TOUGHREACT_Debugging
Hello Everyone:
I am pretty new to TOUGHREACT. I am trying to build a model to simulate Ba++ and So4-- reaction in porous media. The attachments are the input files I modified based on P8_EOS1 from samples.
However, when I try to run them using EOS1, the running results are shown in the picture. There is no information about the error. Could anyone please help me with this or tell me where I should look for. Thanks very much!
Have a great day!
Regards,
4 replies
-
I am pretty sure something wrong with flow.inp formatting. Please try entering 2 empty lines before keyword RPCAP.
-
Kenny
Thank you very much! Appreciate your help.
As you see, I have no idea how to debug based on the CMD running log. It is so amazing that you know there is something wrong with RPCAP. Could you please tell me which file contains the error info that is helpful for debugging? Or which file points out what specific keyword is wrong?
I just followed your instructions and reran the file. There is no info about the error as shown in the picture. I wonder which file I should check to find out the wrong keyword? Thank you!
Have a great day!
Regards,
-
-
Hi Shijun,
Everything should be done in a stepwise approach, first flow, then injection, chemistry, perm change. etc., until you are sure the files and the results are correct. Look at the .out files to figure out where there was a problem in the inputs. Even if there is no error message, the last thing written in the .out file will be the last parameter read in successfully, so you can tell where the mistake was made. I made a lot of fixes to all your files to get it to run. It starts, but runs only for a little while, because temperatures go to zero. I'm not sure what you are trying to model, so that's as much as I can do. Some comments (look at my changes in the files): Do not put the ELEME, CONNE, and GENER blocks in the flow.inp file. TOUGH reads these and writes them out without the zero's. Then it reads them from the files and does not recognize them. The MULTI was not correct for EOS1. No enthalpies were given for the injection, and there should only be one component. I put in enthalpies for 60C just to get it to run, because you had 60C everywhere else. The initial time step should be always be around 1 second. Concentrations always have to be greater than zero. The initial conditions under the PARAM block were not correct for EOS1. The solute.inp was not correct for the current version. Again, always make sure the flow runs without chemistry, and you get the correct temperatures and pressures, etc. Then incrementally add GENER and then chemistry, etc. Always look at the .out files to see where things are crashing, because of incorrect inputs.
cheers,
Eric
-
Eric Sonnenthal
Thank you so much for your detailed reply. That is so helpful for me to learn the TOUGHREACT.
Here is the story: I plan to use the TOUGHREACT to simulate a geochemistry reaction, Ba+2 + SO4-2 = BaSO4, in a core as shown in fig_1. The core is saturated with Ba+2 before injection. Two injectors are drilled into the core: one injects Ba+2, and the other injects SO4-2. When Ba+2 reacts with SO4-2 and forms BaSO4 solid, the solid deposition will block the pore and induce permeability decrease, which causes core pressure increase. My purpose is to match the pressure increase.
Several model settings: The outlet of the core is exposed to the atmosphere (P_out=14.7psi). Therefore, I make the volume of every grid block at the outlet edge greater than 1E50 m^3 and pressure at 14.7psi. Because the experiment is conducted at constant temperature (60 C), I set the formation heat conductivity very low, and the rock gain specific heat very high to remain constant temperature.
Thanks to your reply, I realize EOS7 is better than EOS1 for doing the simulation. So, I go through the manual and redo the input files. The input files are in the attachments. I still have several questions regarding of initial condition and boundary water:
- I use the 'Default indices for all chemical property zones' method to define the initial liquid composition in the core. Because there are two injectors with different compositions, I set the IZBWDF(injection water zone number) at 0. Then, I use IZBW for two specific blocks to define the injector compositions (as shown fig_2). I wonder whether this way will work or not?
- In the flow.out file, I find the 'Phase Volume in Place' is always zero. How do I set the core is saturated with liquid (Sw=1)?
Hope to get help from you! Thanks
Have a great day!
Regards,
-
