How we calculate
The formulas, where each number comes from, and what is only an estimate.
What this site cannot promise
- The calculator's times are a model we tuned to land near published recipes. They have not yet been validated against measured brews. Your grinder, filter and pouring will change them.
- A time is never the goal. If the cup tastes good, the recipe worked, whatever the clock says.
- We could not read the SCA's own Golden Cup document. Its TDS and extraction ranges are quoted from a peer-reviewed paper; its ratio and temperature come from secondary sources, listed below.
- We found no peer-reviewed data on pour-over timing, so the times stay an estimate even where the strength is sourced.
How to read the labels
Sourced Taken from the page linked next to it.
Common practice Widely used advice, with no single source we can point to.
Our estimate A value we chose. Treat it as a starting point.
Amounts
This is plain arithmetic, so it is exact. 1 ml of water is treated as 1 g.
water = coffee × ratio
coffee = water ÷ ratio
ice = total water × ice share
hot water = total water − ice
The last pour takes whatever remains, so the pours always add up to the total.
Strength and extraction
Sourced From two peer-reviewed UC Davis papers. The site shows these as expectations; it does not measure your cup.
in the cup = water − ( × coffee)
drip: TDS = extraction × coffee ÷ in the cup
immersion ceiling: TDS = ÷ (ratio + )
The Golden Cup range is % TDS and % extraction (Batali et al., 2020). So 1:18 falls inside it, while 1:15 and 1:16 are stronger. That is not a fault: the same research group describes the chart behind this range as outdated.
- Water held by the grounds, and the immersion ceiling, come from Liang et al., 2021. The retention was measured on immersion brews; using it for drip is our approximation.
- In that study immersion took about 20 minutes to level off, so a 2-minute AeroPress stays below the ceiling. We cannot say by how much.
- The ceiling formula matched the study's measurements at 1:5 (4.12% against 4.07%) and 1:15 (1.41% against 1.36%), but overshot at 1:25 (0.85% against 0.75%).
- Between 87 and 93°C, brew temperature had no appreciable effect on taste at fixed strength and extraction (Batali et al.), and immersion extraction did not change from 80 to 99°C (Liang et al.). Our temperature advice is therefore common practice about extraction speed, not about flavor.
Calculator times
Our estimate Every number in this section.
scale = √(hot water ÷ 300), between 1 and 2
pour rate = g/s × scale
drain rate = V60 or Chemex g/s × scale × roast
roast = light , medium 1, dark
wait = (pour ÷ drain rate) − pouring time, at least V60 s or Chemex s
drawdown = same formula, at least s
The total is shown as a range of ±%. A long-brew warning appears above on V60 and on Chemex.
Worked example
V60, 20 g, 1:16, 3 pours, medium roast. This table is calculated live by the same code the calculator uses.
Where the numbers come from
| Value | Basis |
|---|---|
| Golden Cup ratio: about 55 g/L (≈ 1:18) | Sourced Secondary: Gusto's Coffee |
| Golden Cup temperature: 93°C ± 3° | Sourced Secondary: Klatch Coffee |
| AeroPress Original 10 oz (≈ 295 ml) | Sourced AeroPress |
| AeroPress Go 8 oz (≈ 237 ml) | Sourced AeroPress |
| AeroPress XL 20 oz (≈ 600 ml) | Sourced AeroPress help center |
| Space taken by grounds: ml per g | Our estimate |
| Bloom: V60 × coffee, Chemex × coffee | Common practice |
| Grounds hold about × their weight | Sourced Liang et al., 2021 |
| Smallest allowed pour: g | Our estimate |
| Most coffee per brew: V60 g, Chemex g, AeroPress Original g, Go g, XL g. Makers publish no gram limits, so we estimated these from the largest sizes. | Our estimate |
| Bloom wait by roast (light / medium / dark): V60 s, Chemex s | Our estimate |
| AeroPress pour: g/s, at least s. Press: 30 s | Our estimate |
| Size warning above V60 ml, Chemex ml | Our estimate |
| Temperature and grind by roast, Chemex and AeroPress temperatures | Common practice |
| Starting-point suggestions, AeroPress steep times | Our estimate |
| Ice share options | Our estimate |
| "How did it taste?" advice | Common practice |
Published recipes
These follow the linked page at the published dose. At other doses the amounts scale and the times stay as published, which is our choice and not the author's.
- Tetsu Kasuya, 4:6 Method (Hario)
- James Hoffmann, Ultimate V60 (Hario USA). The first pour's length is not stated; we use g/s.
- Charlene de Buysere, World AeroPress Championship 2012 (AeroPress). Only the ~30 s bloom is stated; the pour and press times are our estimate.
Your data
There are no accounts and no tracking code. Saved recipes, notes and your recent brews stay in this browser and are never uploaded. A share link contains the recipe settings but not your notes. Fonts load from Google Fonts and the image export library from Cloudflare.
Found a mistake?
Tell us on X. Corrections are listed below.
Changes
v, 31 October 2026
- RatioCraft is now called Sabba (صبّة), with a new logo. Your saved recipes are kept.
- Fixed: new app icons so the site can be installed on Android.
v2.3.0, 31 October 2026
- New: a limit on the coffee each tool can brew, with a clear error above it.
v2.2.0, 31 October 2026
- New: dark mode, and your last 3 brews one tap away.
- Changed: a simpler design, with the most used settings first.
v2.1.0, 31 October 2026
- New: expected amount and strength in the cup, and calculating the coffee from the amount you want, based on peer-reviewed research.
- Changed: temperature advice no longer claims effects on flavor that the research does not support.
v2.0.0, 31 October 2026
- Fixed: the brew timer did not start; changing the coffee changed the ratio; impossible recipes showed negative amounts.
- Fixed: recommendations were wrongly attributed to the SCA; 1:16.6 was labelled the Golden Cup ratio.
- Changed: calculator times now depend on the amount of water and are shown as a range.
- New: published recipes with sources, saved recipes, share links, taste advice, ice share, and this page.