19.1 Is The New 20.0

Cross posted from my Google+

The prior literature on coffee brewing tends to use mass units for coffee (grams or ounces), and volume for water (liters or fluid ounces, sometimes gallons or cups). Granted, you’ll see teaspoons or tablespoons used sometimes, but none of those are really trying to be scientific.

Lavoisier’s Law of the Conservation of Mass teaches us that mass is a constant. Volume depends on density. If density is a constant, then you can effectively treat volume as a constant in that particular case. In the case of coffee brewing, the density of water is not a constant. Water density decreases at higher temperatures. I have this particular web page bookmarked for when I need to calculate water density at a particular temperature: http://antoine.frostburg.edu/chem/senese/javascript/water-density.html

So when you say “I’m brewing coffee with one liter of water,” if you want to be precise and/or want to use this data to do some coffee brewing math, you need to know what temperature that water is. At room temperature, let’s say 20°C (68°F), one liter is 998.2 grams per milliliter. At 93.3°C (200°F), it’s 963.1 grams. The density decreased, and a given mass of water will expand in volume as it’s heated. This is true, and undisputed.

This is a fact that Vince Fedele has pointed out to the world by integrating it into the ExtractMojo (and MojoToGo) software. Both pieces of software, therefore, uses mass for water instead of volume. If you plug in a volume measurement, it will use its own temperature-density calculator to convert it to mass, before it does its calculations. This a great thing!

So what’s the problem? The problem is, with new units, you have to adjust the chart.

Everyone is still using charts that all read 18-22% as the Gold Cup extraction yield zone. But the 18-22% zone was developed with calculations using volume, not mass, of water. Therefore if you change the units to mass of water, since there’s a density-based Δ (delta, or empirical change), you have to adjust the results of any calculations accordingly.

If using volume as your water number, the extraction yield zone of desirable taste characteristics “by the book” was 18.0-22.0%. Using mass and 93.3°C (200°F), the new corresponding zone is 17.2 to 21.1%. The “sweet spot,” if you’re trying to nail the middle point of that zone, is 19.1% extraction.

Therefore, 19.1 is the new 20.0.

5 Responses to 19.1 Is The New 20.0

  1. Interesting – though I don’t know enough about how the original experiments were done. If drying was used then surely the extraction would still be accurate?

    You could argue that your range is closer to many people’s preferences for even extraction with current grind distributions – I’ve certainly not enjoyed anything over 21% with unsieved grinds coming from the grinders I’ve used.

    • nickcho says:

      James, there is mention of CBI studies that were indeed done by drying, but not the sort of drying you would think. A volumetric sample of the beverage is dried to discover the solids content. Extraction yield is calculated as solids ÷ original coffee mass. Makes me wonder why they didn’t take before-and-after mass measurements of grounds, but the literature makes no mention of it either way.

  2. Ah – for some reason I had assumed they would have done the grounds because it is lo tech, easy (though slow and probably smells a bit) and would have been accurate enough if brewing reasonable sized batches.

  3. Tom Walsh says:

    Nick, I find this discussion interesting as well as confusing. Maybe you can help me out here. I believe that I understand your point about density of water lessening as temperature rises. However, as a home user as well as when I’m in the shop, when brewing a pour over, I “weigh” both the quantity of coffee in grams and quantity of water in ml. I understand that ml is essentially a measure of volume, but a scale cannot weigh volume, just mass (I think?). (As an aside, I assume that scale mfg.’s, when allowing for measurement of fl. oz., or ml, is assuming a standard temp (or density) of the fluid and hence is providing a measurement of mass). So…when we measure water qty on a scale aren’t we really measuring water mass at an assumed density (1.0?) and hence any extraction readings would pertain to the original Gold Cup standard?
    Thanks for any thoughts to help clear this up for me. Excellent post!

    • Nick Cho says:

      Tom, the answer is: it depends. The point is, many people are using the ExtractMojo or MojoToGo software to do their math for them. That software uses mass in its calculations. So your “assumed density” is irrelevant, it’s going to calculate the mass based on density at that temperature. Hope this helps!

Leave a Reply

Your email address will not be published. Required fields are marked *


You may use these HTML tags and attributes: <a href="" title=""> <abbr title=""> <acronym title=""> <b> <blockquote cite=""> <cite> <code> <del datetime=""> <em> <i> <q cite=""> <strike> <strong>