Coffee | One Roast at a Time

250g → 210.5g (15.8% WL)
Cold Start | 250°C Set | Cool at 12:40

Context

Roast 35 is the second roast of this Kenyan AA Natural and the first deliberate calibration roast under the new ventilation setup.

Ventilation change:

  • Previous setup: active exhaust connected to roaster.
  • Current setup: kitchen oven hood extractor running at full power.
  • Roaster exhaust directed toward hood, but not physically connected.
  • Extraction affects room air only, not the roaster airflow path.

This roast was intentionally designed to reduce weight loss from Roast 34 (16.2%) toward the 15.0–15.5% range.

Roast 34 Reference (For Context)

Roast 34:

  • Cool: 13:00
  • Weight out: 209.5g
  • WL: 16.2%
  • Ambient: 23°C
  • Reached 200°C: 04:00
  • Reached 230°C: 06:43

Objective for Roast 35:
Reduce total time by ~20–30 seconds to test WL slope behaviour.

Roast 35 – Data

  • Ambient: 21°C
  • Reached 200°C: 04:14
  • Reached 230°C: 06:53
  • Cooling initiated: 12:40
  • Weight in: 250g
  • Weight out: 210.5g
  • Weight loss: 15.8%

Cooling time: 07:45

Early Phase Behaviour (Drying / Ramp)

200°C was reached 14 seconds later than Roast 34.
230°C was reached 10 seconds later.

This minor shift is consistent with:

  • Slightly cooler ambient
  • Normal roast-to-roast thermal variance

There was no evidence of:

  • Accelerated ramp
  • Heat stall
  • Irregular fan behaviour
  • Environmental airflow interference

Observation:
The oven hood ventilation appears neutral in terms of roast thermodynamics.

Weight Loss Calibration Behaviour

Time reduction:
13:00 → 12:40 (–20 seconds)

WL reduction:
16.2% → 15.8% (–0.4%)

This confirms a predictable slope:

On this system:

  • ~20 seconds ≈ ~0.4% WL
  • ~50 seconds ≈ ~1% WL

This matches previous Santos mapping and confirms consistency across origins.

Key lesson:
The Gene Café responds linearly to total applied time under stable conditions.

Ventilation Observation

This roast confirms:

  • External environmental extraction does not materially alter roast progression.
  • The roaster’s internal airflow remains system-controlled.
  • Milestone temperatures (200°C, 230°C) are stable and repeatable.
  • No WL spike or unexpected acceleration occurred.

Conclusion:
The oven hood setup improves room air quality without destabilising roast behaviour.

This simplifies future calibration.

Structural Position of Roast 35

At 15.8% WL on a 1400–1800m Kenyan natural:

Expected cup structure:

  • Medium development
  • Balanced body
  • Reduced earthiness vs 16.2%
  • Increased clarity vs Roast 34
  • Potentially more fruit definition
  • Slightly less density

This sits between:

  • Bright espresso edge (15.0–15.3%)
  • Fuller, heavier espresso (16.0–16.5%)

Roast 35 becomes the mid-band reference for this bean.

Cooling Behaviour

Cooling time reduced from 08:20 (Roast 34) to 07:45.

This likely reflects:

  • Slightly lower internal bean energy at drop
  • Slightly reduced development momentum

No irregular cooling behaviour observed.

Engineering Lessons

  1. Cold starts remain repeatable.
  2. The oven hood configuration is thermally neutral.
  3. WL response to time is predictable.
  4. The system remains stable without manual intervention.
  5. Kenyan natural behaves consistently relative to Santos WL slope.

Most importantly:

The machine is behaving predictably.

And predictable behaviour is a prerequisite for:

  • Between batch process development
  • Guardrail design
  • Future repeatability work

Open Questions (To Be Tested)

  • Does 15.8% materially reduce the earthy tone noted in Roast 34?
  • Does fruit clarity increase at this WL?
  • Does espresso extraction feel cleaner?
  • Does body drop noticeably?

These will only be answered through tasting at proper rest (minimum 4–5 days).

Current Position

Roast 34: 16.2%
Roast 35: 15.8%

The WL band for this Kenyan is now mapped with two reliable data points.

Next logical calibration point (if required):
~12:25–12:30 drop → ~15.2–15.4% WL

But no further adjustment should occur until tasting confirms direction.

Final Reflection

This roast was not about chasing flavour.

It was about confirming system behaviour under a ventilation change and validating WL slope control.

That objective was met.

Coffee.
One roast at a time.