Missing the point

From time to time we here at VST receive some critical comments. Individuals can become a little defensive about how they brew their coffee, and not wanting to feel like a piece of software, a refractometer or a set of standards from 1950s is telling them how to brew. If this is the argument we are having, however, we are missing the point. Our aim is not to tell you how to brew, but to give you control over the process, and perhaps help to point you in the right direction.

Both ExtractMoJo and MoJoToGo allow the user to set the target extraction yield as a preference wherever you like, and also allows you to define your own ROI range limits.

I’ve tried to locate the original research from the Midwest Research Institute (MRI) who created the region of interest (ROI) adopted by the CBC in the 1960’s.  The CBC used it to replace the earlier ROI created by the National Coffee Association. The MRI ROI survives to this day as the SCAA “Region of Optimum Balance”, and was the starting basis for the SCAE and Norwegian Coffee Association’s ROIs, all pictured on the same chart, below (Fig 1). The archivist at MRI was unable to locate the original research criteria, and I have been unable to uncover it anywhere else, but fortunately, the SCAE has embarked on an effort to re-verify this ROI, and the SCAA and NCA have started an R&D effort to define an ROI for espresso.

Figure 1

The essence of the importance and usefulness of the brewing control chart is lost in many of the criticisms. I can’t tell you the number of specialty cafes in the U.S. and around the world where I have been served coffee that was undrinkable, not even on the page, literally off the chart, i.e., 0.90% @ 14% yield. I haven’t met anyone yet that thought this tasted acceptable. A simple exercise of measurement and plotting adjustments on the chart produces measurable improvements that patrons are thrilled about when they taste sweeter, richer coffee that blooms in the cup after an adjustment is made to move the extraction yield to somewhere near or w/in the region of interest. When you’re off the page, the coffee is usually unacceptable, when you’re in or near or in the region of interest, the coffee nearly always tastes better. Where it tastes best is entirely up to you. How you brew is up to you, but the basic science behind the chart is sound, measurable and repeatable, which is the fundamental point of the tool(s).

The chart assumes proper technique i.e., uniform extraction. Why would anyone encourage uneven extraction (i.e., over-extracting from the center, and leaving dry powder at the perimeter)? Doing so wastes large amounts of coffee (over time), and produces widely inconsistent results, cup-to-cup. There are a number of formulas out there that seem to indicate this is common practice, even encouraged by some, but it does not produce the best nor the most consistent results, IMO. Your opinion may be different, that’s fine. I also don’t share up-dosed brews to raise the concentration, which nearly always produces a stronger, but still under-developed finished result. This navigates up the brewing control chart, rather than up the brew formula line. If uneven extraction methods were used in cupping, professional cuppers would never be able to compare and evaluate results accurately. We can learn from this example. If you up-dose, and extract unevenly, you might get numbers off the chart that could taste okay, but it’s likely they won’t be consistent, and is generally an anomaly. Once you land on a preferred formula, the general idea is to repeat with precision in order to achieve similar results; that’s usually what we’re after.  One simply cannot do that if the ground coffee is unevenly extracted from batch to batch, regardless of the size or technique used to brew.

Even extraction and relatively uniform grind help maintain consistency. Accurate dose of water and coffee weight and technique (contact time, temp, turbulence, wetting) contribute consistency to the final result. Both the SCAA and SCAE offer Gold Cup programs to teach these methods, and they’re invaluable.

A serious problem in the past is that accurate measurements had not been easy to make, because the apparatus with the requisite accuracy and resolution is difficult and expensive to build. Experienced baristas and café owners cannot predict how a particular coffee will extract. The same coffee roasted slightly differently from week to week will extract differently. Green coffees change as they age, roast behavior fluctuates and extraction yields move around with these changes, and must be constantly monitored. Having a way to measure this to keep the targeted area in check is a great way to maintain coffee sweetness and brew quality. There is tremendous room for improvement in commercial and home brewing and grinder equipment. Brewers do not dose water with as much precision as we’d like (time-based rather than measurement bases), grinders are even worse (same reason).  Consider that a -5% error in the weight of coffee along with an error of +5% too much water produces a cumulative Brew Formula error of 10%.  That’s enough to take you from sweet @ 19% to bitter @ 22% Ext Yield.  Precision is important, even more so with smaller batches, especially so w/single-serve.

The brewing control charts used by specialty coffee organizations around the world are all based on the American standard developed in the 1950’s. One of the most important aspects of the charts are the taste defect notes in the regions outside of the ROI in question, i.e., to the left of 18% Extraction Yield and to the right of 22% Ext Yield (Fig 1).

