Ansys Acceleration
Rachal Langwith
Recently I was working with a problem where point mass and gravity are considered. but instead of Gravity, I applied acceleration in the upward direction and the results are same. unable to understand which is correct either gravity or acceleration?
Thank you, Peter, for the clarification. But my question is that y Acceleration should be applied in +Y direction which is the same as -Y direction of Gravity. is there any supporting documents (Or theory which proves).
You can build a simple model to demonstrate. Make a cube, use a fixed support on the bottom, add -Y gravity, solve and request the reaction force at the fixed support. Suppress gravity and apply a +Y acceleration of 9.8xxxxx m/s^2 (there are more digits you can see in the Gravity field), solve and look at the reaction force. It will be the same.
Ansys Mechanical APDL and Mechanical Workbench can perform base acceleration in harmonic analysis on a variety of structure types. Be it for determining the steady-state sinusoidal response to sinusoidal varying loads all acting at a specified frequency. Some load types can be applied with a phase offset.
Figure 1 above shows the Ansys Workbench Project Schematic for a Harmonic analysis, and includes a Modal analysis to characterize the structure. This Harmonic analysis is not using modal superposition, as can be seen in the fact that Solution results from the Modal analysis are not linked.
Many structures are tested per customer specification with base accelerations swept across a range of frequencies. Such a test can be implied by fixing a base in Ansys, and applying an acceleration load to the whole model with an ACEL command.
The resulting movement at a point on the FEA model is relative to a fixed non-moving base, and is not what would be picked up at the point with an accelerometer. To replicate acceleration movements at a base would require an acceleration input at selected geometry, which Ansys does not directly support in Harmonic analysis.
This article illustrates conversion of an acceleration harmonic input into a displacement input, and its use in an Ansys Workbench model. Such an input should result in model movements that replicate what should be picked up by a physical accelerometer placed on the product, since they include base movement.
The same geometry and mesh are used in a Harmonic analysis in Workbench. The harmonic load is applied at the fixed end of the beam with X and Z movements fixed, but a UY harmonic displacement applied. An APDL test will show that although the Ansys command D for a non-zero displacement can support velocity or acceleration input in some structural analyses, these inputs are not supported in a Harmonic analysis:
A non-zero displacement load can be applied in a Harmonic analysis. When a specification calls for a velocity or acceleration load to be applied to selected geometry in a harmonic vibration test, Ansys commands can convert the specified velocity or acceleration load into a displacement load and apply it to the geometry with a Table Array as a function of frequency.
When a specification calls for an acceleration to be applied to a base or other selected geometry on a structure, it can be converted to an applied non-zero displacement as a function of frequency in this manner.
In the present example, a constant acceleration value is to be applied across a range of frequencies. A Table Array can be used in Ansys to achieve this. To apply a non-zero harmonic displacement requires Full harmonic analysis. The following APDL Commands Object at the Environment level applies such a load to nodes in the Named Selection (component) called End_Face. In this example, a constant acceleration of 10 mm/sec2 is converted to displacement as a function of frequency.
The frequency range in the APDL code above must span or exceed the range of frequencies to be employed in a sweep of harmonic analysis frequencies. The Table Array method used here will apply displacement amplitudes as a function of frequency during the SOLVE of the model. It is assumed here that there is no phase shift in the applied acceleration. Note the implied 180 degree phase shift in the minus sign on the displacement value.
The Outline for the example shows the Named Selection used in the above APDL code, and the placement of the APDL code in the Harmonic Response environment. Since the above APDL places a non-zero displacement in the UY direction on the face that is fixed in the modal analysis, other objects constrain the face movement in X and Z:
When the input acceleration is a function of frequency, as just above, then the Frequency Response where the load is applied can be traced in the Solution output. For the above variable acceleration example, a plot shows:
Although Ansys and Workbench cannot take an applied velocity or acceleration as a harmonic displacement input at nodes, Workbench can plot the acceleration frequency response, which can be used to verify that the desired acceleration loading was applied via the displacement table array method.
