Bug in HLLC Jacobian calculation

https://github.com/su2code/SU2/blob/60270690cd749bd190e16b4c34cf724c42edbeb1/SU2_CFD/src/numerics/flow/convection/hllc.cpp#L543C3-L543C3

As per the https://www.sciencedirect.com/science/article/pii/S0021999197957934 (original paper), multiplying the Jacobian with \kappa = 0.5 is not right. It is leading to non-convergence.

Further, dpStar_dU[iVar] calculations need to be relooked. As per the original paper, density should be at "j" when it is being used at "i". Though I have not seen a major difference, but currently testing the same more for strong shock cases.

You are right we fixed that for the Roe scheme at some point, if you have time open a PR please.

Sure, will try. Mostly will ask a student, as I am not quite familiar to GitHub working.

…

On Sun, 14 Jan 2024, 00:02 Pedro Gomes, ***@***.***> wrote: You are right we fixed that for the Roe scheme at some point, if you have time open a PR please. — Reply to this email directly, view it on GitHub <#2198 (comment)>, or unsubscribe <https://github.com/notifications/unsubscribe-auth/ADAD2ZP7WA7PL6U764QGRPTYOLHKFAVCNFSM6AAAAABBZKKBO6VHI2DSMVQWIX3LMV43OSLTON2WKQ3PNVWWK3TUHMYTQOJQGY3TKMZXHA> . You are receiving this because you authored the thread.Message ID: ***@***.***>

Compute HLLC on moving grids in the frame of the face, fix its Jacobians

Navigation Menu

Repository navigation

Bug in HLLC Jacobian calculation #2198

Metadata

Assignees

Labels

Type

Projects

Milestone

Relationships

Development

Bug in HLLC Jacobian calculation #2198

Description

Activity

pcarruscag commented on Jan 13, 2024

ashwaniassam commented on Jan 15, 2024

Metadata

Metadata

Assignees

Labels

Type

Projects

Milestone

Relationships

Development

Issue actions