Natural and Processed Coffee Roasting
Natural, honey, anaerobic, carbonic maceration, extended-fermentation, and other processed coffees often behave differently from clean washed lots on Roest. This page summarizes how process, moisture, density, bean size, and fermentation intensity affect profile choice, heat application, first crack, development, and evaluation. It is a practical bridge between the dedicated pages on Natural Process Roasting, Washed Process Roasting, Development Time and Drop Decisions, and Roast Defects Troubleshooting.
Process Label Is Only a Starting Point
The process name on the bag is useful, but it is not enough to choose a profile by itself. Some coffees labeled “washed” have long fermentation, thermal shock, double soak, co-ferment, or yeast-inoculated steps and may behave more like processed coffees than like clean washed lots. Conversely, some naturals are clean, dense, and high-altitude enough to tolerate a profile closer to washed coffee.
A double soak should be understood as a washed-style process with two water-contact stages: one example fermented for 36 hours under water, was washed, then soaked for another 24 hours under water source. Honey and extended-fermentation coffees also vary widely; one red honey was treated as needing a natural-profile approach despite a measured humidity of 9.7% source.
Special processing can change density and seed structure, but roast defects are still produced by roast conditions. Christopher Feran notes that anaerobic cherry, carbonic maceration, extended ferment, and low-density coffees can be roasted without tipping, and that tipping cannot be treated as purely a processing defect because it does not appear under all roast conditions source.
Starting Framework for Processed Coffees
Use this table as the canonical starting point for choosing a first profile direction. It does not replace coffee-specific dial-in; it gives a controlled first roast to evaluate before changing one variable at a time.
| Coffee type | Starting profile direction | First-crack target | Development / drop starting point | Heat and airflow shape | Notes |
|---|---|---|---|---|---|
| Clean, dense washed coffees such as Ethiopia, Kenya, Peru, high-altitude Colombia | Washed / denser-bean profile | About 5–6 min on 150 g-style profiles | About 45–60 sec, depending on roast goal | Washed coffees can tolerate faster heat and higher ET than naturals; high-density washed coffees may need ET values around 260–270°C in some profiles | Washed coffees were repeatedly described as taking heat faster and working in shorter roasts 2 sources. |
| Clean naturals and lower-density coffees | Natural / lower-density profile | Around 6:30–7:30, with some natural profiles extending to 8 min | Start around 35–45 sec; reduce toward 25–35 sec for very light natural Ethiopia-style roasts | Lower end heat than washed; keep ET/air temp controlled near and after first crack | Natural profiles have been shared around 150 g, ±7 min FC, lower charge, and post-FC airflow increases; Aris’ natural profile targets FC around 8 min, 45 sec development, max air temp below 245°C, and BT delta after FC below 5°C 2 sources. |
| Heavy naturals, anaerobics, carbonic maceration, “bin juice,” long ferment, sealed fermentation | Natural / processed profile, then reduce end energy aggressively | Usually not a very fast 4–5 min FC; many successful examples sit around 6:30–8+ min | Often 30–40 sec; sometimes less for very intense anaerobics | Get enough energy in early, then cut heat before or around FC so the roast does not take off | Several trusted users report that 4–5 min roasts did not cup well for naturals/anaerobics, while 6:30–7:30 min roasts worked much better; anaerobics may be dropped around 30 sec into crack 2 sources. |
| Honey, red honey, pulp natural, extended-ferment washed, thermal shock washed | Choose by behavior, not label | If it lags, move toward natural; if it takes heat like washed, use washed | Start around 35–50 sec and adjust by cup/color | Watch whether the coffee darkens early, cracks silently, or takes off after FC | A processed washed Colombia El Jaragual tasted better on the washed profile than the natural profile despite the coffee being anaerobic washed; the better roast had only a 5–7°C higher early inlet peak source. |
| High-moisture coffees | Natural-style control, but expect slow initial heat uptake and vigorous crack | Do not assume late early-phase response means the coffee needs unlimited heat | Shorten or manage development if crack arrives hard | More heat may be needed early, but power/ET should be reduced before the roast runs away | Coffee Quest moisture of 12.2% was described as very wet/high water activity; such coffees can be hard to heat initially, then crack early and vigorously, and were recommended for a natural profile source. |
| Very low-moisture or old-crop processed coffees | Gentle, lower-ET approach; avoid over-drying | May show BT dips or collapse around early/mid roast | Do not chase weight loss alone | Avoid extended drying and excessive end heat | A Honduras anaerobic measured around 8.7–8.8% moisture was described as problematic; the recommendation was to dry less and roast with lower ET source. |
For batch-size-specific setup, especially 50 g, 100 g, 150 g, 185–200 g, and Ultra work, defer to Batch Size Scaling and Roest Ultra Guide. Small batches can give misleading BT behavior, while 150 g and above generally provide more usable feedback.