Consider the brewing control chart as a map – for brewing coffee/espresso. When you go to a foreign city, how do you get around? Usually with a Map or perhaps a GPS.  What’s the first thing you do with either one? Determine where you are. Why? So that you can navigate to where you want to go. This is exactly what the brewing control chart is for, and how it can be used to improve the quality of your brewed coffee. If you’re not brewing to extract in a range at or near the so-called gold cup standard, like it or not, your coffee is probably not as good as it could be. This is really the key essence of the tool. Use it wisely, and you will improve and maintain higher quality coffee, whatever your purpose.

Note that each specialty coffee organization does not share the same range of concentration (i.e., strength a.k.a. % TDS).  What they all do share in common, on the other hand, is that coffee extracted to much less than 18% is considered [by most people] to be under-developed (a taste defect that tends to sour as concentrations are increased, as in espresso) and they also agree that coffee extracted to much greater than 22% is considered [by most people] to be over-extracted (a taste defect that tends to bitter as concentrations are increased).  Also note that one can have weaker or stronger concentrations of sour or bitter, these taste defects first identified in the 1950’s are still relevant today.

Strength, on the other hand is purely a personal preference.  Take a look at the brewing control charts with ROIs  the SCAA, SCAE, NCA (Fig 1):

                                                Min      Max     Min      Max
                                                Ext      Ext      TDS     TDS
Norwegian Coffee Association                    18        22        1.30     1.55
Specialty Coffee Association of Europe          18        22        1.20     1.45
Specialty Coffee Association of America         18        22        1.15     1.35

While each has a particular preference for the range of preferred strength, they all agree to a common range of extraction, not surprising since no one wants a sour or bitter cup of coffee, while most don’t mind slightly stronger or weaker strength–as long as it’s sweet.

Bitter (over-extracted) taste defects are the primary reasons people add sugar or milk/cream to their coffee beverages. Milk contains lactose, molecularly, similar to sucrose (both are C12H22O11, with different solubility’s in water), IOW, milk is a sweetener. If we brewed our coffee to about 19% extraction yield, it would be naturally sweet (assuming ripe cherries ended up in the final green). Under-extraction does not fully develop the sugars and desirable soluble flavor components. Over-extraction [measurably] lowers the pH as more acidic and bitter components are over-extracted and overpower the natural sugars (sweetness).

Has this chart moved, no but the effective ROI has, but not due to a change in where most people perceive the sweet spot. Rather, it’s due to a small error introduced when commercial drip brewers began to dominate the brewing process. The original CBI/CBC brewing control charts never took into consideration the density of water as a function of temperature. When you build in this correction, the range shifts about a full percent to the LEFT.

When the brewing control chart was first created under the direction of Earnest Lockhart in the 50’s at the CBI/CBC, water amounts were measured in volume and calculations done at ambient temperature. Water measured in units of volume weighs more at ambient than when measured at brewing temps such as in today’s tank-style brewers, so in actuality, the sweet spot is really closer to 19%, and the left edge of the start of acceptability is closer to 17%.  So if corrected for this shift, the charts should really be showing the ROI (Region of Optimum Balance) in the range of 17-21, which many skilled cuppers I know have noted by taste, i.e., that they find 17% acceptable even though it’s outside the 18-22% range (because 17 is the “old” 18), and the most skilled cuppers I know all think 19% is the center of the sweet spot.  Further, 22% is slightly bitter to most people.  The ExtractMoJo and MoJoToO software takes into account the density of water as a function of temperature, and keeps the brew formula constant regardless of starting water temp, so, for example, if water is measure at ambient, you’ll note that there is about 4% more coffee required than if measured in volume at brewing temp. In effect, this means the actual ROI is really 17-21%.

Both ExtractMoJo and MoJoToGo allow the user to set the target extraction yield as a preference wherever you like, and both also allow you to define your own ROI range limits. MoJoToGo even alerts you if you fall below the min or above the max limit you’ve selected, whatever it is. So the software tools are there for you to do as you please, not, as implied in much of the criticism, forcing a fixed 18-22% ROI that dates to the 50’s on the printed version of the charts, although with the caveat stated above, I believe 17-21% (a.k.a. 18-22, circa 1952) is indeed the sweet spot for nearly all coffees, regardless of brewing method.

