How the simulator models Cold Brew extraction

Cold Brew in the simulator

Cold Brew is the most different method of all those implemented in the simulator: it uses no heat. Ground coffee steeps in cold water for many hours. Extraction is slower and more selective: it extracts fewer volatile acids and more sweet and full-bodied compounds.

The simulator models Cold Brew with three basic parameters: grind, ratio and total steep time in hours. Advanced mode adds fridge temperature and water hardness.

Model note: the simulator does not measure real extraction, but a heuristic index (0–100) that combines the brewing variables. That index determines both the position on the extraction map and the shape of the flavor radar. It is not a TDS percentage or a real extraction yield (EY).

---

Simulator variables: Cold Brew

Grind (0–100)

Grind in Cold Brew has the lowest extraction weight of all drivers in the model (up to 25 points), reflecting that in cold extraction grind affects the final result less than in hot methods. This is because low temperature already makes extraction slower and more uniform.

However, grind still matters:

• Fine grind with long steep: risk of over-extraction. More bitter and astringent.
• Coarse grind: slower extraction. Appropriate for long steeps (20–24 h).
• Very coarse grind with short steep: likely under-extraction.

Typical range in real Cold Brew: very coarse, coarser than French Press. In the simulator, values of 20–50 are representative.

Ratio (1:4–1:10)

The Cold Brew ratio in the simulator goes from 1:4 to 1:10 (water-to-coffee). Cold Brew is typically prepared as a concentrate (1:4–1:8) that is then diluted before serving. Some prefer to drink it straight at 1:10.

• Low ratio (< 1:6): intense concentrate. The model adds up to 20 extraction points for low ratios and directly increases body in the radar.
• Standard ratio (1:6–1:8): usual concentrate zone.
• High ratio (> 1:8): direct drink, less concentrated and lighter.

Ratio carries high weight in the model (20 points), comparable to steep time. In Cold Brew, final concentration depends on both ratio and time.

In the model, ratio does not only dilute the drink: it also increases the extraction index by raising the water-to-coffee ratio available during the steep. A higher ratio means more water in contact with the same coffee for the same number of hours, extracting more solubles even at equal grind and time.

Total steep time (8–24 hours)

Total steep time is the most characteristic parameter of Cold Brew. The simulator range is 8–24 hours.

• Short steep (8–12 h): less extraction. More acidic and lighter flavor.
• Medium steep (14–16 h): balanced zone for most coffees with standard grind.
• Long steep (20–24 h): more extraction. Can produce bitterness and astringency with fine grind.

Normalized time is calculated as (totalTimeH − 8) / 16. This means the model only captures variation between 8 and 24 hours; times outside that range are clamped.

Time contributes up to 18 points, the highest individual weight in the Cold Brew model (more than grind or ratio).

Fridge temperature (advanced, 2–8 °C)

Fridge temperature affects extraction speed. In the simulator, the range is 2–8 °C.

• Higher temperature (6–8 °C): slightly faster extraction. Up to 8 additional points compared to 2 °C.
• Lower temperature (2–4 °C): slower and more selective extraction.

In real practice, many Cold Brews are steeped at room temperature (~20 °C) and then refrigerated. The simulator only models refrigerator steeping; room-temperature steeping would produce very different results (much faster) that are outside the model's range.

Water hardness: GH and KH (advanced)

GH also directly affects extraction in Cold Brew. The model includes GH as an extraction variable (waterN = min(waterGH / 12, 1), with up to 5 additional contribution points). More mineralized water extracts slightly more when cold.

KH is not modeled in the Cold Brew sensory profile the same way as in hot methods; the Cold Brew model has a more simplified sensory projection.

---

How it reflects in the extraction map

The map positions the recipe in the grind-ratio space:

• The balanced zone for Cold Brew is reached with medium-coarse grinds (25–55) and medium ratios (1:5–1:8) with steep times of 14–18 hours.
• Steep time is the main secondary modifier: more hours clearly shift the point toward more extraction.
• Fridge temperature has a more subtle effect.

---

How it reflects in the flavor radar

The radar does not represent isolated flavors, but the overall balance of the simulated extraction. Each axis reflects how the extraction index and the modifiers for roast, process and concentration combine in the specific Cold Brew you have configured.

