# Pressure and flow bench test

The physical procedure below has not been performed. Its computational run-through is complete: see `outputs/virtual-bench/report.html`, `completion.json`, and `virtual-bench-inputs.json`. Those results are synthetic/theoretical and identify adverse cases as well as passing conditions.

1. Confirm the actual liquid is Newtonian over the relevant shear range. Measure density and dynamic viscosity at a controlled temperature. Use the measured values in `inputs.json` before comparing.
2. Measure the bore and outlet land, inspect for burrs, and document surface finish. Add upstream and downstream pressure-tap sections whose geometry is included in the model. The current model has no fittings; do not attribute their pressure loss to the nozzle.
3. Install calibrated pressure transducers across the modeled passage. Record zero offsets, range, uncertainty, temperature and tubing arrangement. Fully fill and de-air the system.
4. For each candidate, set 10, 20 and 30 mm3/s. Verify flow independently by collecting and weighing discharge over a measured interval, converting mass to volume using measured density. Repeat each operating point at least three times after stabilization.
5. Compare mean static pressure difference with the corresponding computed point. Propagate pressure, flow, viscosity and geometry uncertainties. Investigate disagreement beyond the combined uncertainty rather than fitting viscosity to force agreement.
6. At the upper viscosity, check whether the pressure remains below the 6500 Pa reference budget. Recalculate the Reynolds number for every real case. Record leaks, air, drift and pulsation separately.
7. Evaluate shutoff, drips, cleaning and delivery repeatability in separate tests. They are not predicted by this steady single-phase model.