[Nek5000-users] solver tolerances and norm of residual

Tue Nov 16 14:06:38 CST 2010

Hi Paul,

it worked!!

The crucial difference between my file and yours was:

<       parameter (lz2=lz1-2)
>       parameter (lz2=lz1  )

I am a little ashamed for not having seen it but the output was not giving
me any clue.

Thank you so much.

>
> Hi Alex,
>
> I just tried your case - it ran w/o any difficulty.
>
> I wonder if there is an issue w/ your SIZE file?  I've
> placed the one that I used in
>
> www.mcs.anl.gov/~fischer/SIZE
>
> which you can just grab.  Then makenek 3dtest followed by
> the usual run process.... Let me know if this resolves the
> issue for you.
>
> Cheers,
>
> Paul F.
>
>
> On Tue, 16 Nov 2010, nek5000-users at lists.mcs.anl.gov wrote:
>
>> Hi,
>>
>> I think this is not the reason because in a previous attempt to face the
>> 3D problems I tried to solve the 3D lid-driven cavity (no periodic BC)
>> and
>> I obtained the same error: NaN in Helholtz residual.
>>
>> Maybe I should start from the turbChannel example and simplify the
>> problem
>> to reach a laminar 3D poiseuille flow.
>>
>> Anyway, thanks a lot for your answers.
>>
>>> A quick look of the *rea file did not bring up anything.  I recall an
>>> issue with periodic BC's, following is an email from one of the
>>> developers:
>>>
>>>
>>> Hi Frank,
>>>
>>> Yes -- we currently don't support periodic bcs for
>>> anything smaller than a 3 x 3 x 3 array of elements.
>>>
>>> There's a technical reason for this --- it can be
>>> circumvented if really needed, but it would not
>>> fit within the std. context.
>>>
>>> Sorry about that.
>>>
>>> Paul
>>>
>>> On Tue, 2010-11-16 at 12:56 -0500, nek5000-users at lists.mcs.anl.gov
>>> wrote:
>>>> Hi Frank,
>>>>
>>>> The rea file has been attached.
>>>> Meanwhile I'll check the BC...
>>>>
>>>> Thanks
>>>>
>>>>> Hello Alex,
>>>>>
>>>>> I looked through your files & didn't see anything amiss (by the way,
>>>> it
>>>>> would be best to post your *.rea file).  Since it blows up on the
>>>> first
>>>>> attempt to solve for pressure, perhaps the BC's don't satisfy
>>>>> conservation of mass?
>>>>>
>>>>> Cheers,
>>>>> Frank
>>>>>
>>>>>
>>>>>
>>>>> On Tue, 2010-11-16 at 11:40 -0500, nek5000-users at lists.mcs.anl.gov
>>>>> wrote:
>>>>>
>>>>>> Hi Nek users,
>>>>>>
>>>>>> I am new in using Nek5. I have succesfully simualted the lid-driven
>>>>>> flow,
>>>>>> rayleigh-bÃƒÂ©nard in a square cavity and laminar poiseuille flow.
>>>> All of
>>>>>> them were 2d cases. I though it would be straightforward to move
>>>>>> from
>>>> 2D
>>>>>> to 3D simualtions but I have not been able to run a 3D poiseuille
>>>>>> channel.
>>>>>> I am obtaining an error on Helmoltz solver even in the first
>>>> iteration.
>>>>>>
>>>>>> Please, find output, usr and box files attached.
>>>>>> Thanks in advance.
>>>>>>
>>>>>> ..........................................
>>>>>>            Alex Fabregat
>>>>>>         Post-doctoral fellow
>>>>>>         University of Ottawa
>>>>>> Department of Mechanical Engineering
>>>>>> _______________________________________________
>>>>>> Nek5000-users mailing list
>>>>>> Nek5000-users at lists.mcs.anl.gov
>>>>>> https://lists.mcs.anl.gov/mailman/listinfo/nek5000-users
>>>>>
>>>>> --
>>>>> Frank Herbert Muldoon, Ph.D. Mechanical Engineering
>>>>> Technische UniversitÃƒÂ¤t Wien (Vienna University of Technology)
>>>>> Inst. f. StrÃƒÂ¶mungsmechanik und WÃƒÂ¤rmeÃƒÂ¼bertragung (Institute
>>>>> of
>>>> Fluid
>>>>> Mechanics and Heat Transfer)
>>>>> Resselgasse 3
>>>>> 1040 Wien
>>>>> Tel: +4315880132232
>>>>> Fax: +4315880132299
>>>>> Cell:+436765203470
>>>>> fmuldoo (skype)
>>>>> http://tetra.fluid.tuwien.ac.at/fmuldoo/public_html/webpage/frank-muldoon.html
>>>>> _______________________________________________
>>>>> Nek5000-users mailing list
>>>>> Nek5000-users at lists.mcs.anl.gov
>>>>> https://lists.mcs.anl.gov/mailman/listinfo/nek5000-users
>>>>>
>>>>
>>>>
>>>> ..........................................
>>>>            Alex Fabregat
>>>>         Post-doctoral fellow
>>>>         University of Ottawa
>>>> Department of Mechanical Engineering
>>>> _______________________________________________
>>>> Nek5000-users mailing list
>>>> Nek5000-users at lists.mcs.anl.gov
>>>> https://lists.mcs.anl.gov/mailman/listinfo/nek5000-users
>>> --
>>> Frank Herbert Muldoon, Ph.D. Mechanical Engineering
>>> Technische UniversitÃ¤t Wien (Vienna University of Technology)
>>> Inst. f. StrÃ¶mungsmechanik und WÃ¤rmeÃ¼bertragung (Institute of Fluid
>>> Mechanics and Heat Transfer)
>>> Resselgasse 3
>>> 1040 Wien
>>> Tel: +4315880132232
>>> Fax: +4315880132299
>>> Cell:+436765203470
>>> fmuldoo (skype)
>>> http://tetra.fluid.tuwien.ac.at/fmuldoo/public_html/webpage/frank-muldoon.html
>>>
>>> _______________________________________________
>>> Nek5000-users mailing list
>>> Nek5000-users at lists.mcs.anl.gov
>>> https://lists.mcs.anl.gov/mailman/listinfo/nek5000-users
>>>
>>
>>
>> ..........................................
>>           Alex Fabregat
>>        Post-doctoral fellow
>>        University of Ottawa
>> Department of Mechanical Engineering
>>
>> _______________________________________________
>> Nek5000-users mailing list
>> Nek5000-users at lists.mcs.anl.gov
>> https://lists.mcs.anl.gov/mailman/listinfo/nek5000-users
>>_______________________________________________
> Nek5000-users mailing list
> Nek5000-users at lists.mcs.anl.gov
> https://lists.mcs.anl.gov/mailman/listinfo/nek5000-users
>

..........................................
           Alex Fabregat
        Post-doctoral fellow
        University of Ottawa
Department of Mechanical Engineering