Every empirical correlation in WaveBench names its source and its validity range in the XML doc comment where it is implemented — that is a non-negotiable of the build contract, not a convention. This file gathers them in one place so a reader can check the provenance of a result without reading the source, and so a generated report's citation list has something to point at.
If a correlation is used outside its stated validity range, the output is marked. WaveBench does not silently extrapolate.
The wider bibliography the design drew on is in
WaveBench-Master-Plan.md, Part 13. What follows
is narrower and more useful: what the code actually computes with.
| Used for | Source | Validity |
|---|---|---|
| Shock-tube verification of the Riemann solver | Sod, G. A. J. Comput. Phys. 27 (1978) 1–31 | Exact solution; the anchor, not a correlation |
| MUSCL-Hancock + HLLC construction | Toro, E. F. Riemann Solvers and Numerical Methods for Fluid Dynamics, 3rd ed., Springer 2009 | — |
| Well-balanced treatment of area change | Plan §2.2; verified against a stationary taper | Spurious velocity < 1e-10 m/s |
Measured behaviour of the scheme, including its −3 dB bandwidth at a given
mesh, is in numerics.md.
| Used for | Source | Validity |
|---|---|---|
| Darcy friction factor | Haaland, S. E. J. Fluids Eng. 105 (1983) 89–90 | Turbulent, Re > 4000; explicit approximation to Colebrook within 2% |
| Wall heat transfer in ducts | Colburn analogy, Nu = 0.023·Re^0.8·Pr^(1/3) | Fully developed turbulent pipe flow |
| Wall thermal node, solved between cycles | Plan §2.3 | A steel wall's time constant is ~10 s against a ~20 ms cycle |
Three correlations ship; the model states which was used.
| Correlation | Source | Validity |
|---|---|---|
| Woschni | SAE 670931 (1967) | SI/CI engines near the conditions of Woschni's diesel data |
| Hohenberg | SAE 790825 (1979) | A later fit, generally better at part load |
| Annand | Annand, W. J. D., Proc. IMechE 177 (1963) | Nu = a·Re^0.7 with a ≈ 0.35–0.8 |
They disagree by a few per cent of heat loss, which is worth seeing rather than hiding: the choice is exposed as a field.
| Used for | Source | Validity |
|---|---|---|
| Burn-rate law | Wiebe function, a = 5 convention (99.3% at the stated duration) | A shape, fitted per engine, not a prediction |
| Laminar flame speed | Metghalchi, M. & Keck, J. C. Combustion and Flame 48 (1982) 191–210 | Fuel-specific coefficients; the stated pressure/temperature range per fuel |
| Knock induction time | Douaud, A. M. & Eyzat, P., SAE 780080 (1978) | Native units are atm; a typed overload converts |
| Knock onset integral | Livengood, J. C. & Wu, P. C., 5th Symposium on Combustion (1955) 347–356 | ∫dt/τ = 1 at onset. The RANKING between fuels is verified; the absolute value is not |
| Used for | Source | Validity |
|---|---|---|
| Compressible orifice flow | Heywood, J. B. Internal Combustion Engine Fundamentals, §6.3 | Through to choking |
| Generic valve C_d against lift | Plan §2.5, generic curve | The dominant error source. Replaced by measured flow-bench data when supplied |
| FSAE restrictor | Plan §4.6.4 | Chokes at theory; verified |
| Used for | Source | Validity |
|---|---|---|
| Three-leg junction loss | Idelchik, I. E. Handbook of Hydraulic Resistance — converging and diverging wye forms | Steady; branch-angle dependence carried by these forms |
| Junction as a shared boundary Riemann problem | Plan §2.7 | Verified to 0.07% under a pulse of 69% of mean pressure |
| Collector configurations and pulse timing | Blair, G. P. Design and Simulation of Four-Stroke Engines, SAE 1999, ch. 6 | — |
Bassett, Winterbone & Pearson (2001) UNSTEADY junction coefficients are a
documented standing deferral: branch-angle dependence is currently carried by
the Idelchik forms. See physics.md.
