Viewfactor > Model Creation for a Thermal Radiation Problem > 3.7 Patran Thermal TEMPLATEDAT Files for Surface Property Description
XXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXX''">XXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXX''">   
3.7 Patran Thermal TEMPLATEDAT Files for Surface Property Description
Patran Thermal supports an arbitrary finite number of wavebands for spectral dependent material properties for thermal radiation interchange. Within each waveband the material properties are assumed to be independent of wavelength (gray) and diffuse. Completely gray surfaces are also supported without the added work on your part to specify the waveband as the entire spectrum. This section describes the data needed by Viewfactor in the form of the Patran Thermal TEMPLATEDAT file and VFAC template to make the Patran Thermal thermal network resistors for modeling thermal radiation interchange.
Thermal Radiation Wavebands as Used in MSC Patran Thermal User’s Guide
If you only plan to use gray surfaces, then the information on thermal radiation wavebands and wavelength dependent network resistors may be ignored. If the wavelength dependence is not specified by entering the wavebands, then Viewfactor assumes that the enclosure is gray and you need not be concerned with wavebands.
Thermal radiation wavebands are defined in the Viewfactor code for use by P⁄THERMAL in terms of their beginning and ending wavelengths in units of microns. The spectrum begins at zero wavelength and extends to infinity. Since infinity is not a convenient quantity with which to work in computers, we will use some finite, but large number to represent the upper end of the spectrum, for example 1.0E10 microns. In general, the thermal radiation above this wavelength is completely negligible for engineering problems. (This may not be true in certain physics problems.)
Patran Thermal evaluates the black body function in each waveband and at the temperature of each surface. This is an improvement over other methods which use some mean temperature between the surface pairs (e.g., geometric mean, for the black body temperature). The heat flow between two surfaces 1 and 2 as represented by the Patran Thermal network equations is
(3‑1)
where:
i
=
Waveband index
s
=
Stefan-Boltsmann constant
nbands
=
Number of wavebands
F
=
Black body function from at temperature T
T1,T2
=
Temperatures of surfaces 1 and 2, respectively
R
=
Effective radiative resistance between surfaces 1 and 2, taking into account possibly time, temperature, and waveband
t
=
Time
l
=
Wave length
An example of spectrally-dependent surface emissivities for two surfaces is plotted in Figure 3‑22. Also shown are the approximate wavebands and constant waveband properties that might be used to represent the surface properties of these two surfaces. The approximating properties are shown as thin dashed lines. Refer to Wavebands and Enclosures, 25 for a discussion on the need for a consistent set of wavebands throughout an enclosure.
Figure 3‑22 Modeling Spectrally Dependent Properties
Patran Thermal’s QTRAN will accept overlapping wavebands and/or inactive or missing regions in the spectrum. This is both a blessing and a curse. It gives you the latitude to model surface properties with piecewise constant basis functions and leave out inactive regions of the spectrum from the analysis, but no checking is performed for a nonoverlapping and/or incomplete spectrum. Thus you must be responsible for the correctness of the waveband model and data.
Radiation Resistor Types Used in Patran Thermal
This Guide deals specifically with radiation resistors. Refer to the Introduction (Ch. 1) in the Patran Thermal User’s Guide Volume 1: Thermal/Hydraulic Analysis for a complete discussion of Patran Thermal resistors.
Patran Thermal’s radiation resistors are classified and identified according to their type and subtype. Two types are allowed: being gray (no spectral dependence) and waveband dependent (spectral dependencies as described in the previous section). The gray resistors are identified by the letter R and the waveband dependent resistors are identified by the letter W. Each type has a number of possible subtypes, the set of possible subtypes being identical for each type. This being the case, the subtypes are described without reference to the types and you may then associate the subtypes with the various types as desired. The different subtypes are identified by integer IDs.
Viewfactor creates unformatted data records for the Patran Thermal radiation resistors it creates. These records are in the VFRESDAT file. Since the data is unformatted, it is not easily read by the user. The unformatted form is used to save space. For our discussions here, the data used to describe the resistors is represented in a form readable by the user.
