JDM Solution · Session Report
Revision 4 — WOT fuel correction + timing pull at 220 kPa
July 27, 2026 · 1.75 h against the engagement
- Vehicle
- 1993 Nissan Skyline GT-R (BNR32)
- Engine
- RB26DETT · 2.6 L · stock internals
- ECU
- Haltech Elite 2500 (NSP)
- Induction
- Twin Garrett GT2860RS · external wastegate
- Fuel
- 93 octane pump
- Injectors
- 1,000 cc @ 3 bar
- Engagement
- Initial Dial-In — week 4 of 6
- WOT target
- 11.4:1 (ceiling 11.6:1)
- Program ceiling
- 20 psi
Log analysed
session-04_wot-pulls.csv
12 min · 20 Hz · 3rd and 4th gear pulls, 3,000–7,000 RPM · 22 °C ambient, dry
1. Safety gates
The same go/no-go checks run against every log before a single value is changed. Two of them moved this session; both are addressed in the changes below.
| Gate | Limit | This session | Verdict | Note |
|---|---|---|---|---|
| WOT air/fuel | never leaner than 11.6:1 | 11.9:1 | Marginal | 0.4 s at 4,800 RPM on pull 3, as the turbos came fully on. Corrected in changes 1–4. |
| Knock, any cylinder | 0 events above 0.30 V over noise floor | 3 events, 0.42 V peak | Action taken | Cylinders 5 and 6, 5,600–6,400 RPM, end of pull 3 only. Timing pulled 1.0° in changes 5–7. |
| Injector duty cycle | ≤ 85% | 71% peak | Clear | Headroom for the boost step and the flex-fuel work later in the program. |
| Boost overshoot | ≤ 1.5 psi over target | 0.6 psi | Clear | Feed-forward wastegate duty is doing the work; closed-loop correction stayed under 4%. |
| Oil pressure at 6,500 RPM | ≥ 60 psi | 68 psi | Clear | Steady across all three pulls, no drop on the over-run. |
| Intake air temperature | ≤ 55 °C | 47 °C peak | Clear | Climbing pull-to-pull, which is why the IAT timing correction is being steepened. |
| Long-term fuel trim drift | ≤ ±5% | +2.1% | Clear | Stable session to session — the base fuel model is tracking, not chasing. |
2. Map review — verdict
The fuelling model is sound but consistently lean through the 4,800–5,600 RPM band under full boost. The VE table's shape is correct — it tracks the wideband everywhere outside that band — it is simply about 4% short where the turbos come fully on, which is the classic signature of a table that was extrapolated rather than measured at boost. Above 5,600 RPM the limiter is timing, not fuel: three knock events on cylinders 5 and 6, all at the end of the third pull once intake air temperature had climbed past 45 °C. That pattern reads as heat sensitivity, not a mechanical or fuelling fault, and it is why the IAT correction is being steepened alongside the flat 1.0° pull.
Nothing in this log suggests a hardware problem. Oil pressure held 68 psi at 6,500 RPM across all three pulls, injector duty peaked at 71% with real headroom left, and long-term fuel trims have drifted +2.1% over four sessions — stable, not chasing. The wideband and the ECU's model agree closely everywhere except the corrected cells, which is what we want to see before adding boost.
Next step: enter changes 1–9 and re-run the same capture set from your logging checklist — same road, same gears, same three-pull pattern, ideally at a similar ambient temperature. Hold changes 10 and 11 (the boost step) until that log confirms the WOT band sits at 11.4:1 ±0.15 with the knock window clear. One variable at a time; the data decides when we go up, not the calendar.
3. Itemized changes
Exactly what to type in, and why. These rows are what appear in your portal with a tick box each — no interpretation required, and no change without a reason attached to it.
| # | Table | Cell | Old | New | Reason |
|---|---|---|---|---|---|
| 1 | Main Fuel (VE) | 4,800 RPM · 220 kPa | 88.4 | 92.3 | Pull 3 logged 11.9:1 against an 11.4:1 target — the leanest point in the session. +4.4% VE puts the cell on target. |
| 2 | Main Fuel (VE) | 5,200 RPM · 220 kPa | 90.1 | 93.7 | Same pull, 11.85:1 measured. +4.0% VE — the table's shape is right, its magnitude is short where the turbos come fully on. |
| 3 | Main Fuel (VE) | 5,600 RPM · 220 kPa | 91.6 | 94.4 | 11.75:1 measured across all three pulls. +3.1% VE onto target. |
| 4 | Main Fuel (VE) | 6,400 RPM · 220 kPa | 89.2 | 86.5 | Richer than commanded at 11.05:1 — costing a little power without buying safety. −3.0% onto target; still well inside the 11.6:1 lean ceiling. |
| 5 | Ignition Angle | 5,600 RPM · 220 kPa | 19.5° | 18.5° | Three knock events on cylinders 5 and 6 in this window, peaking 0.42 V over the noise floor, all on the third pull once IAT passed 45 °C. −1.0° at the cell that knocked. |
| 6 | Ignition Angle | 6,000 RPM · 220 kPa | 20.0° | 19.0° | The knock window carried into this cell. Matching the 1.0° pull keeps the curve smooth rather than putting a notch in it. |
| 7 | Ignition Angle | 6,400 RPM · 220 kPa | 20.5° | 19.5° | Precautionary 1.0° so the top of the curve stays consistent with the two cells below it. We revisit this once we have an IAT-matched log. |
| 8 | IAT Ignition Correction | 45 °C | −1.0° | −2.0° | IAT reached 47 °C by the end of pull 3, which is where the knock appeared. Steepening the correction gives timing back automatically on heat-soaked pulls instead of relying on you to lift. |
| 9 | Main Fuel (VE) | 240 kPa row · 4,800–6,400 RPM | copy of the 200 kPa row | corrected 220 kPa values +1.5% | At 18 psi the ECU interpolates between the 220 and 240 kPa rows. Seeding 240 from the corrected 220 values means the first pull at the new target isn't lean while we wait for data to land there. |
| 10 | Boost Target (4th gear)Hold | 4,000–7,000 RPM | 17.0 psi | 18.0 psi | The approved +1.0 psi step toward the 20 psi program ceiling. HOLD THIS ROW: enter it only after the validation log shows changes 1–9 landed — WOT band at 11.4:1 ±0.15 and the knock window clear. |
| 11 | Wastegate Duty (feed-forward)Hold | 5,000 RPM | 42% | 45% | Paired with change 10 — reaches the new target on feed-forward instead of leaning on closed-loop correction, which is what kept overshoot at 0.6 psi. Enter it at the same time, not before. |
Rows marked Hold are staged deliberately: they are the next step up, and they only get entered once the log proves the step before them landed.
4. Next capture set
What to drive and what to log, so the next session measures the same thing this one did.
- Same route, same 3rd/4th gear pulls, three of them, from 3,000 RPM to the 7,600 RPM limiter.
- Log at 20 Hz with the standard channel list — no changes to channels or sample rate this round.
- Two cool-down minutes between pulls so IAT starts each pull in the same place.
- Upload the log to your portal as the validation log for Revision 4 before making any further changes.