[CIG-LONG] Help! SNAC Benchmark: Parallel-Plate Viscometer
Eunseo Choi
echoi at ldeo.columbia.edu
Thu Aug 12 11:39:02 PDT 2010
Hi, Jean
I realized that the back and front vel bcs were not specified in your
input file.
After I added free-slip conditions for those z-normal boundaries (my
input is attached), vel fields now look similar to those given in the
manual (see the attached figure).
Stress fields still look different, though. My guess is that the
definition of stress invariant is not consistent, which has been changed
during the course of SNAC development.
However, I wouldn't worry too much about what was the definition as long
as velocities are reasonably accurate.
Do you have access to Jaeger's book referred to in the manual?
Eunseo
Jean Liu wrote:
> Hi Eunseo,
>
> Is it possible for you to run my input file for the benchmark and send
> me the output files you get? I want to check to see if it's a problem
> with my installation of SNAC.. I attached my input file to this email.
>
> Thank you for all of your help and patience! =)
>
> - Jean
>
>
> Quoting Eunseo Choi <echoi at ldeo.columbia.edu>:
>
>> Hi, Jean
>>
>> The "tetra shear stress" is the second invariant of stress and "max.
>> shear stress" is 0.5*(Sig1-Sig3), where Sig1 and 3 are the largest and
>> the least principal stresses.
>> For comparison with the figure in the manual, please plot "tetra shear
>> stress".
>>
>> "shear/normal stress @ 30d" are used only in special cases.
>> Sorry for the confusing naming. I'll try to improve them in the future.
>>
>> Eunseo
>>
>> Jean Liu wrote:
>>> Hi Eunseo,
>>>
>>> In the no-slip case, my velocities are still greatest on the wall
>>> boundary condition (instead of where it's supposed to be pushing
>>> out in the x-direction), and I get the shear and normal stress from
>>> the SNAC output, visualized on Paraview.
>>>
>>> How do you get stress invariant as an output? When I load the .vts
>>> and .pvts files into Paraview, I get the option of plotting:
>>>
>>> - plastic strain
>>> - strain rate
>>> - pressure
>>> - hydrostatic pressure
>>> - tetra shear stress
>>> - max. shear stress
>>> - von mises stress
>>> - shear stress @30d
>>> - normal stress @30d
>>> - failure potential @30d
>>> - phase
>>> - viscosity
>>> - velocity
>>> - force
>>> - temperature
>>>
>>>
>>>
>>>
>>>
>>> Quoting Eunseo Choi <echoi at ldeo.columbia.edu>:
>>>
>>>> I see. Then, the stress value you compute must be different from
>>>> what's
>>>> plotted in the manual.
>>>>
>>>> First of all, velocity field looks better in the no-slip case?
>>>> And, what does the plot of stress invariant look like?
>>>> How did you get the shear and normal components of stress?
>>>>
>>>>
>>>> Eunseo
>>>>
>>>>
>>>>
>>>> On 08/11/2010 05:04 PM, Jean Liu wrote:
>>>>> Hi Eunseo!
>>>>>
>>>>> I tried that before, as well, and I ended up getting an
>>>>> asymmetric pattern for stress.. (I attached the screenshot to
>>>>> this email)
>>>>>
>>>>> - Jean
>>>>>
>>>>>
>>>>>
>>>>>
>>>>> Quoting Eunseo Choi <echoi at ldeo.columbia.edu>:
>>>>>
>>>>>> Hi, Jean
>>>>>>
>>>>>> No-slip conditions should be added on the moving walls.
>>>>>> So, insert the following block of XML code into each of the top and
>>>>>> bottom BC sections:
>>>>>> ==================
>>>>>> <struct>
>>>>>> <param name="name"> vx </param>
>>>>>> <param name="type"> double </param>
>>>>>> <param name="value"> 0.0 </param>
>>>>>> </struct>
>>>>>> <struct>
>>>>>> <param name="name"> vz </param>
>>>>>> <param name="type"> double </param>
>>>>>> <param name="value"> 0.0 </param>
>>>>>> </struct>
>>>>>> ==================
>>>>>>
>>>>>> And please update me on the results!
>>>>>> Thanks.
>>>>>>
>>>>>> Eunseo
>>>>>>
>>>>>> ps. Do you mind posting this reply on the cig-long list?
>>>>>>
>>>>>>
>>>>>> Jean Liu wrote:
>>>>>>> Hi Eunseo!
>>>>>>>
>>>>>>> I thought it was straightforward, too, but when I ran it, my
>>>>>>> velocity vectors looked a bit different from yours: my
>>>>>>> velocities were greater at the wall velocity boundary
>>>>>>> condition, then at the part where the material is being
>>>>>>> squeezed out. Plus, to compare to the second invariant of
>>>>>>> stress, I just added the shear and normal components on
>>>>>>> paraview and got different numerical values (but similar
>>>>>>> patterns).
>>>>>>>
>>>>>>> Could you take a look at my input file? I'm not sure if it's a
>>>>>>> problem with my installation of SNAC (when I ran make check,
>>>>>>> there were quite a few Fails), or if everything is okay..