A Patran Thermal radiation resistor record consists of 11 pieces of data, not all of which are required for each resistor type and subtype. A radiation resistor is described by:
RESTYP, SUBTYP, NODE1, NODE2, NODE3, MPID, DATA1, DATA2, DATA3, LAMBDA1, LAMBDA2
Input Data
Description
RESTYP
Resistor type, character, R or W.
SUBTYP
Resistor subtype, integer, 1 through 12.
NODE1
First node in the resistor record, integer.
NODE2
Second node in the resistor record, integer.
NODE3
Third node in the resistor record, integer (not always used, but must be present as a place holder).
MPID
Material property ID for Patran Thermal, integer.
DATA1
Data for use in calculating the resistor value, real.
DATA2
Data for use in calculating the resistor value, real (not always used, but must be present as a place holder).
DATA3
Data for use in calculating the resistor value, real (not always used, but must be present as a place holder).
LAMBDA1
Beginning wavelength of waveband, real (not used for type R resistors, but must be present as a place holder).
LAMBDA2
Ending wavelength of waveband, real (not used for type R resistors, but must be present as a place holder).
Radiation Resistor Subtypes
Subtype 1
R = (1.0 - EPSILON) / (EPSILON * AREA)
This resistor is used between a gray surface and a radiosity node, with an emissivity, EPSILON, which is evaluated from a material property, MPID. If the material property is temperature dependent, it will be evaluated at the temperature of NODE1. Typically, NODE1 is the surface node and NODE2 is the radiosity node. The variable, AREA, is the area (nodal subarea) associated with NODE1. NODE3 is not used for this resistor.
Subtype 2
R = 1.0 / ( FF * AREA * TAU )
This resistor is used between radiosity nodes and has a participating media transmissivity, TAU, evaluated directly from a material property, MPID. If the transmissivity material property is temperature dependent, it will be evaluated at the temperature of NODE3. Typically, NODE1 and NODE2 represent the radiosity nodes and NODE3 the participating media node. DATA1 contains the area, AREA, associated with NODE2. DATA2 contains the viewfactor, FF, from NODE2 to NODE1. If one of the radiosity nodes is an AMBNOD, then it will typically be NODE1. Currently, this subtype is not created by Viewfactor since subtype 9 is adequate for the present requirements and requires less computational time.
Subtype 3
R = 1.0 / ( FF * AREA * ( 1.0 - TAU ) )
This resistor is similar to subtype 2, except this subtype is used to represent the radiant interchange between a radiosity node, NODE2, and a participating media node, NODE1. Typically, NODE3 and NODE1 are the same, but this is not required. Currently, this subtype is not created by Viewfactor since subtype 10 is adequate for the present requirements and requires less computational time.
Subtype 4
R = 1.0 / ( DATA1 * DATA2 )
This is a general purpose resistor between NODE1 and NODE2 whose value is determined by multiplying two constants, for example the viewfactor and the area for a case with no participating media. NODE3 is not used for this resistor. Currently, this subtype is not created by Viewfactor since subtype 5 is adequate for the present requirements and requires less computational time.
Subtype 5
R = 1 / DATA1
This is another general purpose resistor between NODE1 and NODE2 and is the simplest and computationally fastest resistor. It is useful whenever material properties are known constants and do not require access to the Patran Thermal material property data. Two typical uses are (1) between two radiosity nodes with DATA1 = FF * AREA * TAU, or (2) as an emissivity resistor between a non-black surface node and a radiosity node with DATA 1 = (EPSILON * AREA ) / ( 1.0 - EPSILON ). NODE3 is not used for this resistor.
Subtype 6
R = ( 1.0 - DATA2 ) / ( DATA2 * DATA1 )
This is the constant known property version of subtype 1 where DATA2 is typically EPSILON and DATA1 is AREA. NODE3 is not used for this resistor. Presently, this subtype is not created by Viewfactor since subtype 5 is adequate for the present requirements and requires less computational time.
