Discussion Forum

Dr. Ivar please....

Hello

I'm certainly not the only one on the Forum today or ? :)
Anyhow for me if you can solver the model library FSI model of a pin moving in a fluid, you should as well get a flap moving the same way. but to get vibrations you need a fine mesh and a fine time step so be prepared to have a loong solver sequence.

To easier debug such sequences, turn on the "plot while solving", and to improve convergence, define initial conditions (at least some) closer to the solution, than just "v=0"

--
Good luck
Ivar

We know, Dr. Ivar will answer . [or will only answer ;-)]
Thanks a lot will do that.
regarding initial conditions, I did not understand.

I am giving prescribed displacement by equation y = Asin(omega*t). what should be the initial value closer to the solution which is not zero? Is there any way to work this out?

Thanks a lot for all your helps.

Wish you a great new year 2012.

Siva

Happy new your to you all too ;)

Well no, I cannot always answer, frist of all I do not know everything, COMSOL multiphysics is covering too many domains, and I'm missing a few interesting modules anyhow (such as acoustics, plasma ...), and 2nd I do not have enought time available to answer everything/everyone
I'm primary supposed to use COMSOL for internal developments where I work (and I'm not from COMSOL either ;). Now, I have found out that one learn a lot by studying the problems the FORUM users submit here, so for me I call this training. In fact I can only advise that you should all try it out more intensively.

This said, back to your issue, now I'm not sure I have understood you right, you have a wing profile and a flap in a FSI fluid flow model, and you want to mode the flap in a prescribed way (sinus oscillation in a time series analysis)and see the wing move up and down, driven by the forces from the fluid ?

My first question what is the fixed referentiel ? a prescribed displacement BC is w.r.t. the external geometrical frame, so if the wing is moving up or down the flap must follow and the expression of its motion must follow the linear motion of the CoG of the wing and the wing Theta_Z orientation (I suppose here you are in 2D). Therefore you could need to add more than just a sinus variation.
On the other hand, to keep the wing in the middle of the image you need springs or other forces to compensate for the drag.
As I said I'm not sure I got it all."My way" would be to impose a inlet velocity left to right, an rather large air area, with symmetry on upper bottom boundaries and a fixed wing with a moving flap via a parametric sweep (or a time series as you propose).
To get a reasonable drag force you need a good, dense mesh to resolve the gradient correctly and to integrate the structural normal pressure from the fluid. At least that is what I know from the theory ;) But I must admit I havnt played a lot with these types of simulations.

There are others on the FORUM that have done more, try a search

--
Good luck
Ivar

That is Dr. Ivar.
I got the whole picture. I think I can do this now.

I am sorry for taking right in getting answers.
This forum has many evidences that you are indeed 'mentoring' many people. Irrespective of our level of knowledge and ignorance, you show us what is what.
We are very thankful to you.
There may be many experts in the forum clearing doubts for advanced users who are already experts themselves. But there is only you who clears ignorance of the novice users without any expectation. Hats off to you and what you are doing attributes to our true learning. We will learn and make COMSOL to do what we want.

When ever I use comsol (or even any cfd software), I am surprised that even for a simple 2D flow past circular cylinder, one has 1000s of options to be set to get results that match with theory. Wont it be nice, if makers of these software realise this?
Some how I always remember this story whenever use CFD for my physics:

"Two people came to a small town and told the local people who suffer from starvation that they can make soup from stones. People don't believe them, so they started to put stones in a pan and put some water in it and started cooking and tasting the coup while they're cooking. They give the expression that the soup was so delicious and people started to wonder, then the two said to each other, if we could have just a bit of tomato, the soup will be perfect, someone out of curiosity brought a tomota for them to put in the soup, then they said if we could have just a bit of ...... it'll be...... evenly, the soup was made - all using the material given by local people......
It is said that people selling CFD is like cooking soup using stone - ask you for more and more and eventually make something out of stuff given by the users themselves....."

But your replies to various posts enlighten novice users to put what 'is' needed than what is available.
Hopefully one day we will also learn this and make delicious soups with addition without becoming 'soup boys'.

With lots of regards
Siva

Hi

Thanks for the encourraging words, but you know, from my own mythology (northern scandinavian) we have a nice tale just like yours, but it has been modernised slightly, it was not from stones (those are too common up there) but from nails that the the visitor managed to get the houseowner to fire up an excellent soup (the tale of "spikersuppa")

Also if I'm interested to get more people to study physics and to use COMSOL, as it's really the best way to illustrate what is happenng around us, it is also because I've entered my last decade of active work and with the challenges we all have ahead, we need more skilled persons out here, at large.
Finally, I have more fun Comsoling than plying a "war game" of some sort. What if all those people trying to kill their virtual ennemies, to later better "kill" their true "ennemies" as if the supremacy over the other people here is the greatest, would seach for new solutions and a better understanding of the physics going on all around them at every instant...
For me the important is rather to really understand what is going on on earth, in our universe, that is far richer and interesting than most of these chronophage game plays ;)

This said, for CFD, mostly one can guess quite well the initial conditions (a human being though, not COMSOL alone) and the COMSOL default "all = 0" is mostly a poor starting point. Often its worth spending 5-10 min setting up a first inital guess for velocity and coresponding pressure drop, that increases the speed to convergence drastically.

But I agree this takes some time, and requiress to dig your way back in the textbooks on flow, diffusion, convection heat transfer and so on, just to discover the many ways history has pushed the same equation into different expressions hardly to be recongnised. Here again COMSOl allows us to unify our view, as unit conversion is quickly done.
From my point of view there is a good and intersting job to be done with COMSOL within chemistry, CFD and thermal: transform all these "correlation and dimensionless named numbers, major weapons that engineers use to confuse scientists, named from 19th century scientist such as Sherwood, Stanton, Schmidt, Lewis, Prandtl, Reynolds, Grashof, Péclet, Damköhler, Biot, Fourier ... numbers" (cf. p250 "Diffusion" by E.L. Cussler, Cambridge, 3rd Ed 2011). These number have appeared as they simplify the way to analyse diffusion ,mass transfer and heat transfer (all based on the diffusion diff equation) and can be nicely illustrated with COMSOL. And are powerful ways to validate A CFD model from some simpler hand calculations.

Another topic for your case: Euler and Lagrangian references (and why does one consider solid and fluics so differently ?, apart from historical reasons). Fixing a wing in a moving flow, or fixing a fluid and moving the objects are two ways to look at the same "problem", but one can be easier to solve than the other.

Looking forward for further challenges and more fun COMSOLing in 2012 too ;)

--
Good luck
Ivar