Upstream condition

[David Wilkinson]

In a free surface panel method for yacht hull flows, can anyone recommend a
method of preventing upstream propagation of surface waves other than using
a Dawson-style upwind differencing?

Is there any sort of upstream boundary condition on velocities or
free-surface singularity values that will do the trick?

[Bruce Rosen]

I'm very interested to see the answers to this question as well.

Post by David Wilkinson
In a free surface panel method for yacht hull flows, can anyone recommend a
method of preventing upstream propagation of surface waves other than using
a Dawson-style upwind differencing?
Is there any sort of upstream boundary condition on velocities or
free-surface singularity values that will do the trick?

*********************************************************************
Bruce Rosen, President E-mail: ***@panix.com
South Bay Simulations, Inc. http://www.panix.com/~brosen
44 Sumpwams Avenue
Babylon, NY 11702 USA Tel/Fax: [phone]
*********************************************************************

[David Wilkinson]

Bruce

I was hoping you would tell me the answer!

When I run an optimisation for the hull shape for minimum drag at a given
Froude number I can almost get the wave drag to go negative, and come up
with some unlikely hull shapes. I suspect the optimisation is homing in on
errors in the upwind difference formula rather than on a real optimum. I
would like to change to a central difference formula of higher accuracy and
stop the forward waves by some other method. I can do this with my original
pressure panel method but not yet with source or doublet panels. Probably
not trying hard enough!

David

Post by Bruce Rosen
I'm very interested to see the answers to this question as well.

Post by David Wilkinson
In a free surface panel method for yacht hull flows, can anyone recommend a
method of preventing upstream propagation of surface waves other than using
a Dawson-style upwind differencing?
Is there any sort of upstream boundary condition on velocities or
free-surface singularity values that will do the trick?

*********************************************************************
South Bay Simulations, Inc. http://www.panix.com/~brosen
44 Sumpwams Avenue
Babylon, NY 11702 USA Tel/Fax: [phone]
*********************************************************************

[Bruce Rosen]

David,

Well, I rely heavily on upwind, Dawson type of approach to ensure
no upstream waves.

I suppose other methods may work. I am not too familiar with
the way others have tackled this. Seems to me they are stuck
with imposing "no waves at upstream infinity" at only a finite
number of points upstream, which opens the door for spurious waves
which can wiggle through those points and the conditions imposed
at them. Thus for a given Froude number they can identify certain
panel spacings that have to be avoided, to eliminate instabilities.

I wonder if anyone else will share their views on this?

Cheers,
Bruce

*********************************************************************
Bruce Rosen, President E-mail: ***@panix.com
South Bay Simulations, Inc. http://www.panix.com/~brosen
44 Sumpwams Avenue
Babylon, NY 11702 USA Tel/Fax: [phone]
*********************************************************************