Notes · the working surface
Notes from a working surface.
- active
- 8
- last touched
- Apr 28, 2026
FOCUS Numerics from Zingg's two books, an espresso build on the bench, a Z4 N54 ignition pass, and an X5d on the lift for a guibo diagnosis.
Active · in motion
-
Working through the Zingg two-book arc
The two-book numerics arc — analytical foundations, then shock-capturing — rebuilt on my own time. CFDLab is the artefact that came out the other side.
Promote FLOMG transonic to a fully-narrated atlas section; add a JST-vs-Roe contact-only crop.
■■■■■□ Python · ARC1D · FLOMG · upwind1d Touched 2026-04-28 -
BMW E70 X5d — drivetrain rumble, suspect the guibo
Diesel X5 with a low-rpm rumble that builds with load. First suspect is the driveshaft flex disc — the guibo. Diagnosis pass before any parts go on.
Inspect the front guibo on the lift, mark and rotate, look for cracking or play before ordering the replacement.
■□□□□□ E70 X5d · lift · hand tools Touched 2026-04-28 -
BMW X1 28i (2017) — coils + plugs reset
Full ignition baseline reset on a 2017 X1 28i — coils and plugs replaced as a single pass so any subsequent misfire diagnosis starts from a known-good state.
Drive a couple of cycles, scan for codes, confirm idle and part-throttle behaviour.
■■□□□□ X1 28i (2017) · hand tools Touched 2026-04-28 -
Yamaha Enticer 250 — bogey wheel sitting on the track edge
Belt replaced and ski repair done. The remaining thread is the bogey wheel that's been riding the inside edge of the track — a geometry problem, not a parts problem.
Pull the rear suspension, measure wheel alignment against the track, and reset the offending bogey before another belt walks.
■□□□□□ Enticer 250 Touched 2026-04-28 -
Gaggiuino — PID and profiling on a Gaggia Classic Pro
Open the Classic Pro, replace its internals with an open-source PID + pressure-profiling controller, close it back up. Started for the value-per-dollar; staying for the repeatability.
Finish the dimmer + pressure-transducer wiring; pull the first profiled extraction.
■■□□□□ Gaggia Classic Pro · Gaggiuino · ESP32 Touched 2026-04-26 -
Ahmed body — y⁺ pass on the baseline grid
Wall-resolved versus wall-modelled. What the residual histogram says about the inflation layer on the M25 baseline.
Re-run with y⁺ < 1 in the wake; compare drag convergence and y⁺ map against the M25 baseline.
■■□□□□ OpenFOAM · Python Touched 2026-04-25 -
What SBP operators are actually doing at a boundary
A field engineer's read of summation-by-parts: where the dual norm earns its name, and where it doesn't.
Tighten the dual-norm bookkeeping for the third-order Gregory closure on a non-uniform grid.
■■■■□□ MATLAB · LaTeX Touched 2026-04-22 -
BMW E89 Z4 — maintenance log
Oil filter housing gasket, ignition coils and plugs, catalyst downpipes, upgraded charge pipe, factory diverter removed for a TiAL BOV. Done in the driveway.
Verify boost behaviour with the new charge pipe + TiAL setup; check for any post-install codes.
■■■□□□ E89 Z4 · hand tools Touched 2026-04-20
Notes that feed back into Work
- Working through the Zingg two-book arc CFDLab — the codebase that came out of it · SBP / Gregory — the MEng numerics context
- Ahmed body — y⁺ pass on the baseline grid aeroAUTO — the Ahmed body workflow · multi-node CFD server — what runs the cases
- What SBP operators are actually doing at a boundary SBP / Gregory operators — the MEng project
Lanes
- ∂ research · active
- turbulence, numerics, CFD analysis
- ✈ fieldbook
- where flying meets the solver
- 🛠 machine · active
- diagnosis, teardown, tuning, hardware
- ☕ extraction · active
- espresso, profiling, repeatable craft
- ※ atmosphere
- place, light, weather
Notes are short — a working artefact, not a publication. For the finished projects, see /work. For the broader story, /about.