CFD with the error bars left in.
Open-source OpenFOAM simulations, each compared with published experiment. Every result carries its grid-convergence status and its limits, and the disagreements stay on the page.
Cases
Impinging jet heat transfer
Impinging jet: a transition model does not fix the stagnation region
Stagnation Nu/Re^0.7: simulation 0.1384 vs experiment 0.1223, +13.2 %
Compared
Natural convection in a tall cavity
Tall-cavity natural convection: mean Nusselt number 20 % below the quoted experiment, and refining the grid does not close the gap
Mean Nu: simulation 4.658 vs experiment 5.85, −20.4 %
Compared
Windsor squareback drag
Windsor squareback: an extrapolation a fourth grid did not support
Drag coefficient, finest grid: simulation 0.2856 vs experiment 0.3070, −7.0 %
Caveated
Micro-channel heat sink, conjugate heat transfer
Micro-channel heat sink: the model reads 0.95 to 2.09 °C hot at all four thermocouples
TC1 temperature: simulation 41.08 °C vs experiment 40.13 °C, +0.95 °C
Compared
Ahmed body, single and in tandem
Tandem Ahmed bodies: steady RANS follows the drag trend, not the close-gap values
Isolated Cd: simulation 0.2798 vs experiment 0.30, −6.7 %
Caveated
NACA 0012 at 10 degrees, k-omega SST
NACA 0012 at 10 degrees: lift extrapolated from three grids sits +2.14 % above a reference code and +3.4 to +3.8 % above experiment
Lift, grid-extrapolated: simulation 1.10083 vs CFL3D 1.07781, +2.14 %
Caveated
Cylinder at Re 150: laminar vortex shedding
Laminar cylinder wake at Re 150: the Strouhal number is within 0.84 % of a published simulation on all four grids
Strouhal number: simulation 0.1839 vs reference 0.1841, −0.13 %
Caveated
APC 11x5.5E propeller, rotating-frame RANS
A small propeller: tracing a +27.8 % thrust error to the blade pitch, with power still unexplained
Power coefficient: simulation 0.01696 vs UIUC 0.0152, +11.6 %
Caveated
Other work, listed in the ledger
These families have no case page yet. Each links to its row in the ledger, with its numbers and limits.
| Simulation family | Status |
|---|---|
| Ahmed body turbulence-model study (25 / 35 deg)External aero bluff body, SST / realizable k-eps / URANS, ~1M cells, wall functions | Caveated |
| Isolated rotating wheel (Fackrell)Wheel + ground, MRF hub, RANS and URANS, 3.9M to 9.3M cells | Negative |
| Zerihan ground-effect wingMoving belt, h/c sweep, simpleFoam SST, 1.9M cells | Negative |
| Solid mechanics verification (Cook's membrane, cantilever, plate-hole)solidDisplacementFoam / solids4foam, solid only (no fluid-structure coupling run) | Caveated |
| Method of manufactured solutions (laminar Navier-Stokes)Code verification of the solver setup | Verified |
| Sailing hydrofoil sections vs XFOIL / TMR2D simpleFoam SST / SST-LM, GCI ladder 17k to 61k cells | Caveated |
| Free-surface foil (Pernod / Duncan)interFoam VOF 2D, 91k cells | Caveated |
| Cavitation: Kermeen hydrofoil and Clark-YinterPhaseChangeFoam, Schnerr-Sauer | Negative |
| Hull and strut free-surface resistance3D VOF interFoam, DTC and DSYHS hulls | Unfinished |
| motorBike grid-convergence laddersimpleFoam, five grids 167k to 1.71M cells | Negative |
| Curved-flow wing (SRF)SRF simpleFoam, 2D then 3D | Negative |
| 30P-30N multi-element airfoil2D SST, structured grids | Unfinished |
| DLR-F11 high-lift (HLPW-2)3D high-lift configuration | Blocked |
| SAE notchback blind testExternal aero | Unfinished |
| Aeromap surrogate pipelinePython tooling (Gaussian-process surrogate, lap-sim export) | Caveated |
The ledger lists all 24 simulation families, including the ones that failed.
How to read the status
- VerifiedCode verification against an exact (manufactured) solution, with an observed order of accuracy.
- ComparedCompared with experiment, with a numerical uncertainty band. Not claimed as a formal validation.
- CaveatedA result exists, with named limits that change how far it can be trusted.
- NegativeThe model disagreed with experiment, or an earlier agreement did not survive refinement. Kept on purpose.
- UnfinishedRunning, paused, or never run.
- BlockedNo result: the setup failed a mesh or numerics gate.