Hi everybody,
does anyone know if there is a plug in to simulate an object being dragged through the water?
I would like to design a wobbler by my own:
Maybe flexhopper?
Thanks for every hint/comment.
Have a nice saturday night everybody!
Maybe flexhopper?
Thanks for every hint/comment.
For “accurate” (kinda) results get a CUDA thing. For rational results mimic some existed wobbler. Big question: why bother digging into the rabbit hole? Bigger question: why killing fish for fun? (what about becoming vegan?)
This might be a starting point:
I already suspected something like that, and that there is a reason why wind and water today still often is simulated in physical models.
Never said I am going to kill them. But if I wouldn’t kill/eat them, wouldn’t I’m just fishing for fun, which would be even worse than catching because I am hungry? hehe.
I am not really into religious/political discussions at all. But I will seriously think about it;)
thanks @martinsiegrist. I will check when I am back at the pc. Looks like something that my 10 year old lenovo laptop could handle:)
For those recommending not to dig into rabbit holes, how would one then catch the rabbit. In other words, try, fail and try again till you have it solved
If it is wobbling through the water, as the name suggests, only full CFD codes are likely to produce accurate results. Anything less would be a very rough approximation.
Even that it’s more or less questionable: Think the mighty Empire (Mercedes/AMG) with budget 10+ times the GNP of Nigeria plus rather unlimited brain (2+ K) resources.
Then think the W13 disaster (not to mention the FIA idiot who “masterminded” the ground effect [i.e. pigs do fly] rules).
Plus (at 1.06) - using then a CRAY:
thanks for your answer!
Is there something you can recommend(having in mind I am no able to spend 10.000s of dollars)?
Could UE be considerable?
Thanks in advance!
The software cost can be zero (OpenFoam), but the opportunity cost will far out-weigh it. Will be cheaper to design and 3D print a few then drag through a pool and iterate till you’re happy with the design.
1 Like
Thanks for the information! So I defintely will get my hands get dirty, which also is nice.
Been doing a lot of CFD in both openfoam and in cloud like simscale. It “should” never be more expensive than physical tests. I’m doing urban wind comfort and a slot in a physical validated windtunnel like this requires weeks if not months scheduling and is a $5,000 -10,000 thing each time.
Cheapest would be go for one or more Epyc cpus as they have a great memory band. Install linux and call your stuff over SSH. You could look into this development for the SSH part although its aimed for daylight with Radiance it could inspire to an openfoam workflow.
Openfoam requires quite some digging for the right turbulence models etc and also requires geometry to be super clean to run. Can take many man hours to cleanup a geometry even if one had a template with all the technical parameters.
For grasshopper there’s ODS Swift, Butterfly and Eddy3d for wind simulations although you’ll need to do proper tweaking to setup simulations that are not “wind around buildings”.
If you want the easiest solution, go with simscale, they have really good templates, examples and documentation for many use cases including urban comfort, combustion engines, Fire evacuation, CPU coolers etc etc…
2 Likes
Dragging though air or water are completely two things as their viscosity is times different.
You don’t need a high-end computer if you don’t want a very precise result, nor in a reasonable time.
2 Likes
Do you mean I can use kangaroo for this, or do you have something else in mind?
I think this conversation is a bit all over the place because the aim isn’t very clearly defined yet.
With something like CFD simulation of drag on a vehicle or building, you need to simulate the full movement of the fluid with measurable properties and return values for the forces exerted on the parts of the object.
If the thing moving through the fluid is also flexible, then you need to also simulate its solid deformations and the 2 way interaction between the solid and the fluid (and even much professional dedicated CFD software doesn’t include this capability).
If you’re creating a presentation animation of fluid flowing around something where you’re showing the fluid, then you might need to simulate some plausible looking ripples and vortices, but not necessarily in any quantitatively meaningful way that you could engineer from. This sort of plausible looking fluid motion is a big and continuously developing field in its own right, quite separately from engineering CFD.
Finally, if you want to show an animation of how something flaps and moves in a steadily flowing fluid, but not the fluid movement itself, you can often skip simulating the fluid altogether and model it as a constant linear force projected onto area elements, and the movement is surprisingly similar. For something like a flag in the wind, the characteristic flapping motion happens without including anything like turbulence or vortices in the simulation. Part of the flag angled against the wind gets more force applied, causing it to pull other parts of the flag around, and so on, and the waves emerge just from that. (eg Kangaroo Wind Control - #3 by DanielPiker)
3 Likes
You’re right and I want to apologize for that! I am just starting with this topic, and the goal may change in the discourse. Since I have already seen so much, partly crazy, things here on the forum and the people here have so many different hobbies, I thought, maybe someone already has experience with it.
that should not be necessary for the time being:)
I’ll try this among all the other things that were prosposed in this thread. Unfortunately I had the last few days very little time and must make up for the weekend
Maybe I’m wrong, but I don’t think you can achieve that by Kangaroo. Water is just different from Wind (supported by Kangaroo). And Kangaroo cannot simulate turbulence, which plays a key role with small objects.