Subtype 7
R = 1.0 / ( FF * AREA * TAU )
This resistor is used between radiosity nodes and has a participating media transmissivity, TAU, evaluated using Beer's law ( TAU = EXP( - KAPPA * DISTANCE ) ) and an extinction coefficient from a material property, MPID. If the extinction coefficient material property is temperature dependent, it will be evaluated at the temperature of NODE3. Typically, NODE1 and NODE2 represent the radiosity nodes and NODE3 the participating media node. DATA1 contains the area, AREA, associated with NODE2. DATA2 contains the viewfactor, FF, from NODE2 to NODE1. DATA3 contains the mean beam length from the surface associated with NODE2 to the surface associated with NODE1. If one of the radiosity nodes is an AMBNOD, then it will typically be NODE1. Currently, this subtype is not created by Viewfactor since subtype 11 is adequate for the present requirements and requires less computational time.
Subtype 8
R = 1.0 / ( FF * AREA * ( 1.0 - TAU ) )
This resistor is similar to subtype 7, except this subtype is used to represent the radiant interchange between a radiosity node, NODE2, and a participating media node, NODE1. Typically, NODE3 and NODE1 are the same, but this is not required. Currently, this subtype is not created by Viewfactor since subtype 12 is adequate for the present requirements and requires less computational time.
Subtype 9
R = 1.0 / ( FA * TAU )
This resistor is similar to subtype 2, but the FF and AREA have been combined by multiplication into FA and stored in DATA1. DATA2 is not used.
Subtype 10
R = 1.0 / ( FA * ( 1.0 - TAU ) )
This resistor is similar to subtype 3. It is to subtype 3 what subtype 9 is to subtype 2.
Subtype 11
R = 1.0 / ( FA * TAU )
This resistor is similar to subtype 7, but the FF and AREA have been combined by multiplication into FA and stored in DATA1. DATA2 in not used.
Subtype 12
R = 1.0 / ( FA * ( 1.0 - TAU ) )
This resistor is similar to subtype 8. It is to subtype 8 what subtype 11 is to subtype 7.
Patran Thermal MPIDs (Material Property IDs)
Refer to the Introduction (Ch. 1) in the Patran Thermal User’s Guide Volume 1: Thermal/Hydraulic Analysis for a complete discussion of MPIDs (material property IDs). If you choose to use MPIDs for the thermal radiation material properties, then these properties must either already be available in the material property data file or you will need to create the appropriate Patran Thermal material property specifications. Some examples are shown and discussed below as they would appear in the ‘mat.dat’ file.
Figure 3‑23 Temperature Dependent Emissivity
Figure 3‑24 Time Dependent Emissivity
Patran Thermal Material Property Definition
The various ways to define material properties are described in the MSC Patran Thermal User’s Guide. Note that by referencing the MPID as a negative number in the TEMPLATEDAT template the material property will be evaluated as a function of time instead of as a function of temperature.
In Patran, the material properties in Patran Thermal MPIDs (Material Property IDs), 56 are defined under fields:
Finite Element
 
Action:
Create
Object:
Material Property
Method:
General
Click on Input Data, then the desired Patran Thermal material function (e.g., "mpid_linr_tabl" will bring up the input data for a tabular input).
Note:  
Make sure you have selected Patran Thermal as the analysis preference.
VFAC Template Format
The complete specification of the VFAC template is given in TEMPLATEDAT (Surface Pointer Data), 149. Unless the reader plans to interface to the Viewfactor module formally through another computer code, the description of the VFAC template in this section will be all that is needed.
The VFAC template consists of a header line and one or more following data lines. Data on the header line will determine how many data lines must follow. Some of the data is optional, or optional depending on what other data is entered. All optional data defaults to standard values. If the value desired is the default value, then you need not enter the optional default values, as they will be assigned automatically.
KEYWORD TID nbands
The first line of the VFAC template, or header line, consists of three fields. These fields are:
1. The keyword VFAC,
2. Followed by the template ID number, TID, and finally
3. The number of wavebands and nbands used to define the thermal radiation properties. The first two fields are required and the third field is optional and defaults to zero. The TID and nbands must be integers. The fields should be separated by one or more spaces or by commas.
