To render images of a water surface in a simulation i'm looking for computer graphics files of such surfaces, in particular those showing a Kelvin Wake or a Bernoulli Hump. These features are generated by surface or sub-surface vessels moving through water. I see quite a bit of research predicting the Kelvin Wake and the Bernoulli Hump and simulations of these might generate 3DS or DXF files that can be used to simulate how these features might show up in an imager. Would any body have such 3DS or DXF files.
Many thanks for any help, N
You could try a program I wrote called zgreen.
It produces wave patterns created by a single or multiple Havelock sources or sinks.
You should be able to create your own 3DS or DXF files because the wave elevations are saved in text format on a regular rectangular grid.
See for example:
https://www.boatdesign.net/threads/zgreen-wave-patterns-made-by-havelock-sources.36112/
You can find 3DS or DXF files of Kelvin Wake or Bernoulli Hump geometries on academic repositories like ResearchGate or Zenodo, especially within CFD-related research projects. Many researchers share their simulation models as supplementary materials in open-access journals or conference papers. Check GrabCAD and Thingiverse for community-shared 3D models; use keywords like “Kelvin wake 3D model” or “Bernoulli Hump CAD.” Platforms like Sketchfab may have visualizations, though not always downloadable in 3DS/DXF. If unavailable, consider using software like SolidWorks or Blender to manually model based on published diagrams. You can also request data from authors via email or through their institutional GitHub repositories. Lastly, forums like CFD Online often host threads with shared simulation files or model links.
Many thanks for you good points, yes i have a program from Leo, but i'm sure there'll be other good ones. Thank you, N.
This code developed by Leo is excellent. I used it to generate the surface disturbance on water, then used Pyvista to generate Delaunay triangles of the surface which i then read into the millimetre wave scene simulation to generate images of the wake.
I just now need to understand what the Havelock sources and sinks are and how representative they are of a disturbance caused by a body moving through water.
Great code developed by Leo!
Looking at the results, from your input programs, i'd say the source was travelling faster than the speed of the waves on the liquid surface. Can you slow the speed of source down to less than the speed of the waves on the surface? I reduced the speed in the input file down to 1.2, 1.0, 0.8 and 0.1. When the speed is less than 1, there appears very little surface disturbanced behind the source. Is the code still representative of surface waves when the source velocity is less than the surface wave velocity, or speep U in the model of 1.0? Many thanks, N.
Dear Leo, would you happen to have a reference to the way you calculate the wake using the Havelock sources, or to the general method of using these sources to predict wake behaviour? Thank you, Neil
Also curious about how the strength of the source is related to real parameters of physical dimensions and speed of submerged body, any clues as to how these are related? I take is the model is dimensionless, so for example if x,y,z were in metres, the velocity U would be in m/s? Cheers, N.
Running your code generates realistic looking figures. However, i was wondering about the units of the quantities that are in the input files, because i've been comparing predictions of your code with that of other codes in journal papers.
The surface displacements seem to be a factor of two to ten larger than those in these publications and the wavelengths of the surface displacements dont seem to agree either. So i was wondering if perhaps speed was in knots and not m/s, and perhaps dimensions are feet, instead of metres. Any help appreciated. Thanks, N.
by way of validating, if i've run your code properly, i compared the results of running your code for Rankine ovoid with those of simulations in the paper on wakes by, Fuduo Xue, et al, IEEE Access, Vol 8, 2020. Comparisons indicate that the features of the images are almost identical, which indicates dimension in metres and the velocities in m/s are alright. However, levels of the surface elevations from your code are around 1.5 to 5 times greater than those in the paper.
Might this over estimation be a feature of using a Rankine ovoid as a simplifications of a more complex model in Fuduo's paper? Any comments would be helpful.
many thanks, Neil
You can find 3DS or DXF files of a Kelvin Wake or Bernoulli Hump on websites like:
Thingiverse (for 3D models)
GrabCAD (for engineering CAD models)
TurboSquid (for 3D models)
CGTrader (for 3D models)
Thank you Tanvir, i did look for files, but i was looking for the free ones, and couldnt find any. I have subsequenctly created DXF files of vessels, using the code supplied Leo. This is actually quite effective, as you determine the wake from your own parameters associated by the vessel. The only thing now is that i need a reference for the code, ie what's in it and if there are any publications showing how it works. For scientific work you need such information for audit purposes. Many thanks, N
- Badges
- Science topic
- Similar topics
- Physical Sciences
- Atomic, Molecular and Optical Physics
- Atoms