---------- Forwarded message ---------
From: Beyhan Karakaş beyhannkarakas@gmail.com Date: Sat, Jun 20, 2020 at 7:34 PM Subject: Using finite temperature EOS with GRHydro in BNS simulation To: Einstein Toolkit Users users@einsteintoolkit.org
Dear to whom it may concern,
I am using Turing release on Fedora 31. I have been trying to perform a binary neutron star merger simulation via finite temperature equation of state (might be misleading) with GRHydro. As far as I can see, this question was replied by Roland previously in http://lists.einsteintoolkit.org/pipermail/users/2019-February/006756.html.
Parameters were added as has been suggested but If Meudon_BNS_ID does not refer to source code of Meudon_Bin_NS then I could not find where Meudon_BNS_ID or Meudon_Bin_NS ID is located in Cactus, if it does I will try to work on it. In this setting, is it enough to use IOUtil parameter -filereader_ID_vars = "all" - to allow initial temperature, entropy and electron fraction to be read from the file?
I have tried to use APR eos with SFH0 eos table from stellarcollapse.org ( https://stellarcollapse.org/~evanoc/Hempel_SFHoEOS_rho222_temp180_ye60_versi... ) and got "EOS errori n prim2con_hot" in this setting and do not know how this might be overcomed.
Besides that, I suppose WhiskyTHC allows nuclear eos table to be used with Lorene initial data by providing this table in pizza format and uses some conversion of units but do not know what will be needed for Einstein Toolkit.
Parameter, error and output files are attached.
Best regards,
Beyhan.
Hello all,
I forgot to forward the attachments.
Yours, Roland
---------- Forwarded message ---------
From: Beyhan Karakaş beyhannkarakas@gmail.com Date: Sat, Jun 20, 2020 at 7:34 PM Subject: Using finite temperature EOS with GRHydro in BNS simulation To: Einstein Toolkit Users users@einsteintoolkit.org
Dear to whom it may concern,
I am using Turing release on Fedora 31. I have been trying to perform a binary neutron star merger simulation via finite temperature equation of state (might be misleading) with GRHydro. As far as I can see, this question was replied by Roland previously in http://lists.einsteintoolkit.org/pipermail/users/2019-February/006756.html.
Parameters were added as has been suggested but If Meudon_BNS_ID does not refer to source code of Meudon_Bin_NS then I could not find where Meudon_BNS_ID or Meudon_Bin_NS ID is located in Cactus, if it does I will try to work on it. In this setting, is it enough to use IOUtil parameter -filereader_ID_vars = "all" - to allow initial temperature, entropy and electron fraction to be read from the file?
I have tried to use APR eos with SFH0 eos table from stellarcollapse.org ( https://stellarcollapse.org/~evanoc/Hempel_SFHoEOS_rho222_temp180_ye60_versi... ) and got "EOS errori n prim2con_hot" in this setting and do not know how this might be overcomed.
Besides that, I suppose WhiskyTHC allows nuclear eos table to be used with Lorene initial data by providing this table in pizza format and uses some conversion of units but do not know what will be needed for Einstein Toolkit.
Parameter, error and output files are attached.
Best regards,
Beyhan.
Hello Beyhan,
I am using Turing release on Fedora 31. I have been trying to perform a binary neutron star merger simulation via finite temperature equation of state (might be misleading) with GRHydro. As far as I can see, this question was replied by Roland previously in http://lists.einsteintoolkit.org/pipermail/users/2019-February/006756.html.
Parameters were added as has been suggested but If Meudon_BNS_ID does not refer to source code of Meudon_Bin_NS then I could not find where Meudon_BNS_ID or Meudon_Bin_NS ID is located in Cactus, if it does I will try to work on it.
Meudon_BNS_ID refers to the binary neutron star ID reader for LORENE that is found in the toolkit. Indeed the correct name is Meudon_Bin_NS .
In this setting, is it enough to use IOUtil parameter -filereader_ID_vars = "all" - to allow initial temperature, entropy and electron fraction to be read from the file?
Not quite. filereader reads in data from Cactus 3d output files. However that is not what is done with the Meudon thorns. Instead they are regular initial data thorns that read in data from LORENE's output files.
Looking at the file Meudon_Bin_NS/src/Bin_NS.cc then it will set HydroBase::temperature and HydroBase::Y_e if those variables have storage eg if you add
HydroBase::initial_temperature = "read from file" HydroBase::initial_Y_e = "read from file"
Really though the Meudon thorn should extend the valid ranges of those keywords to include "Meudon_Bin_NS" the way tit does for the other initial_XXX parameters.
