Liquid dispensing · Internal fluid flow
Dispensing nozzle
A pressure-drop study for three small dispensing nozzles. It compares bore geometry, retained volume, flow rate, and viscosity using a low-Reynolds-number flow model.
CFD and synthetic bench completed

Implementation
The solver assembles the full axisymmetric Stokes strain terms with P2 velocity and P1 pressure on a body-fitted mesh.
Three nominal bore geometries are exported as STEP and STL. Raw nodal fields, pressure comparisons and input dimensions are included.
The virtual bench executes tolerance-corner FEM checks, pressure-envelope calculations, synthetic instrument fits, obstruction detection and a compliant dosing/shutoff model. The final operating decision is supplied as JSON.
Inputs and assumptions
- Isothermal Newtonian reference liquid: density 1000 kg/m3; viscosity 0.5 to 1.0 Pa s.
- Flow 10 to 30 mm3/s; 1 mm outlet diameter; 20 mm passage; 6500 Pa pump budget and 60 mm3 retained-volume limit.
- Nozzle walls are no-slip. Developed profiles bound the modeled passage; external fittings and supply lines are excluded.
- Combined system envelope: command at most 26.4 mm3/s at 0.5 to 1.0 Pa s with +/-0.01 mm bore radii, +/-0.1 mm outlet-land length, the declared 10 mm feed tube and bounded static meniscus. The original 30 mm3/s endpoint is comparison evidence, not an accepted combined-system operating point.
Results
| Selected geometry | Wide inlet with 14 mm taper and 2 mm outlet land |
|---|---|
| Nominal pressure drop | 3071.31 Pa |
| Pressure reduction from baseline | 42.37 % |
| Nozzle-only pressure at original 30 mm3/s endpoint | 6142.62 Pa |
| Retained liquid volume | 56.941 mm3 |
| Maximum outlet Reynolds number | 0.0764 |
| Virtual-bench flow bound | 26.4 mm3/s |
Verification
21 executed numerical/geometry checks pass, including six regression tests, pipe benchmark, three mesh levels and extended boundaries. Additional virtual bench: 14 verification checks; 243 tolerance/operating cases; 30 synthetic observations; four dosing cases. Original maximum-flow condition fails added system effects; computed bound 26.4 mm3/s.
python run.pyDownload the verification record ↓Additional scenarios
14 outcomes checked through synthetic/theoretical. The report lists the expected and observed outcomes for each check.
Read the validation report ↗Project files
The ZIP contains the source, inputs, results, and instructions. These files are also available individually.
- Axisymmetric FEM solverflow.py · 5 KB
- Geometry and operating inputsinputs.json · 1 KB
- Selected nozzle STEPoutputs/cad/wide-inlet.step · 18 KB
- Selected fluid volume STEPoutputs/cad/wide-inlet-fluid.step · 10 KB
- Dimensioned section viewoutputs/section.svg · 43 KB
- Flow and viscosity sweepoutputs/operating-sweep.csv · 2 KB
- Mesh convergence resultsoutputs/mesh-study.csv · 2 KB
- Physical pressure-test procedurebench-test.md · 2 KB
- Selected operating envelopeoutputs/virtual-bench/selected-operating-envelope.json · 1 KB
- Completed virtual-bench verificationcompletion.json · 5 KB
- Virtual pressure and dosing bench sourcevirtual_bench.py · 23 KB
- Tolerance, instrument and dosing inputsvirtual-bench-inputs.json · 1 KB
- Virtual bench and acceptance reportoutputs/virtual-bench/report.html · 11 KB
- Acceptance outcomesoutputs/virtual-bench/acceptance-matrix.csv · 1 KB
- All 243 pressure-envelope casesoutputs/virtual-bench/tolerance-pressure-envelope.csv · 27 KB
Archive checksum
SHA-256
03355be45a6fb4686f7f489f525db253159ed0d828d7cd25b61cd92f978ab294Original project scope
Archived brief PDF ↗Run locally
Python 3.12 / CadQuery 2.7.0 / scikit-fem 12.0.2 / NumPy 2.3.5 / SciPy 1.17.0 / Matplotlib 3.10.8
Extract the ZIP and open a terminal in the dispensing-nozzle folder. The README includes setup and usage details.
python -m pip install -r requirements.txt
python run.py
python test_flow.pyLimits
- Numerical reference study; no physical dispensing measurements were used.
- Stokes flow excludes inertia. The declared range stays below Re 0.1, but a different fluid or flow rate needs a new regime check.
- CAD uses nominal geometry. Materials, seals, threads, surface finish, tolerances and cleaning validation require a separate manufacturing release.
- Additional static meniscus and lumped compliance models cover pressure-head and shutoff sensitivities. They do not resolve droplet breakup, moving bubbles or physical contact angle.


