Profile Sharing and Starting Points

Profile sharing on Roest is useful for finding a starting direction, but a shared profile is not a guaranteed recipe. Batch size, model, voltage, fan behavior, pressure, probe behavior, and the coffee itself can all change the result, so the goal is to use shared profiles as structured starting points and then adapt them deliberately.

Shared Profiles Are Starting Hypotheses

A shared profile should be treated as a reference curve, not a transferable result. Multiple experienced users report that the same profile can crack on time on one machine and much later, or not at all, on another; even shared inlet profiles can differ by minutes between machines 2 sources. The most common causes are differences in airflow, exhaust setup, pressure behavior, voltage, heater fan setting, model generation, and batch size.

Inlet profiles are generally easier to share than power profiles because they target a measured inlet temperature rather than a raw heater percentage, but they still require machine-specific adjustment. Power profiles are especially sensitive to available heater power and voltage; one user reported needing 88–90% power at an event roaster where the same 185g profile used 65–70% peak power at home on 243V source. For more on this, see Calibration and Environment, Inlet Temperature Management, and Power Curve Strategies.

BT/IT profiles are more adaptive in principle because inlet changes according to bean-temperature feedback, but they are also the most disputed for sharing. Some users find them useful once tailored; others argue that BT readings vary too much between machines, batch sizes, and probe contact conditions for direct transfer 2 sources. See Bean Temperature Profiling before using BT/IT profiles as a shared starting point.

Starting-Point Procedure

Use this section as the canonical starting workflow. Pick the closest profile by batch size, process, model, and control mode; run one test roast; then change one variable at a time.

Starting profile selection

Use case	Better starting point	Initial targets or settings	Notes
New user learning Roest	150g, or 125–150g if coffee is limited	Use an inlet or Roest library profile designed for that weight	Graphs and BT readings become more useful around 120–150g, while 100g and below can look misleading 2 sources.
Washed coffee, 150g	Washed 150g inlet/profile start	Preheat/profile reference around 200°C; washed charge around ±150°C drum temp; aim FC around 5–6 min; start development around 50–60 sec if using a conventional washed approach 2 sources.	Good default for washed Ethiopia, Kenya, Colombia-style profiles, then adjust to taste.
Natural / lower-density / non-washed, 150g	Natural or “anything else” 150g profile	Preheat/profile reference around 170°C; charge around ±125°C drum temp; aim FC around ±7 min; bump air shortly after crack if the profile calls for it source.	One Aris-developed profile is specifically described as working well on naturals and lower-density beans source.
50g samples	50g inlet/time or sample profile	Expect FC around 5:30 on Luca-style 50g profiles, or around 226–230°C ET on one ET-increase sample profile; development may be very short, around 28–34 sec on that ET-increase profile 2 sources.	Best for tiny offer samples and cupping, not for graph interpretation or brewing evaluation. Do not expect RoR/BT graphs to make sense.
100g samples	100g inlet/time profile rather than old ET profiles	Use a known 100g profile and judge by cup, color, and timing; do not expect clean BT/RoR behavior	Older 100g ET profiles predate inlet-temperature workflows and are often discouraged now source.
120–130g	Scaled 100g profile or BT/inlet profile with enough bean mass	Minimum 120g is suggested for BT/inlet profiles that need more accurate BT readings; some users prefer 125–130g over 120g 2 sources.	Useful compromise when 150g is too much green to spend.
165–185g consumption roasting	Profile designed for 166–185g or 175–185g	Use the profile’s intended batch size closely; small weight changes can shift timing	Several later workflows are built around 166–185g batches, especially for repeatable home consumption roasting source.
Ultra / counterflow	Counterflow-specific profile	Reduce inlet substantially versus L100/classic-mode references, often by about 40–60°C, then fine-tune	Counterflow transfers heat differently and can run much faster; simple temperature offsets may not fully map profile shape 2 sources.

