Roaster: Gene Café CBR-301
Batch: 250 g
Set Temp: 250°C
Preheat: None
Fan: 2 → 3 at 4:27
Cooling initiated: 12:51 (771 s)
Output: 210.1 g
Weight Loss: 16.0%
Development (from crack flag): 83 s

Intent:
Repeat Roast 20 and align structural outcome to Roast 19 (~14.3% WL).


Why This Roast Exists

Roast 19 (Honduras Catuai Washed) landed at 14.3% weight loss and felt structurally balanced.

Roasts 18 and 20 (Guatemala Caturra Washed blend) both landed at ~16.1% weight loss.

Roast 21 was intended to bring the Guatemala coffee down into the same structural band as Roast 19 to allow a fair comparison.


Roast Behaviour

The roast behaved almost identically to Roasts 18 and 20.

  • Drying progressed normally.
  • Chaff began exiting in the expected 5–7 minute window.
  • First crack was flagged at 11:28.
  • Development time from flag to cool was 83 seconds.
  • Cooling initiated at 12:51.

No instability.
No unexpected airflow behaviour.
No thermal crash.

From a machine standpoint, this was stable and repeatable.


What Actually Happened

Despite the intention to reduce weight loss, the roast landed at:

16.0% WL

This places Roast 21 structurally in the same band as:

  • Roast 16
  • Roast 18
  • Roast 20

This is now a repeat-confirmed outcome for this coffee under this method.

The system is consistent.


Why It Did Not Align With Roast 19

The assumption going into Roast 21 was:

If development seconds are held similar (~80 s), reducing total roast time slightly would reduce weight loss.

What this roast suggests instead:

For this Guatemala Caturra blend at 250 g / 250°C / fan 2→3, the machine + coffee combination naturally settles around 16% WL when development is ~80 seconds.

Roast 19 was a different coffee (Honduras Catuai Washed).
Even with similar development timing, it reached 14.3%.

The difference is not only development seconds.
It is:

  • Pre-crack energy accumulation
  • Moisture content and density
  • How this specific coffee converts applied energy into mass loss

This is a coffee response difference, not machine instability.


Structural Interpretation

At 16% WL, this roast sits in your established “fuller medium” band.

Expected structural traits (pre-cupping):

  • Increased body
  • Slightly reduced brightness
  • More caramelised sweetness
  • Slight compression of clarity compared to 14–15% band

This aligns with previous roasts of the same coffee.


What This Confirms

  1. The CBR-301 is thermally repeatable under identical settings.
  2. This Guatemala Caturra blend consistently lands at ~16% WL under ~80 s development.
  3. Development time alone is not the primary driver of structural band.
  4. Total applied energy and bean response matter more than crack timing alone.

This roast was not a misfire.

It was confirmation.


What Changes Next

If the goal is to move this coffee into the 14.5% band, one of the following must change:

  • Reduce development to ~60–70 seconds
  • Initiate cooling ~25–35 seconds earlier
  • Or adjust total applied energy pre-crack

Simply “aiming lower” while holding development at ~80 seconds appears insufficient.

That is now an evidence-based observation.


Current Status

Roast 21 reinforces that:

The 16% band is stable for this coffee under current parameters.

The 14.3% band of Roast 19 belongs to a different coffee response pattern.

The dataset is beginning to show coffee-specific structural signatures.

No conclusions about preference yet.

Cupping will determine whether this 16% band is actually optimal for this coffee.