| Used for | Source | Validity |
|---|---|---|
| Open-end radiation impedance | Levine, H. & Schwinger, J. Physical Review 73 (1948) 383–406 | Low-ka form used below the stated limit |
| Transfer-matrix method | Munjal, M. L. Acoustics of Ducts and Mufflers, 2nd ed., Wiley 2014 | Plane-wave regime, below the first cross-mode |
| Atmospheric absorption | ISO 9613-1 | — |
| Frequency weighting and time constants | IEC 61672-1 | — |
| Fractional-octave bands | IEC 61260-1 | — |
| Loudness normalisation for A/B | ITU-R BS.1770 / EBU R128 | — |
| Stationary vehicle noise geometry | ISO 5130, SAE J1287 | Referenced for geometry; no absolute level is claimed |
The TMM is cross-validated against the nonlinear solver to 0.45 dB. Four
psychoacoustic metrics — ISO 532-3, ECMA-418-2, fluctuation strength and
DIN 45681 — are not implemented: each needs verification against published
reference signals that cannot be redistributed. See
acoustics.md.
| Used for | Source | Validity |
|---|---|---|
| Corrected mass flow and speed | SAE J1826 (2022), Turbocharger Gas Stand Test Code | Defines surge, soft-surge and choke lines, and the reference conditions a map must state |
| Pulse vs constant-pressure turbocharging | Watson, N. & Janota, M. S. Turbocharging the Internal Combustion Engine, Macmillan 1982 | — |
| Blade speed ratio, turbine efficiency character | Baines, N. C. Fundamentals of Turbocharging, Concepts NREC 2005, ch. 5 | Radial turbines; peak near U/C ≈ 0.7 |
| Housing rescaling by area ratio | Watson & Janota 1982; Baines 2005 | Corrected flow scales with A/R; the efficiency field is left unchanged |
| Surge dynamics | Greitzer, E. M. ASME J. Eng. Power 98 (1976); Moore, F. K. & Greitzer, E. M. (1986) | B-parameter form; scaling below critical B |
A map's reference conditions are required and never defaulted. The two common gas-stand references are 1.69% apart in corrected speed before that propagates into pressure ratio, and the error is invisible in the answer.
Every turbocharger in the shipped library is analytic, with its own
Source and Licence saying so. WaveBench ships no manufacturer maps without
written permission (plan §4.7).
| Used for | Source |
|---|---|
| Torque and power correction | SAE J1349 |
| Standard atmosphere | U.S. Standard Atmosphere (1976) |
| Physical constants | CODATA 2018 |
| Atomic weights | IUPAC 2021 |
| Quantities and units | ISO 80000-4 |
| Rotor balance grades | ISO 1940 |
Every design warning names the source of its limit, because a limit without one is this tool's opinion.
| Limit | Source |
|---|---|
| Diffuser half-angle ≤ 7° before separation | Claywell, M. & Horkheimer, D., SAE 2006-01-3654 |
| Merge collector branch angle 10–30° | Idelchik, converging wye |
| Collector area ratio 0.7–1.0 × combined primaries | Blair 1999, ch. 6 |
| Surge margin ≥ 10% | Plan §4.2 |
| Turbine inlet temperature against the rated limit | Plan §4.3 |
| L/D above about 1 for plane-wave theory | Plan §5.3 |
What has been checked against something outside this tool, and what has not,
is stated in every generated report and in physics.md. The
standing open cases are:
- Transient spool against a measured case. No redistributable dataset found; self-consistency (mesh convergence, energy balance, sensitivity-band behaviour) is checked in its place.
- Measured order levels. Order structure is verified; absolute measured levels are not.
- Diabatic correction against a measured on-engine outlet temperature.
The first published-data case is the open-access CSU thesis runner-length
study, reproduced in validation/.