Brewing method has nothing to do with the chemistry, assuming the coffee was brewed uniformly, and with good practices. Different brewing methods do, however, impart different mouth feel, clarity/sediment, lipids and oils, but that’s not the point of this post. I think you’ll find that when accurately measured, very few people skilled in the art would find a 15% extraction yield nor a 23% one acceptable with any coffee nor method. On the other hand, nearly all would be happy with a good coffee served in the range of 17-21%, and I’d lay odds that they would find the sweetest spot at about 19% Ext Yield for most coffees, regardless of roast degree, region, terroir, etc….

I originally intended to sit on this until the SCAE completed their Region of Interest research (out of curiosity to see if they agreed in the end, w/o any bias), but since it keeps coming up, el gato está fuera de la bolsa. This is documented in several presentations I have given, most recently at the SCAA Symposium in Anaheim this past month. You’ll notice that when ExtractMoJo was first introduced, there was a region called GHCC.  George Howell, and his daughter Jennifer at Terroir, along with Ben Kaminsky at Ritual are three of the top cuppers in the world I have had the pleasure of working with. Jennifer placed a cliff-hanger 2nd to Ben in this years U.S. Cup Tasting Championship in Anaheim, and Ben took 1st for a second year running. The competition couldn’t have been closer. Dave Walsh is also a skilled cupper, and winner of the Cup Tasting Championship in Ireland this year. While I am not even close to their skills as cuppers personally, over three years tasting and cupping hundreds of coffees with them, we all tend to agree that the sweetest coffee is almost always at 19% Ext Yield (at any strength). Interestingly, the cupping standard used today around the world (developed originally by George Howell), nine times out of ten measures between 19.0-19.7% Ext Yield and 1.25-1.30% TDS consistently over hundreds of measurements (Fig 2).

Figure 2

Hopefully this sheds some light on why that is, in other words, the work done in the 50’s is actually 17-21% when taking into account the density of water as a function of temp while keeping the brew formula constant and using modern brewers (Fig 3). Personally, after more than two full years and thousands of measurements at hundreds of cafes and coffee houses all over the world, as well as in the lab, my personal preference for maximum sweetness is also 19% so I use a target of 19 % with a range of  +/- 1%. I have found dozens of the world’s best cuppers, roasters, café operators and baristi nearly all agree on that number, which is further assurance that the gentlemen who conducted their science in the 50’s did their homework, and understood very well the chemistry of brewing coffee. They deserve a great deal of credit for providing a repeatable method of visually charting it.

Figure 3

I have had the pleasure of personally designing and setting up literally hundreds of cafes in an around Boston over the past three years, and can tell you that year one (prior to ExtractMoJo) was a extremely frustrating, whereas the past two years has offered a vast improvement to quality and repeatability, now that repeatable tools are available. Making coffee at home is relatively easier than running a café or coffee house. Consider a typical café serving just three coffees, each in four batch sizes (including ICED). That’s twelve (12) combinations that all need to be set up and calibrated so that the 2-Liter batch is extracted to the same TDS and Yield as the 4-L & 6-L batch, and the same as the single-serve pour-over. It’s a lot of work, and nearly impossible to manage correctly w/o the instrumentation and tools to help measure and set it up. Once set up, however, it’s far easier to maintain.

There were no regions of interest for espresso on the Universal Brewing Control Chart, until ExtractMoJo was released in April, 2009, but the same general range of 17-21% applies to espresso [IMO], again, with personal preferences for the concentration required for the particular drink being prepared.  A significant educational contribution I learned about espresso was that the taste defects developed in the ‘50’s are greatly amplified at the palate by the concentrations of  typical espresso, so under-developed becomes decidedly sour and over-extracted becomes very sharply bitter at these higher concentrations.

There is huge value in using the brewing control charts, and in educating the industry to the basic high school chemistry of brewing coffee. If you don’t believe in it, I suggest you actually try it in your daily work. I think you’ll be pleasantly surprised. The MoJoToGo software for the iPhone, iTouch and iPad is now only one-tenth the price it was just 60-days ago for the PC version, so it’s much more within reach for the rest of us. Yes, you still need a decent refractometer, but the software now includes a TDS calculator, so one (1) refractometer works for both Coffee and Espresso, yet a further reduction in cost.


Vince F


One Response to “Missing the point”

  1. […] This post was mentioned on Twitter by David Walsh and JacobOrr, VST inc. VST inc said: Blog post – Missing the point – http://vstapps.com/blog-2/mojotogo/missing-the-point/ […]