PFLOTRAN in dfnWorks: Newton solver failed to converge in steady-solve, reason: -3

Hai Pham

unread,

Nov 16, 2017, 2:26:46 PM11/16/17

to pflotran-users

Dear all,

I got a lot of errors "Newton solver failed to converge in steady-solve, reason:-3" (more logs in attached files) when running PFLOTRAN in within dfnWorks 2.0. I guess one of the reasons is I extract too much water from a model cell with low permeability. What should be the best solver/option to run PFLOTRAN with DFN? How can I print convergence profile for every iteration? I am using the option below but it failed to converged very often.

Thank you so much,

Hai Pham

#================= flow solver ===============================================

NEWTON_SOLVER FLOW

# MATRIX_TYPE AIJ

# PRECONDITIONER_MATRIX_TYPE AIJ

RTOL 1.d-08

ATOL 1.d-08

STOL 1.d-30

MAXIT 10000

MAXF 50000

ITOL_UPDATE 1.d0

# NO_INFINITY_NORM

PRINT_DETAILED_CONVERGENCE

/

failed_logs_printed_from_screen

logs_printed_from_PFLOTRAN.out

Hammond, Glenn E

unread,

Nov 20, 2017, 10:14:38 AM11/20/17

to pflotran-users, dfnw...@lanl.gov

Hai,

Los Alamos will need to help you with the below. I have CC'ed their mailing list. Hopefully, they will reply soon.

Glenn

From: pflotra...@googlegroups.com <pflotra...@googlegroups.com> on behalf of Hai Pham <pvh...@gmail.com>
Sent: Thursday, November 16, 2017 11:26 AM
To: pflotran-users
Subject: [EXTERNAL] [pflotran-users: 4635] PFLOTRAN in dfnWorks: Newton solver failed to converge in steady-solve, reason: -3

Karra, Satish

unread,

Nov 20, 2017, 12:02:11 PM11/20/17

to Hammond, Glenn E, pflotran-users, dfnw...@lanl.gov

Send your input deck.

Hai Pham

unread,

Nov 29, 2017, 7:33:49 PM11/29/17

to pflotra...@googlegroups.com, Hammond, Glenn E, dfnw...@lanl.gov

Thank Glenn and Satish. Attached please find the PFLOTRAN input file. Sorry I did not see your reply until today.

Hai.

On Mon, Nov 20, 2017 at 9:02 AM, Karra, Satish <satk...@lanl.gov> wrote:

Send your input deck.

From: <owner-d...@maillist.lanl.gov> on behalf of "Hammond, Glenn E" <geh...@sandia.gov>
Date: Monday, November 20, 2017 at 8:14 AM
To: pflotran-users <pflotran-users@googlegroups.com>
Cc: "dfnw...@lanl.gov" <dfnw...@lanl.gov>
Subject: Re: [EXTERNAL] [pflotran-users: 4635] PFLOTRAN in dfnWorks: Newton solver failed to converge in steady-solve, reason: -3

Hai,

Los Alamos will need to help you with the below. I have CC'ed their mailing list. Hopefully, they will reply soon.

Glenn

iflow_base.txt

Karra, Satish

unread,

Nov 30, 2017, 1:39:58 PM11/30/17

to Hai Pham, pflotra...@googlegroups.com, Hammond, Glenn E, dfnw...@lanl.gov

I can only look at this in January. I’ll be on travel and vacation most of Dec.

—Satish

From: <owner-d...@maillist.lanl.gov> on behalf of Hai Pham <pvh...@gmail.com>
Date: Wednesday, November 29, 2017 at 5:33 PM
To: "pflotra...@googlegroups.com" <pflotra...@googlegroups.com>
Cc: "Hammond, Glenn E" <geh...@sandia.gov>, "dfnw...@lanl.gov" <dfnw...@lanl.gov>
Subject: Re: [EXTERNAL] [pflotran-users: 4641] PFLOTRAN in dfnWorks: Newton solver failed to converge in steady-solve, reason: -3

Thank Glenn and Satish. Attached please find the PFLOTRAN input file. Sorry I did not see your reply until today.

Hai.

On Mon, Nov 20, 2017 at 9:02 AM, Karra, Satish <satk...@lanl.gov> wrote:

Send your input deck.

From: <owner-d...@maillist.lanl.gov> on behalf of "Hammond, Glenn E" <geh...@sandia.gov>
Date: Monday, November 20, 2017 at 8:14 AM
To: pflotran-users <pflotra...@googlegroups.com>
Cc: "dfnw...@lanl.gov" <dfnw...@lanl.gov>
Subject: Re: [EXTERNAL] [pflotran-users: 4635] PFLOTRAN in dfnWorks: Newton solver failed to converge in steady-solve, reason: -3

Hai,

Los Alamos will need to help you with the below. I have CC'ed their mailing list. Hopefully, they will reply soon.

Glenn

Reply all

Reply to author

Forward