Adaptation steps

Match the batch size first. Do not run a 50–100g profile at 180–200g, or a 180g profile at 100g, and expect a usable result. Batch-size changes alter graph shape, power needs, airflow behavior, and probe contact 2 sources.
Match the control mode. If the source profile is inlet/time, adapt inlet values. If it is power, adapt power. If it is BT/IT, evaluate whether BT readings on the target batch size are stable enough before trusting the adaptive behavior.
Run the first attempt close to the shared profile only if the machine conditions are close. If the machine is known to run fast, slow, unusually positive/negative in pressure, or on a different voltage, treat the first roast as a calibration roast rather than a production roast.
Use the first-crack timing as a coarse profile-fit check, not an absolute truth. For a profile targeting FC around 7–8 minutes, one suggested adjustment is to increase inlet values by 3–5°C if the coffee does not crack in that window, or decrease by about 3°C if it cracks before 7 minutes source. For profiles targeting around 6 minutes, adjust toward the stated target rather than applying the 7–8 minute rule.
Use offsets before redrawing the profile. Common first moves include adding or subtracting a uniform inlet offset, adjusting charge temperature, or changing development/drop by a few seconds. Several users recommend changing one parameter at a time and logging the effect 2 sources.
When scaling batch size, use small, explicit changes. Moving from 100g to 125g has been described as needing extra energy, with one recommendation of a maximum 3–5% increase source. For 150g versus 100g power profiles, guidance cited in the community suggests increasing the overall profile by about 5–10% power source.
For 110V machines, consider inlet compensation. One shared-profile note recommends adding 10°C to all inlet points for US/110V machines source. Treat this as a starting offset, then verify by roast timing and cup quality.
Evaluate with cup, color, and weight loss, not graph shape alone. Graphs can mislead, especially at smaller batch sizes. Color plus graph is more useful than graph alone, and weight loss depends on profile and moisture 2 sources.

Choosing a Profile Type

Inlet profiles are the most common shared starting point because they are more machine-independent than raw power percentages and are considered one of Roest’s strengths source. They still need adjustment for batch size, airflow, pressure, and model differences.

Power profiles can be simple and useful, especially for learning, but they are less transferable. A profile that cracks at 5 minutes on one machine may crack at 7+ minutes on another, and percentage power is not equivalent across voltage and machine conditions 2 sources.

ET or air-temperature profiles are historically important and can work, but several users caution that ET shape changes with bean, batch weight, RPM, and sensor behavior. Older 100g ET profiles are therefore better treated as historical references than as default starting profiles source.

BT/IT profiles can be powerful when BT readings are reliable, but the community disagrees on their portability. They are more appropriate at batch sizes where BT readings are stable, and less appropriate for 50–100g on L100/S100-style machines unless the user understands the limitations. See Bean Temperature Profiling and Batch Size Scaling.

Meaningful Conflicts

CONFLICT (Unresolved): BT/IT profile sharing. One view is that BT/IT sharing is effectively impossible because machine and BT behavior vary too much; another view is that it is not impossible, but must be tailored to the machine source. This is best resolved practically: copy the shape, not the promise, and verify with your own timing, color, and cup.

CONFLICT (Unresolved): shared profile transferability. Some contributors say profiles cannot be shared for exact results between machines, while others report that profiles can work well if exhaust and pressure conditions are set up similarly 2 sources. The safe synthesis is that profiles are shareable as starting points, not as calibrated recipes.

CONFLICT (Temporal Evolution): older ET sample profiles versus newer inlet workflows. Earlier 50–100g and ET-based profiles were used before inlet-temperature workflows became common; later guidance often favors inlet/time or inlet-based profiles for small batches and standard 150g inlet profiles for more stable data 2 sources.

A profile link alone is rarely enough for another user to reproduce a result. Useful shared-profile notes include:

Roest model and mode, including L100/L100+/S100/Ultra/P3000 and classic versus counterflow.
Batch weight and process.
Heater fan RPM or heater fan percentage if changed from stock.
Exhaust/fan behavior and any pressure reference if measured.
Voltage or region if relevant.
Charge drum temperature, first crack time, drop time or drop BT, color, and weight loss.
Green moisture or density if available.
Cup notes after a stated rest period.

Well-labeled numerical data is especially useful when tracking weight loss and moisture content source. For post-roast interpretation, see Cupping and Sensory Evaluation, Color Reading and Measurement, and Weight Loss Targets.

Practical Mindset

A good starting profile should reduce the number of unknowns, not eliminate the need to roast. The strongest community pattern is to pick a sensible profile, hold batch size steady, make one change at a time, and judge the result by taste, color, weight loss, and repeatability. For a first Roest learning path, see Getting Started With Roest; for diagnosing bad outcomes after a starting profile, see Roast Defects Troubleshooting.