How Naturals and Processed Coffees Differ in the Roast
Naturals and heavy-process coffees often need more attention near first crack because they can darken, speed up, or burn at the surface more easily. The common defect pattern is not simply “too dark”; it is an uneven combination of exterior darkening, tipping, black embryo/tip marks, muted fruit, roasty aftertaste, or a coffee that tastes both developed and underdeveloped.
A repeated community observation is that high heat or high inlet too late in the roast can burn sugars on the exterior of naturals and anaerobics. Several roasters describe the main risk on heavy-process coffees as sugars burning on the outside, sometimes producing black tips, charred notes, or a “black chimney” appearance near the embryo 2 sources.
Processed coffees can also release moisture and pressure differently. Naturals and processed coffees may have faint or silent first crack, early outlier pops, RoR dips or humps, and ET/exhaust dips during the drying phase. These signals should be interpreted together with color, weight loss, aroma, and cup quality rather than treated as precise first-crack markers.
First Crack Is Less Reliable on Naturals and Anaerobics
First crack can be faint, delayed, irregular, or misleading on naturals and anaerobics. Some naturals barely crack, while some processed coffees crack early, vigorously, or continue cracking into the drop. One contributor explicitly uses fixed BT points for such beans because auto first-crack detection is not helpful in that case source.
For repeated work on the same coffee, crack count may help compare batches. For comparing different coffees, crack count is much less reliable because different lots can crack at different intensity and temperature. Patrick J. notes that counted cracks may be useful when repeating a roast with the same bean, but are irrelevant for comparing two different coffees source.
For 120 g and larger batches, some roasters shift toward drop temperature, color, and taste as more stable end markers. Denis reports using two profiles—one for naturals and one for washed—then dropping by BT, testing color, cupping, brewing, and adjusting drop temperature or charge rather than rebuilding the whole profile source.
Development and Drop Decisions
Development on processed coffees is often shorter than on washed coffees, but not universally. Naturals have been reported as fully developed at 25–45 seconds in many Roest contexts, while washed coffees more often tolerate 45–60 seconds. Heavy anaerobics may overdevelop quickly after first crack, especially when crack is quiet or late.
A practical starting range is:
| Coffee behavior | First adjustment |
|---|---|
| Natural tastes heavy, chocolatey, oily, mellow, or lacks fruit | Drop earlier, reduce post-FC heat, or reduce development by 5–10 sec. |
| Natural tastes thin, grassy, tea-like, weak, or under | Add a little more early energy or extend development modestly; avoid only adding late heat. |
| Anaerobic tastes burned, walnut/chocolate, toxic, smoky, or charred | Reduce heat entering/after FC, use shorter development, and inspect tips/interior. |
| Anaerobic tastes raw, leafy, beet-like, or sharply sour | Add structure before FC, avoid excessive drying, and consider slightly more development only if defects are not from green quality. |
| Washed coffee tastes flat from a natural-style profile | Use a washed profile with more appropriate early/mid energy and longer development. |
Several community recommendations put naturals around 30–45 seconds and washed coffees around 45–60 seconds, but the same contributors also report narrower or shorter targets for very light profiles and larger batches. Treat these as starting ranges, then use Development Time and Drop Decisions, Weight Loss Targets, and Color Reading and Measurement to decide the next roast.
Airflow, Pressure, and End-Roast Control
Airflow and pressure matter strongly for processed coffees, but fan percentage alone does not transfer between machines. The same fan percentage can produce different pressure depending on the roaster, venting, batch size, heater fan, and power setting. Detailed setup belongs on Airflow and Fan Settings and Pressure Management.
A recurring processed-coffee strategy is to avoid large, abrupt airflow/pressure jumps unless needed for smoke, chaff, or first-crack control. Low airflow or near-neutral pressure can increase body and internal development, while more negative pressure can clean up smoke and chaff. Denis describes a pressure target of around 0 to yellow, slightly negative after yellow, and no more than about -10 Pa into development source.
Naturals and chaff-heavy coffees may need enough ventilation to remove chaff, but excess exhaust can lift beans toward the top exhaust area or contribute to movement problems. Chaff-bin checks are a simple diagnostic: empty the chaff bin, roast, then inspect whether chaff is missing, dark, or burned source. curated Do not reduce airflow or negative pressure so far that smoke or chaff accumulates in the roast chamber or exhaust path; for chaff-heavy naturals, start with a clean chaff bin and exhaust path, monitor for burning chaff or sustained smoke, and abort/cool safely if chaff ignition is suspected.