TID
Valid TIDs are positive integers. TIDs will be associated with UIDs from the Patran Viewfactor LBC form.
nbands
Valid values of nbands are non-negative integers. If nbands is zero (the default), then this template will cause R-type resistors to be created; otherwise, W-type resistors will be created when this template is referenced.
The number of data lines following the header line must be exactly the number of nbands, except that if the nbands field is blank or zero there must be exactly one data line following. Comments may also be placed at the end of header or data lines by placing a semicolon after the last data field on the line.
Some example header lines are:
VFAC
99
 
which specifies that this is VFAC template 99 and represents a gray material property, and
VFAC
1001
6
which specifies that this is VFAC template 1001 and represents a material with 6 wavebands.
The data lines contain eight fields each. The fields are separated by one or more spaces or by commas. The first field, CONSTANT_EPSILON, is the only required field. Trailing blank fields may be omitted. Blank fields between nonblank fields are not permitted and will be ignored, thus corrupting the user’s data. The fields are:
CONSTANT_EPSILON, CONSTANT_TAU, EMPID, TMPID, LAMBDA1, LAMBDA2, KFLAG, COLLAPSE
Input Data
Description
CONSTANT_EPSILON
This field is required and must be a real number. It is the value of the surface’s constant emissivity, or if the emissivity is not constant, then it must have the value 0.0. Valid values of emissivity are greater than 0.0 and less than or equal to 1.0. If the emissivity is not constant, then the material property ID which describes the property must be given in the EMPID field.
CONSTANT_TAU
This field is optional and defaults to 1.0. It is the value of the constant transmissivity of the participating media if any (the transmissivity of a vacuum is 1.0, hence the default value). If this value is not constant, then it must be given the value 0.0 here. Valid values of constant transmissivity are greater than 0.0 and less than or equal to 1.0. If the KFLAG is set to 1, then this value will represent the extinction coefficient for absorption according to Beer's Law and valid values are greater than 0.0. If the extinction coefficient is not constant, then this value must be set to zero. If this property, either transmissivity or extinction coefficient, is not constant, the material property ID which describes the property must be given in TMPID field.
EMPID
This field is an optional integer MPID (material property ID) and defaults to 0. It must assume its default value if a constant emissivity is specified by CONSTANT_EPSILON. If the emissivity is not constant, as indicated by the value 0.0, then EMPID must be nonzero. Positive values of EMPID denote temperature dependence, while negative values of EMPID will be evaluated as functions of time.
TMPID
This field is an optional integer MPID (material property ID) and defaults to 0. It must assume its default value if a constant transmissivity or extinction coefficient is specified by CONSTANT_TAU. If a constant is not specified, as indicated by its value of 0.0, then the TMPID must be nonzero. Positive values of TMPID denote temperature dependence, while negative values of TMPID will be evaluated as functions of time. The TMPID will evaluate to a transmissivity if the KFLAG is zero and to an extinction coefficient if the KFLAG is one, just as the CONSTANT_TAU does.
LAMBDA1, LAMBDA2
These are optional real fields, but either they both must be present or both not present. They are the beginning and ending wavelengths for the present waveband. Note that the wavebands do not necessarily have to be in order of increasing wavelength but must be in the same order for every surface in an enclosure. These fields default to 0.0, the value for the case when nbands is 0. If nbands is greater than 0 (i.e., the properties have spectral dependence), then LAMBDA2 should be greater than LAMBDA1, which should be greater than 0.0. The units used for wavelength are microns or micrometers.
