[Nek5000-users] Nek5000-users Digest, Vol 103, Issue 12

nek5000-users at lists.mcs.anl.gov nek5000-users at lists.mcs.anl.gov
Mon Sep 18 07:03:59 CDT 2017


Daniel,


If you want a custom distribution in any of the directions, you just list

a positive value, nel,  for the number of elements in that direction, and then

provide a list with nel+1 entries.   Note, the list of entries must be shorter than

80 characters, so you have to wrap the line if it goes over 80 characters.  This

is because I wrote a parser to allow for comments in the box files and as such

it only parses 80 character strings.    Just so long as the sequence of strings has

nel+1 values, it will keep reading the strings till satisfied.  It will then go on to

the next required read (e.g., the next list of element locations for the next

direction).


2D example:


-2    9    # 2 elements in x, 9 in y

0 1  0.5     x0,x1, ratio:  element 2 is half the size of element 1

0 .1 .2 .3

.4 .5 .6 .7

.8 .9              coordinates for y  (comments only on last string)


etc


Paul


________________________________
From: Nek5000-users <nek5000-users-bounces at lists.mcs.anl.gov> on behalf of nek5000-users at lists.mcs.anl.gov <nek5000-users at lists.mcs.anl.gov>
Sent: Monday, September 18, 2017 4:02:28 AM
To: nek5000-users at lists.mcs.anl.gov
Subject: Re: [Nek5000-users] Nek5000-users Digest, Vol 103, Issue 12


Hi Paul,


Sorry for not replying earlier, I didn't see your reply last week unfortunately.  So in my case I'm creating a partition as follows:


Box_1
-20 -50                    nel_1, nel_2
0. 50.  0.8                x0, x1
0. 100.  1.1               y0, y1
W  ,E  ,W  ,W              (bc's) velocity (west, east, south, north)
T  ,E  ,t  ,W              (bc's) temperature (west, east, south, north)
Box_2
-20 -50                    nel_1, nel_2
50. 100.  1.2              x0, x1
0. 100.  1.1               y0, y1
E  ,W  ,W  ,W              (bc's) velocity (west, east, south, north)
E  ,T  ,t  ,W              (bc's) temperature (west, east, south, north)


This makes two boxes, separated in the x-direction.  Is what you're saying, then, that my two 20x50 boxes are stacked one on top of the other to, in effect, create one 20x100 box?  If that's the case, how do I take my element map (which is just a list of 2000 numbers mapping local to global elements) and make sure that my two boxes are concatenated in the x-direction, instead of the z?


Thanks in advance,


Daniel

________________________________
From: Nek5000-users <nek5000-users-bounces at lists.mcs.anl.gov> on behalf of nek5000-users-request at lists.mcs.anl.gov <nek5000-users-request at lists.mcs.anl.gov>
Sent: 15 September 2017 13:05:25
To: nek5000-users at lists.mcs.anl.gov
Subject: Nek5000-users Digest, Vol 103, Issue 12

Send Nek5000-users mailing list submissions to
        nek5000-users at lists.mcs.anl.gov

To subscribe or unsubscribe via the World Wide Web, visit
        https://lists.mcs.anl.gov/mailman/listinfo/nek5000-users
or, via email, send a message with subject or body 'help' to
        nek5000-users-request at lists.mcs.anl.gov

You can reach the person managing the list at
        nek5000-users-owner at lists.mcs.anl.gov

When replying, please edit your Subject line so it is more specific
than "Re: Contents of Nek5000-users digest..."


Today's Topics:

   1. Re: time integration scheme for temperature/scalar equations
      (nek5000-users at lists.mcs.anl.gov)
   2. Re: Nek5000-users Digest, Vol 103, Issue 9
      (nek5000-users at lists.mcs.anl.gov)


----------------------------------------------------------------------

Message: 1
Date: Fri, 15 Sep 2017 13:02:20 +0200
From: nek5000-users at lists.mcs.anl.gov
To: nek5000-users at lists.mcs.anl.gov
Subject: Re: [Nek5000-users] time integration scheme for
        temperature/scalar equations
Message-ID:
        <mailman.839.1505473384.10190.nek5000-users at lists.mcs.anl.gov>
Content-Type: text/plain; charset="windows-1252"; Format="flowed"