>>>>>>>
>>>>>>> I also attached a screenshot of my Paraview results.
>>>>>>>
>>>>>>> Thanks for all of you help again, Eunseo!! I'm sorry for being
>>>>>>> such a nuisance..
>>>>>>>
>>>>>>> - Jean
>>>>>>>
>>>>>>>
>>>>>>>
>>>>>>>
>>>>>>>
>>>>>>> Quoting Eunseo Choi <echoi at ldeo.columbia.edu>:
>>>>>>>
>>>>>>>> Dear Jean
>>>>>>>>
>>>>>>>> I produced the benchmark results but don't have the original
>>>>>>>> input file.
>>>>>>>> As far as I remember, it should be straightforward to construct an
>>>>>>>> input file for that problem.
>>>>>>>> What are the problems you ran into?
>>>>>>>>
>>>>>>>> Eunseo
>>>>>>>>
>>>>>>>> Jean Liu wrote:
>>>>>>>>> Hi!
>>>>>>>>>
>>>>>>>>> I was wondering if anyone has the input file for the SNAC
>>>>>>>>> benchmark parallel-plate viscometer problem that was used
>>>>>>>>> for the figures in the SNAC manual?
>>>>>>>>> I'm trying to get a comparison for my output, but I'm running
>>>>>>>>> into a few problems...
>>>>>>>>>
>>>>>>>>> Thanks!
>>>>>>>>>
>>>>>>>>> - Jean
>>>>>>>>>
>>>>>>>>> _______________________________________________
>>>>>>>>> CIG-LONG mailing list
>>>>>>>>> CIG-LONG at geodynamics.org
>>>>>>>>> http://geodynamics.org/cgi-bin/mailman/listinfo/cig-long
>>>>>>>>>
>>>>>>>>
>>>>>>>>
>>>>>>>> --
>>>>>>>> =====================================
>>>>>>>> Eunseo Choi
>>>>>>>> Post-Doctoral Research Scientist
>>>>>>>>
>>>>>>>> Lamont-Doherty Earth Observatory
>>>>>>>> 305c Oceanography
>>>>>>>> P.O. Box 1000
>>>>>>>> 61 Rt. 9W
>>>>>>>> Palisades, NY 10964-8000
>>>>>>>>
>>>>>>>> http://www.ldeo.columbia.edu/~echoi
>>>>>>>> Phone: 845-365-8813 Fax: 845-365-8156
>>>>>>>> =====================================
>>>>>>>
>>>>>>>
>>>>>>> ------------------------------------------------------------------------
>>>>>>>
>>>>>>
>>>>>>
>>>>>> --
>>>>>> =====================================
>>>>>> Eunseo Choi
>>>>>> Post-Doctoral Research Scientist
>>>>>>
>>>>>> Lamont-Doherty Earth Observatory
>>>>>> 305c Oceanography
>>>>>> P.O. Box 1000
>>>>>> 61 Rt. 9W
>>>>>> Palisades, NY 10964-8000
>>>>>>
>>>>>> http://www.ldeo.columbia.edu/~echoi
>>>>>> Phone: 845-365-8813 Fax: 845-365-8156
>>>>>> =====================================
>>>>>
>>>>>
>>>>
>>>>
>>>> --
>>>> =====================================
>>>> Eunseo Choi
>>>> Post-Doctoral Research Scientist
>>>>
>>>> Lamont-Doherty Earth Observatory
>>>> 305C Oceanography
>>>> P.O. Box 1000
>>>> 61 Rt. 9W
>>>> Palisades, NY 10964-8000
>>>>
>>>> http://www.ldeo.columbia.edu/~echoi
>>>> Phone: 845-365-8813 Fax: 845-365-8156
>>>> =====================================
>>>
>>>
>>
>>
>> --
>> =====================================
>> Eunseo Choi
>> Post-Doctoral Research Scientist
>>
>> Lamont-Doherty Earth Observatory
>> 305c Oceanography
>> P.O. Box 1000
>> 61 Rt. 9W
>> Palisades, NY 10964-8000
>>
>> http://www.ldeo.columbia.edu/~echoi
>> Phone: 845-365-8813 Fax: 845-365-8156
>> =====================================
>
>
--
=====================================
Eunseo Choi
Post-Doctoral Research Scientist
Lamont-Doherty Earth Observatory
305c Oceanography
P.O. Box 1000
61 Rt. 9W
Palisades, NY 10964-8000
http://www.ldeo.columbia.edu/~echoi
Phone: 845-365-8813 Fax: 845-365-8156
=====================================
-------------- next part --------------
A non-text attachment was scrubbed...
Name: Results_EC.png
Type: image/png
Size: 38620 bytes
Desc: not available
Url : http://geodynamics.org/pipermail/cig-long/attachments/20100812/6a8105f9/attachment-0001.png
-------------- next part --------------
A non-text attachment was scrubbed...
Name: benchmark2_EC.xml
Type: text/xml
Size: 6244 bytes
Desc: not available
Url : http://geodynamics.org/pipermail/cig-long/attachments/20100812/6a8105f9/attachment-0001.bin
More information about the CIG-LONG
mailing list