BH App Archive

Coffee Extraction

Quick start – brewing coffee involves many choices and trade-offs. Here we simulate many of the key brewing processes to see the influences of the various parameters.

Coffee extraction is a…

complex process and in our search for the perfect cup of coffee, some of the choices we can make are the technique, the temperature, the grind, the volume of water, the roast of the coffee and the time taken to do the extraction.

Here we let you make each of those choices so you can see what effect they have on the overall process using a set of simple but plausible models inspired by the many excellent complex models found in the coffee literature.

You have a chance to see everything that is going on. If at first the outputs are too complex, de-select the T & Flow and Extraction options, but here we’ll assume that you can see the full outputs, discussed in turn

  • Rate This is how fast the extraction is taking place. For a given T & Flow and grind it starts rapidly (many of the components are readily available on the outside of the grounds) and then decreases because there is less to extract and with an increase in concentration in the brew there is a reduced gradient driving the extraction.
  • Strength The more water that flows through the system, the more the coffee is extracted. But, of course, you are also diluting the coffee. So there is a trade-off. In some cases it is simple – you just get a more dilute coffee. In others there is an optimum of extraction and strength.
  • Temperature and “Flow” The temperature varies during your process and the effect of temperature is assumed to double or halve the rate of extraction as you change up to 100C or down to 80C. “Flow” describes the mixing and turbulence required to keep the extraction process going. For each technique the exact meaning varies but the general trends make sense. The funny oscillations in Pour-over indicate the regular top-ups from the kettle. If you click the 1 option then it’s just a single slow pour, i.e. a filter coffee machine.
  • Extracted The amount that is extracted is different from, but connected to, the strength. It is well known that some components extract faster than others, so the total is divided into two components. What should we call them? Baristas like to call the fast-extracting components “acidic” (a term of praise, different from “acid” which is not) and the slower ones “richer” or “bitter”. A lot of what we think of as a “fruity” taste is from fruit acids, so the faster components might be named “fruity”. Because to many people “acidic” means the same as “acid”, we chose “fruity”.

Now you understand the outputs, what about the inputs?

  • Temperature As mentioned above, the extraction rate is assumed to vary by a factor of two for 10deg above and below 90C. In most processes the temperature will fall over time so the rate of extraction will also fall.
  • Grind The assumption is made that at Grind=3 the rate is “normal” and increases by a factor of 2 for Grind=1 and is halved when Grind=5. In some cases a fine grind can be a catastrophe as it can either open up large channels through which the water passes too quickly or blocks a filter so you can’t complete the brew. Such effects are not included.
  • Time We always have the worry about under-extracting, balanced by the fear of over-extracting. The setting changes the relative times by a “reasonable” amount and the absolute times for each brewing technique are chosen to be “normal” for Time=3. If you happen to disagree with our definition of “normal” feel free to imagine a different time scale.
  • Volume As discussed above, a large volume of water gives you good extraction but a low strength brew, and a small volume extracts less with a stronger brew. The absolute volumes are not specified, the trends are simply indicative of the common trade-offs.
  • Roast This affects the ratio of fruity to acid. Arguably a higher roast will lead to a larger extraction but this slider is more focussed on the fruity/bitter trade-off we all have to make.

What about xxx technique? Yes, we could have added Moka, or Turkish or Syphon or … you name it. But we decided that there are few new principles and Moka is so variable and Turkish so unique (the coffee is ground fine in a very different way) and Syphon so fast and … that it wasn’t worth adding further complication. But we’re open to persuasion!

Good coffee and great coffee. A lot of work and thought went into creating this app, and a lot of the specialist academic literature was consulted. What was surprising at first is that it seems remarkably easy to get a good coffee, with plenty of latitude in the settings. This is because a large part of the extraction takes place very quickly and the processes are naturally rate limiting. As long as you have some fresh, good-quality beans (with the taste balance you like) and don’t go much above 90C

(when you start to destroy the ingredients and extract real nasties) and have a reasonable grind and a reasonable time, you’ll get a good cup of coffee. That’s what the science says and that’s what experience shows. So how about a great cup of coffee? Well, the science suggests that there isn’t a short cut. All those little details that the great baristas have worked out over the years somehow come together to get that extra “something” which transforms a good cup of coffee into a great one.

In one way, this is a bit of a disappointment. Surely we just want to find the perfect science for a great cup of coffee time after time. But in another way it’s good news. It shows that a great cup of coffee needs dedication, training and experience. The science has certainly progressed over time and it’s far more likely that we’ll get a good cup of coffee. But for a great cup we either have to be a great barrista or pay for the privilege of drinking coffee prepared by one.

Brewing Control Charts. For many decades, the industry has made good use of Brewing Control Charts, filled with masses of lines and various dashes and dots to illustrate features of the complex brewing space. We wondered if we should produce one ourselves (or at least reproduce a famous chart such as Lockhart’s). But this app gives nearly all the information that a Brewing Control Chart can provide, while revealing what’s going on behind the scenes. And it tells you the specifics of your chosen settings rather than a lot of lines describing other settings.

What this app does not give is absolute values of extraction % or %TDS (Total Dissolved Solids). Although it would not be hard to add extra inputs to make this happen, in practice the benefits are modest for a significant increase in complexity. But we’re open to persuasion!

Copyright © 2019 Barista Hustle, All Rights Reserved!