Specifications sometimes call for products to be subject to harmonic acceleration loading applied to the base of the product. If absolute accelerations are to be measured at points on the product, a harmonic finite element analysis would be best run with non-zero acceleration inputs at a base. Ansys supports only non-zero displacement loading at nodes in harmonic models. The desired non-zero harmonic acceleration loading can be converted to displacement as a function of frequency, and the desired load applied in Ansys via a Table Array that is a function of frequency (referred to as TIME in the table array).
Frequency response plots at the base where the loads are applied can be used to confirm that the desired displacement and acceleration loads were input. Frequency response plots at other points in the model can show the absolute (relative to the global origin, not relative to the base) displacement and acceleration results elsewhere in the model.
The usual displacement, stress and strain plots can be generated. Users should note that chosen frequencies and phase angles must be manually entered in many of the results plot object details. In the use of APDL commands as illustrated above, the correct units must be used in the accelerations and displacements, and these units must be employed in the SOLVE of the Workbench model.
I'm curious why only such specific graphics cards, like the NVIDIA Teslas, are supported for GPU acceleration in Ansys Mechanical computations. Secondly, I'm using a Quadro RTX 4000, which is tested as a display card for Ansys and of course does a lovely job of keeping the display snappy. I would love to be able to use it to assist in solving simulations as well, and I was wondering if there's still any way to do that (most of what I'm doing at the moment is in static structural, and I can provide more details about the system I'm running etc. if necessary; I just don't know what would be useful to know, as I'm rather new to Ansys). I don't mind if incorporating a GPU like mine wouldn't give a significant performance boost, but I would like to find out how it can be done if it's at all possible. When I check the box to enable NVIDIA GPU acceleration in the advanced solver settings, the solution returns an error once the mathematical model is built. The parts that seem relevant are the following:
No recommended GPU devices have been detected on machine
DESKTOP-5VIE40K. Only Tesla-series or Quadro
P5000/P6000/GP100/GV100/RTX6000/RTX8000 GPU devices are recommended at
this release. For optimal performance, install a recommended GPU
device in this machine. If you wish to use an alternative GPU device,
please review the recommendations in the section titled "Requirements
for the GPU Accelerator in Mechanical APDL" in the Installation Guide
for your platform.
This is followed by a little warning saying that doing this with new, powerful graphics cards that haven't been tested is just fine, but doing it with weaker cards may actually slow performance. Again, I don't mind this (so long as it's reversible , so I want to try it out for fun. I do want to be sure that I set the environment variable correctly, as this is something I don't do often besides adding the odd thing or two to the system path. So, I just create a new system environment variable called ANSGPU_OVERRIDE and give it the value 1? Or is it something to be done in the "Additional Command Line Arguments section" of the mehcanical solving tab?
Finally, after the environment variable override, should I still expect to encounter a snag with the "NVIDIA Library Requested but not Enabled" error that came up in the solver output, or will that be taken care of by the overriding or HPC or something like that?
Thanks for the information; I went and put in the environment variable, and straight away I could use the GPU acceleration setting without raising any errors. What took me a second to reply was that I also decided to do a little ad-hoc testing with a stopwatch and the taskmanager. When I had GPU acceleration off, there was no utilization during solving whatsoever, as expected: no VRAM usage, no CUDA usage, etc. The simulations ran at a normal speed, though I only have 16 GB of RAM at the moment, so when I did larger solutions, I got a lot of disk utilization (we're talking writing 300+ MB per second for about a third of the total time, and reading back at about 2GB per second towards the end). When I enabled GPU acceleration, building the mathematical model looked about identical, but the CPU indeed hovered at less overall utilization, and we got a lot of CUDA usage, showing that the GPU was indeed computing away! Success! Plus, it used a considerable amount of its VRAM, which I think is a reason why there was a lot less disk utilization when I had the GPU enabled. Both tests took about the same amount of time, though I imagine that once I put more RAM in my machine, the disk usage will lower for the CPU-only calculations and the CPU will be faster over all unless you do indeed have a GPU with excellent double-precision flops. For anyone else following my steps, I ended up not needing any additional toolkits or libraries; just the environment variable. This was a good bit of fun!
93ddb68554