Brazil Yellow Bourbon Natural BR2504 — First Restored-Heater Validation Roast
Coffee: Brazil Yellow Bourbon Natural BR2504, 88graines
Roaster: Gene Café CBR-301
Machine state: Newly replaced 220 V heater installed
Batch: 250 g
Output: 212.7 g
Weight loss: 14.9%
Target temperature: 250°C
Start temperature: 24°C
Cooling / data end: approx. 13:13
Why this roast mattered
This roast was not really about producing the perfect Brazil.
It was a calibration roast.
After the earlier repair phase, the machine had been landing much lighter than expected. Roast 98, using a broadly similar Brazil Yellow Bourbon Natural approach, finished around 13:16 but produced only 12.7% weight loss. That made the machine feel underpowered or at least not aligned with its earlier behaviour.
With the replacement 220 V heater now installed, Roast 99 was intended to answer a simple question:
Is the machine back closer to its original roasting behaviour?
The early answer appears to be yes — but with an interesting caveat.
Key roast markers
| Marker | Time | Temperature |
|---|---|---|
| 200°C | 3:58 | 200°C |
| 230°C | 6:04 | 230°C |
| 245°C | 7:41 | 245°C |
| 250°C | 8:15 | 250°C |
| Cooling / end | ~13:13 | 247°C |
Fan behaviour
| Event | Time | Temperature |
|---|---|---|
| Fan 2 engaged | 0:01 | 23°C |
| Fan 2 → 3 | 5:58 | 228°C |
| Fan 3 → 2 | 11:33 | 245°C |
| Fan/stirring 3 at end | 13:13 | 247°C |
What I observed
The roast immediately felt more energetic than the recent post-repair roasts.
Reaching 200°C in under four minutes was the first reassuring sign. That was much closer to older, healthy-machine behaviour than to the weak/light-landing phase.
The next major sign was the pace through the upper band. The machine reached:
- 230°C at 6:04
- 245°C at 7:41
- 250°C at 8:15
That is a strong upper-band climb.
The final weight loss confirmed that this was not just a graph impression. At roughly the same total roast time as Roast 98, Roast 99 landed at 14.9% WL instead of 12.7%.
That is a major difference.
Current interpretation
This looks like a successful restored-heater validation roast.
The machine no longer appears to be behaving like the weak post-repair / non-optimal-heater state.
But I am not yet ready to say it has simply returned to the original machine behaviour either.
The more interesting possibility is that this is now a third machine personality:
- Original / pre-failure machine
The baseline I learned on originally. - Post-repair / non-optimal heater state
Lighter landing, slower or less effective, with Roast 98 as a key example. - Restored 220 V heater state
Strong again, but possibly more energetic in the upper band than expected.
That is only a working idea after one roast. It may disappear once more data comes in. But Roast 99 is strong enough that it deserves to be marked clearly.
Current feeling
Relief, mostly.
This roast felt like the machine had come back to life.
But also caution.
It would be very easy to overreact to one successful roast and declare the whole story solved. I do not want to do that.
For now, this is not a conclusion. It is a signal.
A very encouraging signal.
Current assumptions
These are assumptions, not facts:
- The new 220 V heater has restored much stronger effective roasting energy.
- Roast 98 was likely representative of the non-optimal post-repair heater phase, not the machine’s proper behaviour.
- The restored machine may not behave exactly like the original pre-failure machine.
- The later fan 2→3 transition in Roast 99 may have allowed stronger movement through the 230–250°C range.
- A 13:10–13:15 Brazil Yellow Bourbon Natural roast on the restored machine may now sit around 14.5–15% WL.
All of these need confirmation from more roasts.
What I should not conclude yet
I should not yet conclude that:
- the machine is fully stable again
- this new heater state exactly matches the original machine
- 13:15 will always produce 14.9% WL
- the fan behaviour is fully understood
- the darker-roast experiment can be planned using post-repair timing assumptions
- one roast proves the whole repair story
One roast is enough to raise confidence.
It is not enough to close the case.
Practical next step
The next step should be another calibration roast, not a dramatic experiment.
Ideally:
- same or similar coffee
- 250 g batch
- 250°C target
- cold start
- planned stop around 13:10–13:15
- record 200°C / 230°C / 245°C / 250°C
- record fan behaviour
- record output weight and WL
- compare cup result after rest
If the next calibration roast also lands around 14.5–15% WL with similar milestone behaviour, then the restored-heater baseline becomes much more credible.
Working note to future me
Do not read this note later as certainty.
Read it as the first moment the machine seemed to come back.
Roast 99 suggests the new heater has restored the CBR-301 to a strong, usable roasting state. But the exact personality of that restored state still needs a few more roasts before it can be trusted.
For now:
Promising. Energetic. Possibly a third personality. Not proven yet.