[supermanks]

Hi all,

I have a conceptual question which shouldn't take up too much of your time: I'm working on modelling the flow field around a finite-sized body which is being translated and rotated by a fluid (which is pressure-driven) . If I'm looking to model the steady-state situation (for both particle and fluid), can I get away with using a translating frame of reference alone, or would I need to use a rotating frame as well? Also, why?

Any quick ideas? TIA!

Regards

[duri]

As long as there is no linear acceleration along flow direction it can be modeled using MRF in translation. MRF in rotation it depend on the particle geometry. The pressure distribution around geometry should be insensitive to geometry orientation.

[supermanks]

Thanks for your reply, Duri. My thoughts are the same, but I'd like to clarify them with you (if you've the time )using an example. Let's look at two primitive shapes, like a circle, and an ellipse (2D for now).

For the circle, previous studies have shown that it rotates and translates with constant speed at equilibrium. Now, since the shape is angle-invariant (if I may call it that), then that would be a suitable candidate for using the translating reference frame alone. Correct?

However, for the ellipse, again from literature, the linear speed is shown to be constant, but the angular speed is periodic, in that, it rotates fastest when its major axis is vertical, and slowest when horizontal. Now if we observe a point at a fixed distance from its centroid (from the moving frame), we should see that the velocity is almost constant with time, since the linear and angular components more or less add up to the same value, even though the shape is not angle-invariant. Correct?

I'd be totally grateful for any additional insight into how I can look at this. Thanks a lot!!

Regards