Moisture, Density, Bean Size, and Green Defects
Processed coffees often need profile changes because of moisture, density, and bean structure—not only because of the process label. High-moisture coffees may resist early heat and then crack vigorously; low-moisture coffees may collapse, taste flat, or scorch easily. Larger beans may need more time for heat to reach the center, while small Ethiopian naturals may show unusual BT/ET behavior and more sorting issues.
Green quality can dominate roast results. Ethiopian naturals in the claims frequently required post-roast sorting for quakers or unripe beans. One ET2509 natural required 19 beans to be picked out from a 15 g dose, while ET2512 from the same supplier tasted cleaner and had no baggy/astringent/unripe defect source. See Green Coffee Selection for sourcing and defect-selection guidance.
When a natural or honey roast is uneven, first determine whether the unevenness is green-related. Mixed ripeness, screen-size spread, insect damage, or heavy silverskin can make an even-looking roast unrealistic without over-roasting the better beans.
Troubleshooting Processed Coffee Roasts
| Symptom | Likely cause | First fix |
|---|---|---|
| Black tips, black embryo, charred interior, or “chimney” tips | Too much energy into FC, excessive late ET/inlet, or fragile low-density/processed structure | Lower RoR/energy into FC, cut heat earlier, cap end ET/inlet, and inspect beans internally. |
| Exterior dark but interior light | Too much hot air early, high inlet, or too little time for heat penetration | Reduce early inlet/air intensity or lengthen the roast before FC without adding harsh late heat. |
| Sweet but heavy, chocolate/cocoa, oily, low acidity | Too much development or too much end heat for a natural | Drop earlier, reduce development, or reduce end inlet/ET. |
| Tea-like, grassy, leafy, flat, “green fruit,” weak sweetness | Underdevelopment, too short Maillard, too little useful heat before FC, or poor green | Add structure before FC, avoid overly short yellow-to-FC intervals, and compare after rest. |
| Strong ferment, solvent, rubber, soy/meaty, boozy, or “bin juice” | Green processing character, over-fermentation, or oxidation after opening | Lower brew temperature, evaluate with cupping, and avoid assuming roast alone can remove it. |
| Natural tastes good immediately, then poor on days 2–3 | Resting phase rather than necessarily roast failure | Re-evaluate after 7–14 days or longer; avoid making final decisions on day 2–3. |
| Lots of gas, foam, slow drain, or muted early cups | Fresh roast / insufficient degassing | Rest longer, grind appropriately, or brew lower-temperature for heavy processed coffees. |
For visual defect mapping, use Roast Defects Troubleshooting. For phase timing, use Drying and Maillard Phases.
Resting and Brewing Processed Coffees
Processed coffees can change dramatically after roasting. Some taste useful out of the roaster or within a few hours, become poor around day 2–3, and open again after 7–14 days. Other very light or high-air profiles may need 30–45 days sealed before stabilizing 2 sources.
Heavily processed coffees may also decline quickly after opening if they are prone to oxidation. One Castillo honey was recommended for lower brew temperature than washed coffee and finishing the opened bag within 3–4 days because it developed heavier chocolate, dark, over-fermented, and boozy notes after opening source.
For evaluation, a practical cadence is to taste or cup soon after roasting for an early read, avoid over-weighting day 2–3, and then re-check after 7–14 days or after the longer rest used for the profile. See Resting and Degassing and Cupping and Sensory Evaluation.
Meaningful Disagreements and Context
CONFLICT (Unresolved): natural heat strategy. Some guidance emphasizes lower charge, lower air, and gentler handling for naturals to avoid surface burn and tipping; other successful approaches give naturals more heat earlier and then cut harder before or around first crack. The practical synthesis is to avoid uncontrolled late heat, not necessarily to avoid early energy altogether. If a natural lags and tastes underdeveloped, more early energy can help; if it tips, burns, or tastes heavy, reduce peak/late heat and shorten development.
CONFLICT (Unresolved): fast versus slow naturals. Several roasters report poor results from 4:30–5:00 first crack on naturals and prefer 6:30–8:00+ profiles, while others report clean, fruity fast natural roasts in specific 50–100 g contexts. Treat very fast natural roasts as higher-risk and profile-specific rather than universal guidance.
CONFLICT (Unresolved): development time by process. Trusted contributors report natural development targets ranging from roughly 20–45 seconds on Roest, while some Loring or other-machine references use longer development. Batch size, sensor behavior, roast style, and machine type materially change the meaning of “30 seconds” or “60 seconds,” so final decisions should be tied to cup, color, weight loss, and drop temperature rather than seconds alone.