I noticed that there is a TOVSolverHot thorn in http://svn.einsteintoolkit.org/incoming/TOVSolverHot/ that lets you construct initial data for an isothermal TOV star and that may be a an easier test before going for a full binary simulation.
I have tried to use APR eos with SFH0 eos table from stellarcollapse.org ( https://stellarcollapse.org/~evanoc/Hempel_SFHoEOS_rho222_temp180_ye60_versi... ) and got "EOS errori n prim2con_hot" in this setting and do not know how this might be overcomed.
There is not usually a really good general solution for this. It boils down to gut feeling and experience with this particular type of simulations. I would give the isothermal single TOV a try first and once that works work my way up to a binary.
Besides that, I suppose WhiskyTHC allows nuclear eos table to be used with Lorene initial data by providing this table in pizza format and uses some conversion of units but do not know what will be needed for Einstein Toolkit.
WhiskyTHC has a full example on how to use it in its tarball http://personal.psu.edu/dur566/downloads/WhiskyTHC.tar. Its output is in the regular ET HydroBase variables so not really different from when using GRHydro.
Yours, Roland
Dear Roland,
Firstly, thank you for your considerations and detailed response. Also, thank you for forwarding the message. I will first try to use TOVSolverHot and make it work as you have suggested.
Best regards,
Beyhan.
On Sun, Jun 21, 2020 at 9:56 PM Roland Haas rhaas@illinois.edu wrote:
Hello Beyhan,
I am using Turing release on Fedora 31. I have been trying to perform a binary neutron star merger simulation via finite temperature equation of state (might be misleading) with GRHydro. As far as I can see, this question was replied by Roland previously in
http://lists.einsteintoolkit.org/pipermail/users/2019-February/006756.html .
Parameters were added as has been suggested but If Meudon_BNS_ID does not refer to source code of Meudon_Bin_NS then I could not find where Meudon_BNS_ID or Meudon_Bin_NS ID is located in Cactus, if it does I will try to work on it.
Meudon_BNS_ID refers to the binary neutron star ID reader for LORENE that is found in the toolkit. Indeed the correct name is Meudon_Bin_NS .
In this setting, is it enough to use IOUtil parameter -filereader_ID_vars = "all" - to allow initial temperature, entropy and electron fraction to be read from the file?
Not quite. filereader reads in data from Cactus 3d output files. However that is not what is done with the Meudon thorns. Instead they are regular initial data thorns that read in data from LORENE's output files.
Looking at the file Meudon_Bin_NS/src/Bin_NS.cc then it will set HydroBase::temperature and HydroBase::Y_e if those variables have storage eg if you add
HydroBase::initial_temperature = "read from file" HydroBase::initial_Y_e = "read from file"
Really though the Meudon thorn should extend the valid ranges of those keywords to include "Meudon_Bin_NS" the way tit does for the other initial_XXX parameters.
I noticed that there is a TOVSolverHot thorn in http://svn.einsteintoolkit.org/incoming/TOVSolverHot/ that lets you construct initial data for an isothermal TOV star and that may be a an easier test before going for a full binary simulation.
I have tried to use APR eos with SFH0 eos table from stellarcollapse.org (
https://stellarcollapse.org/~evanoc/Hempel_SFHoEOS_rho222_temp180_ye60_versi...
) and got "EOS errori n prim2con_hot" in this setting and do not know how this might be overcomed.
There is not usually a really good general solution for this. It boils down to gut feeling and experience with this particular type of simulations. I would give the isothermal single TOV a try first and once that works work my way up to a binary.
Besides that, I suppose WhiskyTHC allows nuclear eos table to be used with Lorene initial data by providing this table in pizza format and uses some conversion of units but do not know what will be needed for Einstein Toolkit.
WhiskyTHC has a full example on how to use it in its tarball http://personal.psu.edu/dur566/downloads/WhiskyTHC.tar. Its output is in the regular ET HydroBase variables so not really different from when using GRHydro.
Yours, Roland
-- My email is as private as my paper mail. I therefore support encrypting and signing email messages. Get my PGP key from http://pgp.mit.edu .
users@lists.einsteintoolkit.org
-
Beyhan Karakaş -
rhaas＠illinois.edu -
Roland Haas