XXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXX''"> 13.3 Derivations
Prior to Patran Version 9.0, the results processor attempted to recognize whether a stress or strain tensor was 2D (e.g. Plane Stress:
iz = 0 and Plane Strain:
iz = 0 for
i = 1, 2, 3) or 3D and then calculate principal values based either a 2D or 3D formulation respectively. Starting with Patran Version 9.0 the user must choose to use either the 2D or 3D formulation. The mechanism that is provided to allow the user to choose a specific formulation is to either select the derived quantity that includes “2D” as part of its name, which will cause the 2D formulation to be used, or to choose the quantity that does not contain “2D” as part of its name, which will cause the 3D formulation to be used. This change was motivated by requests from our customers who wanted to control which formulation they wanted to apply.
Examples of these 2D or 3D tensor quantities are shown below.
• Max Principal 2D | • Max Principal |
• Min Principal 2D | • Min Principal |
• Tresca 2D | • Tresca |
• Max Shear 2D | • Max Shear |
For 2D tensors Patran uses the two in plane principal values as the maximum and the minimum regardless if both of their values are either greater or less than zero. Patran calculates the maximum shear stress to be one half the difference between the maximum and minimum principal values. A consequence of this formulation is that for the cases where both in plane principal values have common signs the maximum shear stress can be under calculated. Similarly, Tresca stress could be under calculated, as shown in the following example.
Example:
Smajor = 30Sminor = 10 where Smajor and Sminor are the maximum and minimum 2D in-plane Principal Stresses respectively.
Using a 2D tensor, ”Tresca 2D” will be 30-10 = 20.
Using a 3D tensor, ”Tresca” will be 30 – 0 = 30 (Sminor = 0)
Derivation Definitions
The following table provides additional definitions for the selected result derivations. These include tensor to vector, tensor to scalar, and vector to scalar resolutions.
Transform Type | Derivation Method | Description |
Scalar to Scalar Vector to Vector Tensor to Tensor | None | No transformation is used if the result data type matches the plot tool’s data type. |
Vector to Scalar | Magnitude | Vector magnitude. |
X Component | 1st vector component. |
Y Component | 2nd vector component. |
Z Component | 3rd vector component. |
Tensor to Scalar | XX Component | XX tensor component. |
YY Component | YY tensor component. |
ZZ Component | ZZ tensor component. |
XY Component | XY tensor component. |
YZ Component | YZ tensor component. |
ZX Component | ZX tensor component. |
Min Principal | Calculated minimum principal magnitude. |
Mid Principal | Calculated middle principal magnitude. |
Max Principal | Calculated maximum principal magnitude. |
1st Invariant | Calculated 1st invariant |
2nd Invariant | Calculated 2nd invariant |
3rd Invariant | Calculated 3rd invariant |
Hydrostatic | Calculated mean of the three normal tensor components. |
von Mises | Calculated effective stress using von Mises criterion. |
Tresca | Calculated Tresca shear stress. |
Max Shear | Calculated maximum shear magnitude. |
Octahedral | Calculated Octahedral shear stress. |
Tensor to Vector | Min Principal | Calculated minimum principal vector. |
Mid Principal | Calculated middle principal vector. |
Max Principal | Calculated maximum principal vector. |
Below are the equations and examples of the derivation methods:
Important: | These equations for calculating invariants are not recommended for complex results since phase is not taken into account. |
von Mises Stress
von Mises stress is calculated from the following equation:
Example: The elements shown below have the following stress contributions:
Elem. ID | Node ID | | | | | | |
1 | 1 | 46.2 | 13.01 | 0.00 | 5.13 | 0.00 | 0.00 |
2 | 93.39 | 25.33 | 0.00 | 17.45 | 0.00 | 0.00 |
11 | 68.37 | 12.16 | 0.00 | -19.73 | 0.00 | 0.00 |
10 | 44.32 | 10.40 | 0.00 | -1.01 | 0.00 | 0.00 |
2 | 2 | 93.39 | 25.33 | 0.00 | 17.45 | 0.00 | 0.00 |
3 | 88.67 | 24.41 | 0.00 | 23.95 | 0.00 | 0.00 |
12 | 57.42 | 5.44 | 0.00 | -34.02 | 0.00 | 0.00 |
11 | 59.37 | 10.16 | 0.00 | -20.73 | 0.00 | 0.00 |
9 | 10 | 44.32 | 10.40 | 0.00 | -1.01 | 0.00 | 0.00 |
11 | 67.37 | 11.16 | 0.00 | -18.73 | 0.00 | 0.00 |
20 | 4.72 | 8.15 | 0.00 | -15.28 | 0.00 | 0.00 |
19 | 17.99 | 7.68 | 0.00 | -4.61 | 0.00 | 0.00 |
10 | 11 | 100.37 | 14.16 | 0.00 | -30.73 | 0.00 | 0.00 |
12 | 57.42 | 5.44 | 0.00 | -34.02 | 0.00 | 0.00 |
21 | -5.63 | 5.72 | 0.00 | -22.03 | 0.00 | 0.00 |
20 | 4.72 | 8.15 | 0.00 | -15.28 | 0.00 | 0.00 |
The von Mises stress calculated at node 11 when nodal averaging is done first due to the contribution from each element is 78.96. When the von Mises derivation is done first and then averaging at the nodes takes place, the calculated von Mises stress is 79.02. Thus a difference can arise depending on whether the averaging is done first or the derivation. This can be true for all derived results.
