Fix NASA Polynomial Tests and CConfig Integration by guptapratykshh · Pull Request #2705 · su2code/SU2

guptapratykshh · 2026-01-14T08:30:43Z

Proposed Changes

This PR implements the NASA 7-coefficient polynomial model for calculating specific heat capacity (Cp) and enthalpy in incompressible ideal gas flows. This is a standard format in thermochemistry that allows for temperature-dependent properties with high accuracy across two temperature ranges.

Added:

New Fluid Model: Added CIncIdealGasNASA class implementing the standard NASA polynomial equations for Cp, H, and S.
Robust Inversion: Implemented a Newton-Raphson solver to accurately recover Temperature from Enthalpy (T(h)) with machine precision.
Config Integration: Updated CConfig to parse NASA coefficients (NASA_POLYCOEFFS_LOW, NASA_POLYCOEFFS_HIGH) and temperature ranges (NASA_TEMP_LOW, etc.).
Comprehensive Testing: Added a dedicated unit test suite (UnitTests/.../CIncIdealGasNASA_tests.cpp) verifying mathematical accuracy, smooth range transitions, and thermodynamic consistency.

Related Work

Resolves #2634 ("Cp should use NASA polynomials").

PR Checklist

I am submitting my contribution to the develop branch.
My contribution generates no new compiler warnings.
My contribution is commented and consistent with SU2 style.
I used the pre-commit hook to prevent dirty commits (User to verify).
I have added a test case that demonstrates my contribution.
I have updated appropriate documentation (Pending docs update).

bigfooted · 2026-01-14T10:36:51Z

Note that we already have CP_POLYCOEFFS for polynomial Cp. I think it is best to just re-use this, but use CP_NASA=YES to indicate we want to use the NASA format.
Also note that the nasa polynomials are already in polynomial format, but:

we need to multiply by the universal gas constant R
in the nasa polynomials, Cp is the molar cp and we need mass Cp (J/kg.K).

There is no need to implement the full nasa format here (we can do that with coolprop or cantera), a user can just copy the polynomial coefficients from the database for his temperature range of interest. Just keep it simple here.

pcarruscag · 2026-01-14T15:36:10Z

SU2_CFD/include/fluid/CIncIdealGasNASA.hpp

+ *   Cp/R = a1 + a2*T + a3*T^2 + a4*T^3 + a5*T^4
+ *   H/(R*T) = a1 + a2*T/2 + a3*T^2/3 + a4*T^3/4 + a5*T^4/5 + a6/T
+ *   S/R = a1*ln(T) + a2*T + a3*T^2/2 + a4*T^3/3 + a5*T^4/4 + a7


what is your reference for this? it looks incorrect

equations are correct and follow standard NASA 7-coefficient polynomial format used in CEA (Chemical Equilibrium with Applications) thermodynamic database(around pages 13-14 of the pdf)

link: https://ntrs.nasa.gov/api/citations/20020085330/downloads/20020085330.pdf

I think nasa-7 is fine, nasa-9 also has a T^-1 and T^-2 term for cp but most chemistry databases use nasa-7.

One is a subset of the other, I would say we should just make it handle the full thing. Users can put 0 for the coefficients they don't use.

updated implementation to support full NASA-9 coefficient format, which includes extra T^-1 and T^-2 terms.

…explicit coefficient variables

…ility

pcarruscag · 2026-01-15T18:33:00Z

SU2_CFD/include/fluid/CIncIdealGasNASA.hpp

+    for (int i = 0; i < N_COEFFS; ++i) {
+      if (i < config->GetnPolyCoeffs()) {
+        coeffs_[i] = config->GetCp_PolyCoeff(i);
+      } else {
+        coeffs_[i] = 0.0; 


Looks good.
How is nondimensionalization handled?

since NASA coefficients (a1-a9) are already dimensionless and define Cp/R directly, they don't require standard temperature-power nondimensionalization. the model simply uses these raw coefficients with nondimensional gas constant and temperature. I have verified this aligns with NASA reference documentation and confirmed it works correctly in tests.

What about the temperature we are using to evaluate the polynomial?
Perhaps you can add some regression tests for a simple airfoil, dimensional and nondimensional.
And please update the config_template with the new options.