The Cold Brew radar is the simplest and most different of all methods. The Cold Brew sensory projection model is more simplified than for other methods:

• Acidity: fixed value by roast. Light = 30, Medium = 15, Dark = 8. Does not directly depend on the extraction index. This reflects that Cold Brew extracts very little volatile acidity regardless of other parameters.
• Sweetness: ascending function of extraction (45 + extraction × 0.4), with a +10 bonus for natural process. Increases with more extraction up to a ceiling.
• Bitterness: very gently ascending function (extraction × 0.2 − 5). Cold Brew produces far less bitterness than hot methods even at high extractions.
• Astringency: very gentle function (extraction × 0.1). Almost imperceptible in the normal range.
• Body: function of ratio (40 + ratioN × 20). Body in Cold Brew depends primarily on ratio, not extraction.

This simplification reflects a real property of Cold Brew: the sensory profile is much more dominated by concentration (ratio) than by extraction dynamics.

---

Typical combinations and common readings

| Scenario | Symptom | Why it happens | Suggested adjustment | |---|---|---|---| | Acidic, watery flavor | Under-extracted | Short time or coarse grind → low index; acidity in cold brew is fixed by roast | More steep time or finer grind | | Excessive bitterness | Over-extracted | Very long time + fine grind → high index → bitterness starts appearing | Less time or coarser grind | | Very intense concentrate | Very low ratio | Low ratio = less dilution + higher extraction index | Dilute before serving or raise ratio | | Sweet, smooth profile | Natural process + ratio 1:6 | Natural adds +10 to sweetness; standard ratio gives balanced body | Target zone | | Silky body | Low ratio + balanced steep time | In cold brew the engine ties body directly to ratio, not extraction | Cold Brew standard combination |

---

Balanced, under-extracted and over-extracted in Cold Brew

Cold Brew thresholds differ from other methods, reflecting that cold extraction operates in different ranges:

Under-extracted: extraction below 35. The cup is acidic and very light, without the characteristic sweetness of Cold Brew. In practice, this occurs with combinations of short steep time, coarse grind and high ratio.

Balanced: extraction 35–50 for light roast, 35–60 for medium, 35–65 for dark. The upper threshold varies with roast because darker coffees in cold water tolerate more extraction before becoming unbalanced due to the higher solubility of their bitter compounds. A balanced Cold Brew has gentle sweetness, minimal acidity and silky body.

Over-extracted: extraction exceeds the threshold. Bitterness and astringency begin to appear, although in Cold Brew this effect is less dramatic than in hot methods.

---

Technical section: how the simulator models Cold Brew

General heuristic

extraction = 5 + grindN×25 + ratioN×20 + timeN×18 + tempN×8 + waterN×5

Where:

• ratioN = (ratio − 4) / 6 (range 4–10)
• timeN = (totalTimeH − 8) / 16 (range 8–24 h)
• tempN = clamp((fridgeTempC − 2) / 6, 0, 1) (range 2–8)
• waterN = min(waterGH / 12, 1)

The starting point is only 5 (the lowest of all methods), reflecting that cold extraction without any additional factor produces very low extractions. Time is the highest-weight driver (18 points), followed by ratio (20) and grind (25). Fridge temperature carries significant weight (8 points) within the modeled 6 °C range.

Extraction states

Cold Brew thresholds use a different system from other methods: instead of a fixed lower threshold (low) and upper threshold (high), the upper threshold varies with roast:

if extraction < 35 → Under-extracted
threshold = 50 (light) / 60 (medium) / 65 (dark)
if extraction < threshold → Balanced
else → Over-extracted

This asymmetry reflects that darker roasts in cold water produce profiles more resistant to over-extraction due to the higher cold-water solubility of their bitter compounds.

Simplified sensory projection

The Cold Brew sensory projection is deliberately more simplified than the other methods. The model does not apply the same temperature, pressure or time modulators to the radar because in cold extraction these effects are much less pronounced and the profile is dominated by concentration variables:

• Acidity is fixed by roast: Cold Brew does not develop volatile acidity over time (it would at room temperature, but the model only covers refrigerator steeping).
• Sweetness grows with extraction but has a ceiling.
• Body depends on ratio, not extraction.

This simplification is a design decision of the model, not a technical limitation: adding extraction modulators to the Cold Brew radar would add complexity without reflecting perceptible differences in the modeled parameter range.

Simplifications and limits

• The model covers refrigerator steeping (2–8 °C). It does not cover room-temperature Cold Brew (~20 °C), which is much faster.
• The time range is 8–24 hours. Shorter steeps (4–6 h) or longer ones (36–48 h) are not modeled.
• Post-steep dilution is not modeled. The calculated profile corresponds to the concentrate before diluting.
• Water hardness only affects extraction (through GH), not the Cold Brew sensory radar, unlike hot methods.

The model prioritizes interpretability over physical precision: the goal is for the user to understand how to adjust variables and develop cold extraction intuition, rather than exactly reproducing the behavior of a real Cold Brew.