GUIDE · Methodology

From Feel to Data: A Three-Pillar Methodology for Motor Analysis

Most racers rely on experience — "feels good," "sounds right." The problem with experience is that it can't be transferred, reproduced, or verified. This article introduces a three-step methodology that turns motor analysis from "feel" into systems engineering: measurement, comparison, health assessment. It's the conceptual framework behind MotorLab's entire analysis system.

Why motor analysis needs a methodology

Common problems in motor tuning:

  • Testing RPM at different voltages leads to contradictory conclusions
  • Two "broken-in" motors produce data that can't be compared
  • Pre-race the motor feels slower, but you can't say what changed

The root cause isn't "wrong technique" — it's the absence of a shared analytical framework.

Pillar 1: Measurement

The goal of measurement isn't "to get numbers." It's to reconstruct a reproducible test environment.

Measurement standards (must be fixed):

  • Voltage — fixed or stepped control, source impedance recorded
  • Time — consistent warm-up, fixed test window
  • Environment — controlled temperature and load

MotorLab's 5 core measurement metrics:

MetricReflects
CurrentResistance and efficiency
RPMOutput capability
ThermalEnergy loss
StabilitySystem consistency
VibrationMechanical health

Measurement is not data collection. It is environment reconstruction.

Pillar 2: Comparison

A single number means nothing — is 18,500 RPM "good" or "bad"? The answer is always "relative to what."

Three comparison types:

  • Self-comparison (time series) — The most important. The same motor across tests #1, #10, #30. You're not reading absolute values — you're reading trend direction.
  • Population benchmark — Cross-individual. Compared against the population of the same motor type. Output: percentile ranking, baseline deviation.
  • Efficiency comparison — Cross-model. Comparing RPM/Current or Output per Watt — avoids being misled by single extreme values.

Common comparison mistakes: comparing RPM at different voltages, comparing under different loads, reading peak values while ignoring averages, missing the time dimension.

Comparison is not magnitude evaluation. It is deviation calculation.

Pillar 3: Health Assessment

Health isn't "how fast" — it's "can the motor still produce predictable behavior."

Four health dimensions:

DimensionWhat to watch
PerformanceRPM trend, output decay rate
EfficiencyCurrent rise trend, energy loss ratio
StabilityRPM fluctuation, current noise
MechanicalVibration level, abnormal frequency patterns

Three health states:

  • Normal — gradual change, smooth curve
  • Degrading — consistent downward trend, efficiency loss
  • Abnormal — sudden deviation, unstable signals

Health Score (0–100) — composite metric based on baseline deviation:

  • 90–100: Baseline stable
  • 70–90: Normal operation
  • 50–70: Degradation phase
  • < 50: Abnormal / end-of-life region

Health is not a snapshot value. It is a trajectory evaluation.

How the three pillars connect

Measurement → defines reproducible environment
Comparison → calculates deviation from baseline
Health → evaluates trajectory state over time

The three form a complete chain: without measurement there's no basis for comparison, without comparison there's no way to assess health, without health assessment you're back to deciding by feel.

Core idea

MotorLab's real core isn't "making motors faster." It's:

Converting invisible feel into measurable and comparable system states.

Performance improvement is a result — the methodology is the foundation. When you can reproducibly measure, objectively compare, and quantitatively assess health, motor tuning stops being luck and becomes engineering.