Can we use this model for the compressible solver too?

pcarruscag

What about the regression tests?

pcarruscag · 2026-01-16T16:06:18Z

SU2_CFD/include/fluid/CIncIdealGasNASA.hpp

+      if (temp_iter < Temperature_Min) {
+        // Clamp to min but don't break immediately to allow recovery if simple overshoot
+         if (abs(delta_temp_iter) < toll) break; 
+         temp_iter = Temperature_Min;
+      }
+      if (temp_iter > Temperature_Max) {
+         if (abs(delta_temp_iter) < toll) break;
+         temp_iter = Temperature_Max;
+      }


Suggested change

if (temp_iter < Temperature_Min) {

// Clamp to min but don't break immediately to allow recovery if simple overshoot

if (abs(delta_temp_iter) < toll) break;

temp_iter = Temperature_Min;

}

if (temp_iter > Temperature_Max) {

if (abs(delta_temp_iter) < toll) break;

temp_iter = Temperature_Max;

}

temp_iter = fmin(fmax(Temperature_Min, temp_iter), Temperature_Max);

pcarruscag · 2026-01-16T16:08:31Z

SU2_CFD/include/fluid/CIncIdealGasNASA.hpp

+    const su2double t_inv2 = t_inv * t_inv;
+
+    // NASA-9: Cp/R = a1*T^-2 + a2*T^-1 + a3 + a4*T + a5*T^2 + a6*T^3 + a7*T^4
+    su2double Cp_over_R = a1*t_inv2 + a2*t_inv + a3 + a4*T_dim + a5*T_dim*T_dim + a6*T_dim*T_dim*T_dim + a7*T_dim*T_dim*T_dim*T_dim;


Use the pow function, this is ugly and not even optimal

…clamping

…stness

Fix NASA Polynomial Tests and CConfig Integration

87b3fdc

guptapratykshh changed the base branch from master to develop January 14, 2026 08:32

Apply pre-commit style fixes

238142d

bigfooted added the changelog:feature label Jan 14, 2026

guptapratykshh added 3 commits January 14, 2026 16:33

Refactor NASA polynomials to use CP_POLYCOEFFS and single CP_NASA flag

56c0f4d

Apply clang-format fixes

448854f

Allow variable number of coeffs for polynomial CP, MU, KT

8ce89c9

pcarruscag reviewed Jan 14, 2026

View reviewed changes

guptapratykshh added 2 commits January 15, 2026 14:19

Add NASA polynomial reference citation and improve code clarity with …

3701aa2

…explicit coefficient variables

Implement full NASA-9 polynomial format with NASA-7 backward compatib…

eca925a

…ility

guptapratykshh force-pushed the feature/nasa-polynomials-cp-2634 branch from 00e0013 to eca925a Compare January 15, 2026 17:00

guptapratykshh requested review from bigfooted and pcarruscag January 15, 2026 17:12

pcarruscag reviewed Jan 15, 2026

View reviewed changes

guptapratykshh force-pushed the feature/nasa-polynomials-cp-2634 branch from 2da80c3 to 1429185 Compare January 16, 2026 05:18

Fix NASA polynomial temperature scaling for non-dimensional simulations

7fbbf18

guptapratykshh force-pushed the feature/nasa-polynomials-cp-2634 branch from 1429185 to 7fbbf18 Compare January 16, 2026 05:21

Merge branch 'develop' into feature/nasa-polynomials-cp-2634

9a458fe

guptapratykshh requested a review from pcarruscag January 16, 2026 10:48

pcarruscag reviewed Jan 16, 2026

View reviewed changes

guptapratykshh added 2 commits January 17, 2026 11:41

Address code review for NASA polynomials logic: use pow and simplify …

0cc637a

…clamping

Code cleanup: check linter suggestions

db284ac

guptapratykshh requested a review from pcarruscag January 17, 2026 06:12

guptapratykshh and others added 3 commits January 17, 2026 11:42

Merge branch 'develop' into feature/nasa-polynomials-cp-2634

a9272db

Complete NASA-9 model: add entropy, improve AD compatibility and robu…

9e203c8

…stness

Merge branch 'develop' into feature/nasa-polynomials-cp-2634

206abcf

Conversation

guptapratykshh commented Jan 14, 2026 • edited Loading Uh oh! There was an error while loading. Please reload this page.

Uh oh!

Proposed Changes

Related Work

PR Checklist

Uh oh!

bigfooted commented Jan 14, 2026

Uh oh!

Choose a reason for hiding this comment

Uh oh!

Choose a reason for hiding this comment

Uh oh!

Choose a reason for hiding this comment

Uh oh!

Choose a reason for hiding this comment

Uh oh!

Choose a reason for hiding this comment

Uh oh!

Choose a reason for hiding this comment

Uh oh!

Choose a reason for hiding this comment

Uh oh!

Choose a reason for hiding this comment

Uh oh!

pcarruscag left a comment

Choose a reason for hiding this comment

Uh oh!

Choose a reason for hiding this comment

Uh oh!

Choose a reason for hiding this comment

Uh oh!

Choose a reason for hiding this comment

Uh oh!

Choose a reason for hiding this comment

Uh oh!

Reviewers

Assignees

Labels

Projects

Milestone

Development

Uh oh!

3 participants

guptapratykshh commented Jan 14, 2026 •

edited

Loading