KFLAG
This optional field signals whether the transmissivity is evaluated directly (KFLAG = 0), either from the constant value or from the MPID referenced in TMPID, or the transmissivity is evaluated using Beer’s Law and an extinction coefficient evaluated from either the constant value or from the MPID referenced by TMPID. The default KFLAG value is 0 and the data must be integer. Beer’s Law may be stated as:
Transmissivity = EXP( - Extinction_Coefficient * Distance )
COLLAPSE
This optional field signals whether radiosity nodes associated with a given surface node should be collapsed into a single radiosity node. If COLLAPSE is zero (the default value), then the radiosity nodes will not be collapsed. The COLLAPSE_ID associated with a nodal subarea surface is transferred to that nodal subarea’s radiosity node. Then radiosity nodes connected to the same surface node by way of emissivity resistors and having the same nonzero COLLAPSE_ID will be collapsed into one node. The resulting parallel emissivity resistors will be merged if possible. The COLLAPSE_ID must be a non-negative integer.
The main advantage of using COLLAPSE to collapse radiosity nodes is that this will result in a much smaller number of radiation resistors in the model. The effect of using COLLAPSE for small resistor networks is shown in Figure 3‑25. The effect is more pronounced for larger networks or for 3-D networks. A smaller number of resistors usually means that the thermal analysis will proceed faster. In the best cases, the number of radiation resistors may be reduced by about a factor of four for 2-D Cartesian or axisymmetric models and by about a factor of 16 for 3-D models.
The main advantage of using COLLAPSE to collapse radiosity nodes is that this will result in a much smaller number of radiation resistors in the model. The effect of using COLLAPSE for small resistor networks is shown in Figure 3‑25. The effect is more pronounced for larger networks or for 3-D networks. A smaller number of resistors usually means that the thermal analysis will proceed faster. In the best cases, the number of radiation resistors may be reduced by about a factor of 4 for 2-D Cartesian or axisymmetric models and by about a factor of 16 for 3-D models.
Our experience is that the loss of accuracy is quite small for fine meshes and lower temperatures. The user may wish to try the examples in Example Thermal Radiation Problems, 11, using the COLLAPSE field modeling techniques. Other existing models may also be rerun using the new COLLAPSE flag. Then the results can be compared with previous results and provide the user with a basis for deciding when to use or not use the COLLAPSE feature.
Figure 3‑25 Effect of COLLAPSE on Radiation Resistor Network
The COLLAPSE is applied on a template by template basis and is applied separately to each individual waveband in the template. This versatility gives the user full control over which surface will have their corresponding radiosity nodes collapsed. In order to collapse the radiosity nodes on one surface, but not on another surface of the same material, the user will assign two different template IDs (TIDs), to the two surfaces. Then in the VFAC templates, specifying the COLLAPSE_ID for each surface will determine whether the radiosity nodes made for that surface are collapsed.
Examples of TEMPLATEDAT Files for Thermal Radiation
The following examples of VFAC templates range from simple to complex. Each example is described briefly.
This VFAC template will be referenced whenever a surface which was VFAC LBCed with UID 99 is referenced. This template defines a gray surface (nbands defaults to 0) and a constant emissivity of 0.89 and a transparent media (default value of 1.0 for TAU).
The following template also describes a gray surface, but this time there is a participating media also with a constant transmissivity of 0.95.
This is a slightly more complicated template. It is still for a gray surface, but now the emissivity is defined by the material property with MPID 4000.
The following template defines both the emissivity and transmissivity in terms of material property data records identified by MPIDs (EMPID and TMPID). Since the TMPID is negative, it will be evaluated as a function of time.
 
The next template has the KFLAG set to one, and thus needs all of the preceding data on the line defined (recall that embedded blank fields are not allowed). This template is similar to the previous one, except that the transmissivity will be evaluated by Beer’s Law and the extinction coefficient will be determined from the time dependent material property definition in MPID 3000.
The next example shows the previous example without the supporting comments and spaces. Clearly the previous example is easier to read and we recommend that some standard and clear format be followed by the user.
The following template is for a surface with three wavebands, the first from 0.0 to 2.0 microns, the second from 2.0 to 5.0 microns, and the third from 5.0 to 1.0E6 microns (approximately infinity for wavelengths). Each waveband defines the properties in a different way and serves to exemplify the versatility of the wavebands and VFAC templates.
The following templates show the use of the COLLAPSE_ID. Note that when a COLLAPSE_ID is given, the other fields of the template must be filled in with appropriate or default values as placeholders.