Book 2 / Part 4 — Physical Transformations: RoR, DTR & Profile Design

Heat and chemistry get the headlines — but the roast you taste is ultimately a physical event. This is the part of the bean’s story where structure, colour, rate and timing become the levers you actually pull.

By the time you reach Part 4 of The Roasting Bible, Book 2, you already understand the heat pathways and the chemistry inside the bean. Now comes the part that turns understanding into control. Drying, browning and development aren’t just chemical stages — each one is a physical transformation you can measure, predict and steer in real time.

Part 4 is the operational heart of Book 2. It moves from the question “what is happening inside the bean?” to the more practical one: “how do I shape it on purpose?” Glass transition, colour, Rate of Rise, Development Time Ratio and origin-specific profile design are the tools that separate roasters who guess from roasters who repeat. This post walks through that toolkit.

Chapter 14–16 · Structure

Glass transition: why beans expand instead of explode

Have you ever made hard candy? You heat sugar syrup until it flows, then watch it cool into something brittle and glassy. Warm it again and it softens before it ever melts. That softening — rigid to flexible without liquefying — is the glass transition, and it is exactly what happens inside a coffee bean as it roasts.

Below its glass transition temperature (Tg), the cell-wall matrix of cellulose, hemicellulose and pectin is locked rigid. Push gas pressure into a rigid bean and the walls eventually fail catastrophically — splits, uneven shapes, violent cracks. But once the bean crosses Tg into its “rubbery” state, those walls can stretch. The bean inflates gradually, like a balloon rather than a pressurised pipe. That controlled expansion is what creates the porous, spongy structure that later decides how water flows through your coffee.

The subtle part — the part most roasters miss — is that Tg is a moving target. As the bean dries from roughly 11% moisture toward 4%, its glass transition temperature climbs in real time, because water actually lowers Tg. Your bean temperature has to keep pace. Fall behind, and the bean can briefly re-enter a glassy state mid-roast, setting up cracking defects when it crosses back over under pressure. This is one reason confident early heat matters: you are racing a rising glass transition temperature.

Chapter 17 · Colour

Colour is the odometer, not the engine

Colour is the roaster’s most visible feedback — and the most misunderstood. The brown comes from two sources: melanoidins, the complex polymers built by Maillard reactions from around 140°C, and caramel polymers from sugar breakdown at 160–180°C. The Agtron scale puts a number on the result by measuring infrared reflectance, from 0 (black) to 100 (lightest), with whole-bean readings as the gold standard because grinding exposes the lighter interior.

But here is the lesson Book 2 keeps returning to: colour is not flavour. Two beans at Agtron 60 can taste completely different — one bright and nuanced, one flat and underbaked — because what matters is how you arrived at that colour. Think of colour like an odometer: it tells you how far the roast travelled, never how you drove. The gap between whole-bean and ground readings tells its own story too — 2–4 points means an even roast; 6–10 points or more means some beans ran ahead of others.

Chapter 18 · Rate of Rise

Rate of Rise: the accelerator pedal of the roast

If bean temperature is the speedometer, Rate of Rise (RoR) is the accelerator pedal — it tells you not where you are, but how fast you are getting there and whether you are speeding up or slowing down. The maths is trivial: degrees of change divided by time. The insight is anything but.

The pattern you want is a smoothly declining RoR — starting high, around 8–12°C/min as cold beans hit the hot drum, then easing down as the bean approaches thermal equilibrium. That shape naturally gives you a clean crack and well-developed flavour. A flat RoR that holds for several minutes means you are stalling, heading toward a baked, papery cup. A rising RoR after first crack is acceleration beyond your control, risking scorched, uneven results. And a sharp crash — cutting heat too hard near the crack — can pull the coffee before its interior is ready.

Once you can read these shapes, RoR becomes a diagnostic tool: every defect has a thermal fingerprint. Better still, you can project forward — final temperature ≈ current bean temp + (current RoR × minutes remaining) — and steer the roast in real time instead of crossing your fingers and hoping the colour lands.

Chapter 19 · Development Time Ratio

DTR: the single number that decides flavour

Development Time Ratio (DTR) is the bridge between what happens thermally and what you taste in the cup. The formula is simple — (development time ÷ total roast time) × 100, where development time is the clock from first crack to drop. If a 10-minute roast hits first crack at 8:00, that’s two minutes of development, or 20% DTR.

The flavour zones are remarkably consistent. Below 12% is underdeveloped — sour, grassy, sharp. The 12–18% band gives bright, origin-forward light roasts. The 18–22% zone is the broad-appeal sweet spot: caramel, toast and chocolate balanced against friendly acidity. Push to 22–25% for deeper, bittersweet espresso-leaning roasts, and beyond 25% you cross into ashy, flat overdevelopment where origin character disappears.

But — and this is what separates advanced roasters from script-followers — DTR is not absolute. A 20% DTR over twelve minutes is 2.4 minutes of slow, patient development; the same 20% over nine minutes is 1.8 minutes happening fast. They taste nothing alike. That is why the book insists you read the RoR shape through development, not just the percentage. You are not measuring time — you are measuring how the coffee opens.

Chapters 20–22 + Supplement · Profile Design

Designing the profile around the bean

Real profiling is not copying one template onto every coffee. Terroir, altitude, processing and harvest timing all shape a bean’s density, moisture and internal structure — so each coffee needs its own heat strategy. Book 2 gives origin-specific starting points, developed on the author’s Giesen drum roasters at Tasse: high-altitude washed Africans want a hot charge and a restrained 12–18% DTR to protect their florals; Brazilian naturals need a cooler, gentler landing to keep their sugars from scorching; delicate Geishas get the lightest development of all, pulled barely past first crack.

Those starting points feed into a systematic five-step methodology: read the green coffee’s physical story, run a deliberately conservative discovery roast for data, cup it against the full SCA protocol, then refine one variable at a time before locking a documented production profile. The discipline is the point. As the book puts it:

“Roasting stops being intuitive guessing and starts being a conversation with the bean — informed by data, driven by single-variable testing, grounded in honest cupping feedback.”

Change one thing, cup it, let the result guide the next single change. Most coffees lock in within three to five iterations. The reward isn’t a one-off great batch — it’s a profile your whole team can repeat, and a process you can diagnose when production drifts.

Who this part of the book is for

Part 4 is for the roaster who has the fundamentals down and now wants control — anyone who has tasted a flat batch and couldn’t explain why, or who wants to move from following someone else’s recipe to designing their own. If terms like RoR, DTR and Agtron are familiar but you’ve never connected them into a single repeatable workflow, this is the part of Book 2 that ties it all together.

Written by Wing Yuen, SCA Authorised Trainer and founder of Tasse Coffee Roastery in Shinjuku, Tokyo. The profiles and methodology in this series are drawn from The Roasting Bible and the daily roasting work at Tasse.