[FLASH-USERS] multiTemp/Gamma EOS implementation in LaserSlab

Wed Feb 12 14:07:52 EST 2020

On Wed, Feb 12, 2020 at 12:44 PM Acharya, Nitish <nachary2 at ur.rochester.edu>
wrote:

> Hey Marissa,
>
> These are great suggestions 🙂Thanks so much.
>
> 1) I did modify the files in the source but I also backed up the original
> files. From now onwards, I'll copy the files to the SimulationMain/.. and
> edit there 🙂. But in the end I need to copy the modified files from the
> SimulationMain/.. back to the source, right?
>

Keep the source pristine in the sense that you haven't touched anything
there. If you keep the files you copied from the source, and edit them in
*SimulationMain/*, when you configure and make in an object directory, *it
should* copy those files into the object directory and not the source files
(meanwhile keeping them pristine). You can then edit those files in the
object directory as you debug, and they won't change the source, just the
ones in the *SimulationMain/*. Just if you change your *Config* file, for
instance, you'll need to resetup rather than just make as you could then be
using different modules.

> 2) Even though I modified few number of lines in the EOS modules, I didn't
> introduce any variables. I just modified the expression for pressure and
> internal energies. So I guess I can skip putting whatever I calculated into
> the UNK variables. Am I right here?
>

I suppose? I guess I am not sure and it's hard to say what is truly
happening over e-mail. Have you tried implementing print statements yet
where you made changes?

> Nitish
> ------------------------------
> *From:* Marissa Adams <marissa at flash.uchicago.edu>
> *Sent:* Wednesday, February 12, 2020 11:56 AM
> *To:* Acharya, Nitish <nachary2 at ur.rochester.edu>
> *Cc:* flash-users at flash.uchicago.edu <flash-users at flash.uchicago.edu>
> *Subject:* Re: [FLASH-USERS] multiTemp/Gamma EOS implementation in
> LaserSlab
>
>
>
> On Tue, Feb 11, 2020 at 2:35 PM Acharya, Nitish <nachary2 at ur.rochester.edu>
> wrote:
>
> Dear flash-users,
>
> I am trying to implement multiTemp/Gamma EOS to one of the materials in
> the LaserSlab and just use the tabulated EOS for the rest. There was no
> error during code compilation. However, I tried to modify couple lines from
> the following EOS routines:
>
> i) eos_idealGamma.F90
> ii) eos_idealGamma3T.F90
>
> These routines were located in .../Eos/EosMain/MultiTemp/Gamma/Ye
>
> I modified the expression for pressure and comibinedGamc calculation.
> However, I dont see any changes while I plot the gamc and pres variable in
> VISIT.
>
>
>
> Are you modifying these files in the *source/physics/blah/blah *(from
> here on I'll refer to it just as *source*, even though the *Simulation*
> module is also apart of the source but that is the part users should really
> just mess with I suppose) code directly, or in the *object* directory in
> which you have made the executable?
>
> I would recommend not touching the source so that you can always maintain
> a copy of what you downloaded, but always making new *Simulation/SimulationMain/
> *directories. In that *Simulation/SimulationMain/* directory copy the
> files from the source that you want to edit. Then when you configure and
> make your flash4 you can make sure that the files you've copied and
> symbolically linked to your *Simulation/SimulationMain/* directory rather
> than the source. From there edit your files and debug until it makes, then
> ./flash4, rinse and repeat until you feel as if you have debugged your
> simulation thoroughly and it provides results you may anticipate.
>
> I think that may alleviate a lot of problems like this because the first
> question that pops into my head is whether or not the *object* directory
> is actually seeing that file you have edited, especially if you're claiming
> results are unchanged.
>
> **** You can also do standard print statement wizardry to see if the code
> it getting to where you want it to be at execution.
>
> **** Also are you putting whatever you've calculated into the UNK
> variables? e.g. U(PRES_VAR,i,j,k) = Nitish_Pressure make sure you put
> those in #IFDEF statements so the compiler is kept in the loop (haha) with
> what you're trying to do.
>
> Someone else who wrote those routines and knows when they are called etc.
> might be better to assist but this is just the quick take that I can put
> together in the 20 mins it took me to write this.
>
> I would recommend the following before answering anything else:
>
> 0. Make sure you're following the recipe I described in the second
> paragraph above, where one makes sure to keep the original source and you
> know exactly what you are manipulating and changing. Then you can diff
> the files you have edited with those from the source. I am just trying to
> encourage appropriate version control, which I know I also fail at keeping
> up with sometimes, but that always seems to be the crux to many of the
> problems I see folks having if they can't nail down a specific question.
>
> 1. Once you have configured and made your executable in such an *object*
> directory, add print statements in the routines you have edited. When you
> run the code do you see them in the standard out?
>
>
> 1) For my case of implementation, how is pressure initialized and
> calculated later in the simulation? It seems I'm modifying the lines in
> wrong EOS routines.
>
> 2) In the params file, I set gammaEle = 2.4, gammaIon = 2.4 and gammaRad =
> 1.33. When I plot gamc in ViSit, it shows 2.348. How is gamc initialized
> and calculated later in the simulation?
>
> I have attached my Config files and .f90 files.
>
> Thanks so much,
>
> Nitish
>
> ---------------------------
>
> Nitish Acharya
>
> PhD Student
>
> Department of Mechanical Engineering
>
> University of Rochester
>
>
>
> Best,
> Marissa
>
> --
> %---------------------------------------------%
> Marissa B. P. Adams
> (she/her/hers)
> Phd Candidate, University of Rochester
> NDVS, University of Chicago
> *E-mail:* marissa at flash.uchicago.edu
> *Website:* https://www.pas.rochester.edu/~madams
> <http://www.pas.rochester.edu/~madams>
> *Twitter:* https://twitter.com/mbpadams
> <https://urldefense.proofpoint.com/v2/url?u=https-3A__twitter.com_mbpadams&d=DwMFaQ&c=kbmfwr1Yojg42sGEpaQh5ofMHBeTl9EI2eaqQZhHbOU&r=cuPKv3tS5I2VBKSHmPxoTpPzBE9dgd_ifIonFFBHfHw&m=V7kLoTJtbt6I_38TyrlMExioenxVwEL01QUjA09vCtg&s=uEJkrN1LQOjPblchPQLXUg93d3T1IFwkslQ_SypGlys&e=>
> *Current Location:*
> 553 ERC
> Flash Center for Computational Science
> 5640 S. Ellis Ave, ERC 553
> Chicago, IL 60637
>

-- 
%---------------------------------------------%
Marissa B. P. Adams
(she/her/hers)
Phd Candidate, University of Rochester
NDVS, University of Chicago
*E-mail:* marissa at flash.uchicago.edu
*Website:* https://www.pas.rochester.edu/~madams
<http://www.pas.rochester.edu/~madams>
*Twitter:* https://twitter.com/mbpadams
*Current Location:*
553 ERC
Flash Center for Computational Science
5640 S. Ellis Ave, ERC 553
Chicago, IL 60637
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://flash.rochester.edu/pipermail/flash-users/attachments/20200212/0e65bda0/attachment.htm>