Hi,
To be more precise. Routines "makeabq" and "makebdq" are called if you
use hemholtz solver for passive scalars. The other option is cvode,
which uses slightly different scheme. I guess hemholtz is default and to
use cvode you have to add CVODE to preprocessing flag. About the link to
doxygen. This one is outdated so it could be dangerous to use it. There
is doxygen showing current beta2 version under
https://adampep.github.io/KTH_Framework/files.html
Regards
Adam

On 15/09/17 12:11, nek5000-users at lists.mcs.anl.gov wrote:
>
> Hi Bolun Xu,
>
> Yes,
>
> you can see that in the conduct.f file( in which convection-diffusion
> equation for temperature is solved):
> http://www.mcs.anl.gov/research/projects/nek5/docs/html/conduct_8f_source.html
>
> In the subroutine "makeabq"  the same Adam-Bashforth coefficients AB()
> are called and in the subroutine "makebdq", the BD() coefficients are
> used again.
>
> best,
> Pierre-Emmanuel
>
>
> On 2017-09-15 11:47, nek5000-users at lists.mcs.anl.gov wrote:
>> Hi! All,
>>
>> I learned from the subroutines makeabf and makebdf in navier1.f that the
>> time integration scheme for momentum equation is the AB-CN scheme, that
>> explicit schemes for nonlinear and forcing term, and implicit ones for
>> viscous term. The prefactors of those schemes are stored in AB() and BD()
>> arrays. Now my question is, whether the temperature/scalar equations also
>> use the same schemes as momentum ones and also share the same prefactors in
>> AB() / BD() arrays.
>>
>> Any help is appreciated.
>>
>> Regards.
>>
>> ----------------------------------------------------------------
>>
>> Bolun Xu
>>
>> University of Science and Technology of China
>>
>> Department of Modern Mechanics
>>
>> Hefei, Anhui, China
>>
>>
>>
>>
>>
>>
>> _______________________________________________
>> 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


-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://lists.mcs.anl.gov/pipermail/nek5000-users/attachments/20170915/b385ae23/attachment-0001.html>

------------------------------

Message: 2
Date: Fri, 15 Sep 2017 07:51:15 +0000
From: nek5000-users at lists.mcs.anl.gov
To: "nek5000-users at lists.mcs.anl.gov"
        <nek5000-users at lists.mcs.anl.gov>
Subject: Re: [Nek5000-users] Nek5000-users Digest, Vol 103, Issue 9
Message-ID:
        <mailman.845.1505477125.10190.nek5000-users at lists.mcs.anl.gov>
Content-Type: text/plain; charset="us-ascii"

Hi Goran,


First of all thanks for having a look, my box file is as follows:


plume_v2_largeIni.rea
2                      spatial dimension
2                      number of fields
#
#
#
#========================================================
#
Box_1
-20 -50                    nel_1, nel_2
0. 50.  0.8                x0, x1
0. 100.  1.1               y0, y1
W  ,E  ,W  ,W              (bc's) velocity (west, east, south, north)
T  ,E  ,t  ,W              (bc's) temperature (west, east, south, north)
Box_2
-20 -50                    nel_1, nel_2
50. 100.  1.2              x0, x1
0. 100.  1.1               y0, y1
E  ,W  ,W  ,W              (bc's) velocity (west, east, south, north)
E  ,T  ,t  ,W              (bc's) temperature (west, east, south, north)

I think where I'm struggling is in working out how the output (0.f#####) files are structured, or how their structure relates to the actual solution.  I've managed to get the (2000, in this case) element mappings out of the files, I just can't then get this mapping to actually map the elements correctly.

Thanks again,

Daniel


________________________________
From: Nek5000-users <nek5000-users-bounces at lists.mcs.anl.gov> on behalf of nek5000-users-request at lists.mcs.anl.gov <nek5000-users-request at lists.mcs.anl.gov>
Sent: 14 September 2017 17:29:13
To: nek5000-users at lists.mcs.anl.gov
Subject: Nek5000-users Digest, Vol 103, Issue 9

Send Nek5000-users mailing list submissions to
        nek5000-users at lists.mcs.anl.gov

To subscribe or unsubscribe via the World Wide Web, visit
        https://lists.mcs.anl.gov/mailman/listinfo/nek5000-users
or, via email, send a message with subject or body 'help' to
        nek5000-users-request at lists.mcs.anl.gov

You can reach the person managing the list at
        nek5000-users-owner at lists.mcs.anl.gov

When replying, please edit your Subject line so it is more specific
than "Re: Contents of Nek5000-users digest..."


Today's Topics:

   1. Problems with post-processing multiple-boxes in MATLAB and
      Python (nek5000-users at lists.mcs.anl.gov)
   2. Re: Problems with post-processing multiple-boxes in MATLAB
      and Python (nek5000-users at lists.mcs.anl.gov)
   3. Re: Problems with post-processing multiple-boxes in MATLAB
      and Python (nek5000-users at lists.mcs.anl.gov)
   4. Re: proj_ortho error in pipe flow
      (nek5000-users at lists.mcs.anl.gov)
   5. problem with applying boundary conditions
      (nek5000-users at lists.mcs.anl.gov)


----------------------------------------------------------------------

Message: 1
Date: Thu, 14 Sep 2017 14:57:27 +0000
From: nek5000-users at lists.mcs.anl.gov
To: "nek5000-users at lists.mcs.anl.gov"
        <nek5000-users at lists.mcs.anl.gov>
Subject: [Nek5000-users] Problems with post-processing multiple-boxes
        in MATLAB and Python
Message-ID:
        <mailman.763.1505402018.10190.nek5000-users at lists.mcs.anl.gov>
Content-Type: text/plain; charset="iso-8859-1"

Hi All,


I am using Nek5000 to simulate a buoyant plume rising from the centre of the floor of a square 2D domain.  I am trying to use MATLAB and/or Python to post-process my data.  Everything works fine when I post-process a simulation that was initialised with only one box in my .box file.  When, however, I run a simulation with multiple boxes (i.e. to cluster nodes), the local to global mapping of elements appears to go wrong and puts data into the wrong places.  When plotting a contour plot of coordinates for example, where I would expect a constantly changing colour gradient, I instead get back colour gradients in four quarters.


I haven't used this list before so if I need to provide examples of images or code I can do.  If anybody else has had this problem and has found a way to overcome it by finding the correct mapping some help would be very much appreciated.


Best regards,


Daniel Ward
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://lists.mcs.anl.gov/pipermail/nek5000-users/attachments/20170914/94fe07e8/attachment-0001.html>

------------------------------

Message: 2
Date: Thu, 14 Sep 2017 15:20:27 +0000
From: nek5000-users at lists.mcs.anl.gov
To: shivang agarwal <nek5000-users at lists.mcs.anl.gov>
Subject: Re: [Nek5000-users] Problems with post-processing
        multiple-boxes in MATLAB and Python
Message-ID:
        <mailman.765.1505402478.10190.nek5000-users at lists.mcs.anl.gov>
Content-Type: text/plain; charset="utf-8"

Hi Daniel,

I've had a similar problem before. Can you send your .box file so we can have a look?

Best regards,

Goran

On Sep 14, 2017, at 11:13, "nek5000-users at lists.mcs.anl.gov<mailto:nek5000-users at lists.mcs.anl.gov>" <nek5000-users at lists.mcs.anl.gov<mailto:nek5000-users at lists.mcs.anl.gov>> wrote:


Hi All,


I am using Nek5000 to simulate a buoyant plume rising from the centre of the floor of a square 2D domain.  I am trying to use MATLAB and/or Python to post-process my data.  Everything works fine when I post-process a simulation that was initialised with only one box in my .box file.  When, however, I run a simulation with multiple boxes (i.e. to cluster nodes), the local to global mapping of elements appears to go wrong and puts data into the wrong places.  When plotting a contour plot of coordinates for example, where I would expect a constantly changing colour gradient, I instead get back colour gradients in four quarters.


I haven't used this list before so if I need to provide examples of images or code I can do.  If anybody else has had this problem and has found a way to overcome it by finding the correct mapping some help would be very much appreciated.


Best regards,


Daniel Ward

_______________________________________________
Nek5000-users mailing list
Nek5000-users at lists.mcs.anl.gov<mailto:Nek5000-users at lists.mcs.anl.gov>
https://lists.mcs.anl.gov/mailman/listinfo/nek5000-users
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://lists.mcs.anl.gov/pipermail/nek5000-users/attachments/20170914/e8a271d0/attachment-0001.html>

------------------------------

Message: 3
Date: Thu, 14 Sep 2017 15:33:07 +0000
From: nek5000-users at lists.mcs.anl.gov
To: "nek5000-users at lists.mcs.anl.gov"
        <nek5000-users at lists.mcs.anl.gov>
Subject: Re: [Nek5000-users] Problems with post-processing
        multiple-boxes in MATLAB and Python
Message-ID:
        <mailman.767.1505403191.10190.nek5000-users at lists.mcs.anl.gov>
Content-Type: text/plain; charset="us-ascii"


Hi All,


Genbox generates a tensor-product array of elements with layout:


   (ix , iy, iz)    (ix fastest moving, iy 2nd fastest, iz 3rd fastest)


If you want to have multiple boxes that continue in this ordering, you could

do the following, say:


box (1)

-5 -5 -5

-1 1 1.0

-1 1 1.0

0 2   1.0


box(2)

-5 -5 -10

-1 1 1

-1 1 1

1.0 9.0


etc


This would stack a 5x5x10 box on top of a 5x5x5 box


and topologically is the same as a 5x5 x (15) box


... i.e., things get concatenated in the z direction.



Also, note that you can use custom distributions in each direction, e.g.,


box

3  3  -5

0 0.1 0.3  10

0  .9   22 33

-5  9  1.5


which would give spacing [ 0 .1 .3 10] in the x direction


[0 .9 22 33] in y


and geometric partition of 5 elements in z on [-5:9]

with size ratio 1.5 between each adjacent z level.


hth


Paul


________________________________
From: Nek5000-users <nek5000-users-bounces at lists.mcs.anl.gov> on behalf of nek5000-users at lists.mcs.anl.gov <nek5000-users at lists.mcs.anl.gov>
Sent: Thursday, September 14, 2017 10:20:27 AM
To: shivang agarwal
Subject: Re: [Nek5000-users] Problems with post-processing multiple-boxes in MATLAB and Python

Hi Daniel,

I've had a similar problem before. Can you send your .box file so we can have a look?

Best regards,

Goran

On Sep 14, 2017, at 11:13, "nek5000-users at lists.mcs.anl.gov<mailto:nek5000-users at lists.mcs.anl.gov>" <nek5000-users at lists.mcs.anl.gov<mailto:nek5000-users at lists.mcs.anl.gov>> wrote:


Hi All,


I am using Nek5000 to simulate a buoyant plume rising from the centre of the floor of a square 2D domain.  I am trying to use MATLAB and/or Python to post-process my data.  Everything works fine when I post-process a simulation that was initialised with only one box in my .box file.  When, however, I run a simulation with multiple boxes (i.e. to cluster nodes), the local to global mapping of elements appears to go wrong and puts data into the wrong places.  When plotting a contour plot of coordinates for example, where I would expect a constantly changing colour gradient, I instead get back colour gradients in four quarters.


I haven't used this list before so if I need to provide examples of images or code I can do.  If anybody else has had this problem and has found a way to overcome it by finding the correct mapping some help would be very much appreciated.


Best regards,


Daniel Ward

_______________________________________________
Nek5000-users mailing list
Nek5000-users at lists.mcs.anl.gov<mailto:Nek5000-users at lists.mcs.anl.gov>
https://lists.mcs.anl.gov/mailman/listinfo/nek5000-users
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://lists.mcs.anl.gov/pipermail/nek5000-users/attachments/20170914/4ab9769c/attachment-0001.html>

------------------------------

Message: 4
Date: Thu, 14 Sep 2017 15:34:39 +0000
From: nek5000-users at lists.mcs.anl.gov
To: "nek5000-users at lists.mcs.anl.gov"
        <nek5000-users at lists.mcs.anl.gov>
Subject: Re: [Nek5000-users] proj_ortho error in pipe flow
Message-ID:
        <mailman.770.1505403282.10190.nek5000-users at lists.mcs.anl.gov>
Content-Type: text/plain; charset="windows-1252"


Hi Steffen,


Thanks for bringing this to our attention.


Adam and I have been looking into this... (no doubt, it's something

stupid done by me awhile back).


Hope to have it resolved by next week.


Best,


Paul


________________________________
From: Nek5000-users <nek5000-users-bounces at lists.mcs.anl.gov> on behalf of nek5000-users at lists.mcs.anl.gov <nek5000-users at lists.mcs.anl.gov>
Sent: Thursday, September 14, 2017 8:18:27 AM
To: nek5000-users at lists.mcs.anl.gov
Subject: Re: [Nek5000-users] proj_ortho error in pipe flow

Hi,

I have made some more tests to find the cause of this issue.
Attached is a laminar pipe flow simulation with temperature and one
passive scalar. The original setup is from Paul's "nusselt.tgz", where
he answered another question of mine.

I found that I can reproduce the same error I see in my turbulent
simulations when there is an additional passive scalar with a different
type of boundary condition, i.e. Neumann boundary condition at the pipe
wall for temperature and Dirichlet boundary condition for the passive
scalar.

Can someone please check my setup and tell me why this causes the
projection to fail?


I hope to save some computational resources by using the projection.
However, if it is not possible, I will keep param(94) = 0.

Kind Regards,
Steffen Straub

--
Karlsruhe Institute of Technology (KIT)
Institute of Fluid Mechanics

M.Sc. Steffen Straub
Doctoral Researcher

Kaiserstra?e 10
Building 10.23
76131 Karlsruhe, Germany

Phone: +49 721 608-43027
E-mail: steffen.straub at kit.edu
Web: http://www.istm.kit.edu

KIT ? The Research University in the Helmholtz Association

Since 2010, the KIT has been certified as a family-friendly university.

-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://lists.mcs.anl.gov/pipermail/nek5000-users/attachments/20170914/3306082d/attachment-0001.html>

------------------------------

Message: 5
Date: Thu, 14 Sep 2017 12:29:10 -0400
From: nek5000-users at lists.mcs.anl.gov
To: nek5000-users at lists.mcs.anl.gov
Subject: [Nek5000-users] problem with applying boundary conditions
Message-ID:
        <mailman.778.1505406553.10190.nek5000-users at lists.mcs.anl.gov>
Content-Type: text/plain; charset="utf-8"

Dear Neks,

I am trying to run a simulation, replicating the Nek5000 run from this
youtube video:
https://www.youtube.com/watch?v=RmmN-KFSvAY

The inlet BC is constant velocity (which is set to 1.0), and as can be seen
from the video, the velocity magnitude reasonably away from the inlet is
zero at starting times. However, in all of my simulations, I see the
velocity magnitude as one at all the interior points away from the inlet
from the beginning (the inlet BC ux=1.0 I am applying is somehow propagated
to all interior elements). Am I doing something wrong somewhere, or is this
the expected behavior? Please see my userbc() and useric() from my .usr
file below:

      subroutine userbc (ix,iy,iz,iside,eg)
      include 'SIZE'
      include 'TOTAL'
      include 'NEKUSE'
      integer ix,iy,iz,eg,e,i,n

CC "My inlet is at x=0"

      if (x.le.1e-4) then
        ux=1.0
      endif

      return
      end

      subroutine useric (ix,iy,iz,eg)
      include 'SIZE'
      include 'TOTAL'
      include 'NEKUSE'

      ux=0
      uy=0

      return
      end

I believe userbc() is executed for only 'v' in my case. For the other
constraints on the elements (like 'W', 'O', 'E'),  I think userbc() would
not run (please correct me if I am wrong, I am learning here). Since the
above functions wouldn't work as expected, I changed userchk() and removed
the above if statements in userbc(). My userchk() is as follows:

      subroutine userchk
      include 'SIZE'
      include 'TOTAL'
      integer i,n

      n = nx1*ny1*nz1*nelv

      if (istep.eq.0) then
        do i=1,n
          if (xm1(i,1,1,1).le.1e-4) then
            vx(i,1,1,1)=1
          endif
        enddo
      endif

      return
      end

This didn't work either. Can somebody point me in the right direction?

Thanks,
Ajay
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://lists.mcs.anl.gov/pipermail/nek5000-users/attachments/20170914/bb155242/attachment.html>

------------------------------

Subject: Digest Footer

_______________________________________________
Nek5000-users mailing list
Nek5000-users at lists.mcs.anl.gov
https://lists.mcs.anl.gov/mailman/listinfo/nek5000-users


------------------------------

End of Nek5000-users Digest, Vol 103, Issue 9
*********************************************
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://lists.mcs.anl.gov/pipermail/nek5000-users/attachments/20170915/322d13e9/attachment.html>

------------------------------

Subject: Digest Footer

_______________________________________________
Nek5000-users mailing list
Nek5000-users at lists.mcs.anl.gov
https://lists.mcs.anl.gov/mailman/listinfo/nek5000-users


------------------------------

End of Nek5000-users Digest, Vol 103, Issue 12
**********************************************
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://lists.mcs.anl.gov/pipermail/nek5000-users/attachments/20170918/3edb650e/attachment-0001.html>


More information about the Nek5000-users mailing list