[Gfs-users] Define density stratification and the rotation in movement From: Ming Ma - 2012-08-01 15:44 Attachments: Message as HTML ```Dear Gfs Users: I have several questions in density definition and solidmoving module, and I will much appreciate for your help. I want to set a density stratified environment with oscillating wings inside. The density gradient is negative along the z direction, and in some places the gradient should be sharp. The locomotion inside is like a heave and pitch wing, so there is a rotation with time except the U and V definitions. I have three questions as follows: (1) I set the density stratification through the alph in PhysicalPara, and I using many Ti (i=1,2,3...) as the interfaces of fluids with various densities, alph=1/(T1*p1+T2*p2+....(1-T1-T2-T3...Tn-1)*pn), so when density gradient becomes very sharp, it needs many Ti, how can I handle this problem? Can that be fulfilled just inside .gfs script? Or there is any better definition of density stratification? (2)I once fulfilled a vertically oscillating sphere inside the stratified flow, but if the interface Ti is not configured properly, especially the interface cuts the sphere's solid boundary, it will always warn: CFL is too large and Gfs-ERROR **: file vof.c: line 1605 (f_over_dV): assertion failed: (GFS_VALUE (cell, p->vpar.v) > 0.), abort! I do not know what this sentence means, because sometimes I move the interface's initial position, it can still run normally, even the sphere interference with the interface later. (3) How to define the movement of rotation with the solidmoving module? I want to add a rotation each time step, except the movement on U,V. Thank you for some suggestions~ -Ming ```
 Re: [Gfs-users] Define density stratification and the rotation in movement From: Geordie McBain - 2012-08-01 16:07 ```2012/8/1 Ming Ma : > Dear Gfs Users: > > I have several questions in density definition and solidmoving module, and I > will much appreciate for your help. I want to set a density stratified > environment with oscillating wings inside. The density gradient is negative > along the z direction, and in some places the gradient should be sharp. The > locomotion inside is like a heave and pitch wing, so there is a rotation > with time except the U and V definitions. I have three questions as follows: > (1) I set the density stratification through the alph in PhysicalPara, and I > using many Ti (i=1,2,3...) as the interfaces of fluids with various > densities, alph=1/(T1*p1+T2*p2+....(1-T1-T2-T3...Tn-1)*pn), so when density > gradient becomes very sharp, it needs many Ti, how can I handle this > problem? Can that be fulfilled just inside .gfs script? Or there is any > better definition of density stratification? How about using the Oberbeck-Boussinesq approximation? http://en.wikipedia.org/wiki/Boussinesq_approximation_%28buoyancy%29 Then alpha is left alone but a buoyancy force is supplied as a GfsSource proportional to the value of a GfsVariableTracer which represents temperature or salinity or whatever it is that is responsible for the stratification. ```
 Re: [Gfs-users] Define density stratification and the rotation in movement From: Stephane Popinet - 2012-08-01 21:35 ```Hi Ming, > (1) I set the density stratification through the alph in PhysicalPara, and > I using many Ti (i=1,2,3...) as the interfaces of fluids with various > densities, alph=1/(T1*p1+T2*p2+....(1-T1-T2-T3...Tn-1)*pn), so when density > gradient becomes very sharp, it needs many Ti, how can I handle this > problem? Can that be fulfilled just inside .gfs script? Or there is any > better definition of density stratification? What are your Ti's? from what you write, I guess they are VariableTracerVOF? why? If your density stratification is continuous (e.g. density is defined through an equation of state dependent on diffusive tracers such as temperature and/or salinity) then you should use VariableTracer. You can then set that as a continuous function of depth (for example). > (2)I once fulfilled a vertically oscillating sphere inside the stratified > flow, but if the interface Ti is not configured properly, especially the > interface cuts the sphere's solid boundary, it will always warn: CFL is too > large and Gfs-ERROR **: file vof.c: line 1605 (f_over_dV): assertion > failed: (GFS_VALUE (cell, p->vpar.v) > 0.), abort! I do not know what this > sentence means, because sometimes I move the interface's initial position, > it can still run normally, even the sphere interference with the interface > later. You should not have this problem if you use VariableTracer. > (3) How to define the movement of rotation with the solidmoving module? I > want to add a rotation each time step, except the movement on U,V. See http://gfs.sourceforge.net/wiki/index.php/GfsSolidMoving http://gfs.sourceforge.net/wiki/index.php/GfsSurfaceBc cheers Stephane ```
 Re: [Gfs-users] Define density stratification and the rotation in movement From: Pei Shui - 2012-08-02 02:39 Attachments: Message as HTML ```About question3 I will release my solid moving module this weekend.. On 1 August 2012 16:44, Ming Ma wrote: > Dear Gfs Users: > > I have several questions in density definition and solidmoving module, and > I will much appreciate for your help. I want to set a density stratified > environment with oscillating wings inside. The density gradient is negative > along the z direction, and in some places the gradient should be sharp. The > locomotion inside is like a heave and pitch wing, so there is a rotation > with time except the U and V definitions. I have three questions as follows: > > (1) I set the density stratification through the alph in PhysicalPara, and > I using many Ti (i=1,2,3...) as the interfaces of fluids with various > densities, alph=1/(T1*p1+T2*p2+....(1-T1-T2-T3...Tn-1)*pn), so when density > gradient becomes very sharp, it needs many Ti, how can I handle this > problem? Can that be fulfilled just inside .gfs script? Or there is any > better definition of density stratification? > > (2)I once fulfilled a vertically oscillating sphere inside the stratified > flow, but if the interface Ti is not configured properly, especially the > interface cuts the sphere's solid boundary, it will always warn: CFL is too > large and Gfs-ERROR **: file vof.c: line 1605 (f_over_dV): assertion > failed: (GFS_VALUE (cell, p->vpar.v) > 0.), abort! I do not know what this > sentence means, because sometimes I move the interface's initial position, > it can still run normally, even the sphere interference with the interface > later. > > (3) How to define the movement of rotation with the solidmoving module? I > want to add a rotation each time step, except the movement on U,V. > > Thank you for some suggestions~ > > -Ming > > > > > > ------------------------------------------------------------------------------ > Live Security Virtual Conference > Exclusive live event will cover all the ways today's security and > threat landscape has changed and how IT managers can respond. Discussions > will include endpoint security, mobile security and the latest in malware > threats. http://www.accelacomm.com/jaw/sfrnl04242012/114/50122263/ > _______________________________________________ > Gfs-users mailing list > Gfs-users@... > https://lists.sourceforge.net/lists/listinfo/gfs-users > > ```