Node 11 | | | | | | | von Mises Stress |
E1 | 68.37 | 12.16 | 0.00 | -19.73 | 0.0 | 0.0 | 71.82 |
E2 | 59.37 | 10.16 | 0.00 | -20.73 | 0.00 | 0.00 | 65.68 |
E9 | 67.37 | 11.16 | 0.00 | -18.73 | 0.00 | 0.00 | 70.45 |
E10 | 100.37 | 14.16 | 0.00 | -30.73 | 0.00 | 0.00 | 108.10 |
Average | 73.87 | 11.91 | 0.00 | -22.48 | 0.00 | 0.00 | 79.02 |
Average then Derive | 78.96 |
Derive then Average | 79.02 |
Important: | It must be noted also that for von Mises and other derived results, the calculations are generally valid only for stresses. Although these operations can be performed for any valid tensor or vector data stored in the database, quantities such as tensor strains are not appropriate for von Mises calculations. To calculate a true von Mises strain the strain tensor must be converted to engineering strains by multiplying the shear components by a factor of two. |
Octahedral Shear Stress
Octahedral shear stress is calculated from the following equation:
From the von Mises example above the octahedral shear stress is:
Octahedral Shear Stress | Node 11 |
Average/Derive | 37.22 |
Derive/Average | 37.25 |
Hydrostatic Stress
Hydrostatic stress is calculated from the following equation:
From the von Mises example above the hydrostatic stress is:
Hydrostatic Stress | Node 11 |
Average/Derive | 28.59 |
Derive/Average | 28.59 |
Invariant Stresses
1st, 2nd, and 3rd invariant stresses are calculated from the following equations:
From the von Mises example above the invariant stresses are:
Invariant Stresses (Node 11) | 1st Invariant | 2nd Invariant | 3rd Invariant |
Average/Derive | 85.78 | 374.44 | 0.00 |
Derive/Average | 85.78 | 373.38 | 0.00 |
Principal Stresses
Principal stresses are calculated from either a Mohr’s circle method for 2D tensors
or from a 3x3 Jacobian Rotation Eigenvector extraction method for a 3D tensors. The User Interface allows for either a tensor-dependent derivation, or a 2D calculation. The tensor-dependent calculation will choose either a 2D or 3D calculation depending on values of each tensor encountered. A 2D calculation will be used when the ZZ, YZ and ZX are exactly zero (which is the case when the analysis code does not calculate these values), otherwise the full 3D tensor will be considered. Both the magnitudes of the principals and their direction cosines are calculated from these routines.
The magnitudes of the two principal stresses from the 2D Mohr’s circle method are calculated according the following equations:
The direction cosines for the 2D Mohr’s circle method are calculated by assembling the following 3x3 transformation matrix:
From the von Mises example above the principal stresses are:
Principal Stresses (Node 11) | Maximum | Minimum |
Average/Derive | 81.17 | 4.61 |
Derive/Average | 81.20 | 4.58 |
Also the principal stress determinant is the product of the three principals and the major, minor, and intermediate principal deviatoric stresses are calculated from:
Tresca Shear Stress
Tresca shear stress is calculated from the following equation:
where
are calculated as mentioned under Principal stress derivations above.
From the von Mises example above the Tresca shear stress is:
Tresca Shear Stress | Node 11 |
Average/Derive | 76.55 |
Derive/Average | 76.61 |
Maximum Shear Stress
Maximum shear stress is calculated from the following equation
where
are calculated as mentioned under Principal stress derivations above.
From the von Mises example above the Tresca shear stress is:
Tresca Shear Stress | Node 11 |
Average/Derive | 76.55 |
Derive/Average | 76.61 |
Magnitude
Magnitude (vector length) is calculated from the components with the standard formula: