[aspect-devel] boundary pb

[Cedric Thieulot]

I confirm tht this seems indeed controlled by alpha. alpha=10^-8 yields a symmetric solution...


On 18 Sep 2014, at 16:06, Rene Gassmoeller <rengas at gfz-potsdam.de> wrote:

> Hi all,
> I can confirm Cedric's observations, and I think it stems from the
> scaling of gravity and thermal expansivity to 10^10 ... if you lower
> this to 10^5 the asymmetry vanishes (at least here in my model). Might
> it have to do with our pressure_scaling, which is calculated as
> viscosity over length_scale and that does not change, when you increase
> gravity, although the pressure will increase by this? Any ideas?
> 
> Best,
> Rene
> 
> On 09/18/2014 02:47 PM, Thieulot, C. (Cedric) wrote:
>> Hi Wolfgang,
>> 
>> sorry for the late reply, I somehow missed your second email.
>> 
>> 
>> Do I interpret your pictures correctly that
>> 
>> - there is/should be an exact left-right symmetry in the velocity field
>> 
>> yes. the initial temp field is given by (1-z)-0.01cos(pi x) sin(pi x)  in a [0,1]x[0,1] box
>> 
>> - the boundary layer exists on only one side?
>> 
>> indeed, always on the right.
>> 
>> I hereby attach three snapshots at three levels of refinements (5,6,7). It becomes more and more visible with a higher resolution.
>> 
>> Best,
>> 
>> C.
>> 
>> [cid:E9CB0E90-9C6D-4F79-ACB6-4F4E24D226C6 at geo.uu.nl][cid:83AB99DD-0EE5-4498-A172-0B00C55E2FF4 at geo.uu.nl][cid:FFC9D13E-F7C6-434B-9BE0-4D6A427B5393 at geo.uu.nl]
>> 
>> 
>> 
>> _______________________________________________
>> Aspect-devel mailing list
>> Aspect-devel at geodynamics.org
>> http://lists.geodynamics.org/cgi-bin/mailman/listinfo/aspect-devel
>> 
> _______________________________________________
> Aspect-devel mailing list
> Aspect-devel at geodynamics.org
> http://lists.geodynamics.org/cgi-bin/mailman/listinfo/aspect-devel