today that compares in
features and price.
Attention Circuit Board Designers
Be sure to check out our new release of our famous Circuit Board Model builder
application PCBUILD 2000. This companion program to TAK 2000 is scheduled to
be released in the summer of 2000.
For those involved in thermal analysis of electronic circuit boards, this is an
indispensable tool. PCBUILD 2000 is a Windows based application that will create a
detailed thermal model of any circuit board. It has a built in thermal solver,
component library, contour plotting capability and an export function that will allow
you to generate a ready to run TAK 2000 thermal model.
30
Download PDF