drivurs logo
Tuning 4 min read

Clean pull vs dirty pull: why comparisons fail

Heat, slope, traction, and driver inputs—how to keep pulls comparable and avoid false conclusions.

Drivurs Team

Key takeaway:

A “dirty” pull adds uncontrolled variables; a clean pull controls the environment so changes reflect the car—not the conditions.

TL;DR

  • Tuning is changing targets and limits so the ECU can control torque safely under your real conditions.
  • Modern ECUs are torque-based: they decide airflow, fueling, and spark to hit a torque request.
  • Validation beats screenshots: consistent conditions and repeat runs are the only way to trust changes.

A mental model (text diagram)

Driver intent → Torque request → ECU torque model → Airflow (boost/throttle) + Fuel + Spark
                 ↑                                  ↓
               Sensors (air, temps, knock, fuel) ← Outcome (torque delivered)

What makes a pull “clean”

  • Stable temperatures (or at least recorded and similar)
  • Same road and direction (grade and wind)
  • Same gear and RPM range
  • No traction events (no spin, no shifts mid-measurement)

What makes a pull “dirty”

  • Heat soak differences
  • Grade/wind differences
  • Traction changes
  • Driver input differences

Validation (repeatability checklist)

Use this checklist any time you change hardware or calibration. The goal is not a single “hero pull.” The goal is repeatable behavior you can trust.

1) Control the variables

  • Use the same gear and the same RPM range for comparisons.
  • Use the same road and direction (grade and wind matter).
  • Keep tires and pressure consistent when testing performance.
  • Watch temperatures: compare runs at similar IAT/coolant/oil conditions.

2) Change one thing at a time

  • If you change hardware and the tune at the same time, you will not know which change caused the outcome.
  • Make one change, log it, validate it, then move on.
  • One run is noise. Multiple runs under similar conditions create a trend.
  • If the first pull is great but later pulls fall off, you likely have a heat/margin problem—not a “peak power” problem.

If you see this, stop (safety signals)

This list is intentionally conservative. Reduce load, verify maintenance and fuel quality, and diagnose before continuing.

  • Persistent knock correction under the same conditions.
  • Misfire under boost (often feels like breakup or “stuttering”).
  • Rapidly rising temperatures run-to-run (heat soak) with worsening behavior.
  • Sudden torque reductions or throttle closures that weren’t present before a change.

If you are not sure what a log means, default to less load and more margin. This page is educational and not legal advice.

Quick checklist (before you change anything)

Use these questions to keep the process disciplined. You can answer all of them in a few lines of notes, but skipping them is how builds become confusing.

  • Goal: What are you optimizing for (daily drivability, track consistency, drag times, or “balanced”)?
  • Baseline: What is the car doing right now, and under what conditions?
  • Constraint: Is the limiter traction, heat, fueling capacity, ignition stability, or torque limits?
  • Variable control: Can you repeat the test with the same gear, road, and temperature range?
  • Single change: What is the one thing you are changing today?
  • Expected outcome: What should improve if the change works (and what tradeoff might get worse)?
  • Stop condition: What would make you back off immediately (knock, misfire, temperature, or intervention)?

Common mistakes

  • Using a single “hero run” as proof.
  • Changing multiple mods and doing one pull.

Diagnostics / what logs tell you (high level)

SignalWhat it usually meansWhat to check
RPMContext for everything elseCompare in the same gear and load range
Throttle angle / torque interventionHow the ECU is controlling torqueLook for closures that explain boost drop
Boost target vs actualControl quality and limitsOscillation can be hardware or control strategy
IAT / tempsMargin and repeatabilityHeat soak changes results dramatically
Knock / timing correctionCombustion safety responseSustained corrections = reduce load/verify fuel/temps

FAQ

How many runs is enough?

More than one. Look for repeatability under similar conditions, not the best number.

Want to keep learning?

Browse the Drivurs Academy hubs for checklists, comparisons, and reference.