Roasted Coffee Color Measurement & Classification

“No one has ever presented a technique for the measurement of roasted coffee color that could be considered for adoption as a standard for use throughout the coffee industry.

No one has ever suggested a standard list of names for different colors of roast. Nor has anyone tried to associate a useful list of names with color values derived from a modern method of measurement.” E.E Lockhart, May 25, 1960, San Francisco, California. *

In the late 1950s a team of researchers from the Department of Food Science and Technology at the University of California in Berkeley published a series of papers on a scientific approach to the color of coffee, its measurement and its relationship to flavor. Four of these papers were published in the Journal of Food Technology and presentations were given at the annual meetings of the Institute of Food Technologists in Chicago, Philadelphia and San Francisco.

At about the same time in 1958, Dr. E.E. Lockhart, with the Coffee Brewing Institute in New York was writing a publication entitled, “The Analysis of Coffee Grinds.” Dr. Lockhart collaborated with the Berkeley scientists and closely followed their results. In 1960 he was citing their work and in 1967 he reprinted their research papers as Publication Numbers 17, 37, 48 and 51 of the Coffee Brewing Center. In March of 1960, one of the Berkeley scientists, Angela Little, contributed an article to The Tea and Coffee Trade Journal entitled “Collaborative Study of the Measurement of Color of Ground Coffee” thus bringing the subject to the attention of coffee industry as a whole. This new research contributed an important milestone in coffee science.

Coincidently, in 1960, Gordon Gould and Thomas Maiman independently invented the first Lasers. It would take almost fifty years to pass, before a laser would be applied to reading the color of roasted coffee but we will get back to that subject later in this article.

For the one thousand years that people have been roasting coffee, they have relied on a variety of sensory clues to determine when to end and stop the roasting process. Consistency and quality are determined by accurately stopping the roast at exactly the right time.

From the earliest days of coffee roasting we determined coffee roast development by eye, noting the visual assessment of color and the swelling of the coffee beans. There is a cracking sound that roasted coffee gives off that also is sometimes used as an audible clue to roast development. Aroma development signals the point of pyrolysis and the point of an endothermic to exothermic process transformation. At a certain point in a dark roast, oil formation on the bean surface and the subsequent smoke emissions can also alert us to the point of terminating a roast.

Often a number of these sensory inputs are used in tandem or simultaneously to monitor the roasting process. These visual references and other sensory clues have been used for the last millennia since Bedouin began roasting in small open pans during encampments. All around the world people have roasted by eye whether in clay bowls or stove top pans.

Most modern sample roasters still encourage us to roast by eye as we pull trier sample after sample for color scrutiny. Likewise most roasting machines, even those totally controlled by computer, still have view ports and a sampling trier as standard issue.

It is still commonplace for a roastmaster to frantically pull samples as he/she nears the end point of the roast. The color changes are much more rapid as the roast progresses toward ebony. There are obvious pitfalls in this subjective method. Individual differences in color perception, ambient lighting, operator fatigue as well as the interpretation of the descriptive terms used for color such as light, medium, and medium dark, French etc.

There is a class of “artisan roasters” who swear by the two audible crack or popping sounds coffee makes as the beans reach a point of stress responding to internal gas pressure and secondly, the point of cellular structure degradation due to charring within the beans. These points are enthusiastically referred to as the “first crack” and “second crack” and are commonly used to monitor and determine when to end a roast.

Aroma development is another clue to roast development progress, especially for those who roast on commercial sample roasters or roast outdoors. The smell of coffee changes considerably as chlorophyll is degraded, sugars caramelize and oils vaporize.

Our dark roast aficionados may also rely on the appearance of surface oils and the blue smoke that rises out of the pyre before they’re ready to call it quits.

All these points of reference are helpful in the attempt to craft consistency where instrumentation is not available. When commercial roasting machine manufacturers began adding thermometers to their effluent gas pipe, mainly to measure pre-heating temperatures, it was an obvious next step to put a thermometer through the faceplate of the roaster to measure the roasting bean’s temperature directly.

Today most roasters have thermocouples in the bean mass (bean temperature probes), which allows monitoring of the roast progress and thus giving us notification as to when the end point temperature is reached. Larger roasting machines will automatically water quench at the end-point temperature and smaller machines without quench, may automatically turn on the cooling tray fan, start the sweep arm and by use of servo control, open the drum door to drop the roasted coffee into the cooling tray. Temperature control is good.

For the professionals and sophisticates there are more complex programs being used such as stepped roasting, ramp roasting and profile roasting. The latter requires computer control of the input roast gasses relative to the bean probe temperature reading.

Computer controlled roasting machines generally arrest the roast at a designated end point temperature that is monitored by either thermocouples or infrared sensors. Water quenching arrests a roast immediately. Without water quench the coffee will still continue to roast somewhat until the cooling tray airflow has had sufficient time to have an effect on the internal bean temperature.

Regardless the drop temperature, Color has always been the gauge of roast development. We have a litany of roast terms from “Cinnamon” for the lightest roast to “French and Italian” to describe our darkest roasts. The following chart is just an example. End-point bean temperatures may range from around 380oF for light cinnamon to about 480oF at the darkest brown color. Many Roasters will supply their own names to the roast descriptors listed here. There is no standard for the center column of names.

Roast Descriptors                   Common Names

    • Very Light                 Light Cinnamon

    • Light Brown              Cinnamon

    • Moderately Light       Scandinavia

      Light Roasts ( Cinnamon, Half City, Light, New England) The bean is light brown and dry with no visible oils on the surface of the bean. The flavor is baked or bread-like, similar to toasted grain. Depending on the coffee, some sour tones may be present and the body of the coffee will be minimal. New England roast ( not very common) and Scandinavian are a little darker than Cinnamon roast.

  • Light Medium              American Traditional “First Crack”

  • Medium                       City Roast

    Medium Roasts (American, Breakfast, Brown, City, Medium) medium light brown beans. The American roast is the most common roast used for cupping and professional coffee tasting. An official Medium or City roast (more common in the Western U.S.) is slightly darker than American (more common in the Eastern and southern U.S.), and is an excellent choice for tasting the differences between most varietals.

  • Medium Dark              Full City                 “Second Crack”

    Medium-Dark to Dark Roasts (Full City, Light French, Viennese)  Medium dark brown beans. Some oily drops will be present on the surface of the bean when roasted Full City. Full City roasted coffee will exhibit some chocolate or caramel undertones. Light French or Viennese is ever so slightly darker than Full City.

  • Moderately Dark        European

  • Dark                           French Roast

    Dark/High Roasts (After Dinner, Continental, European, French, New Orleans and Espresso) At this level, the beans are very dark brown (but not black). French roasted beans are shiny with an oil coating on the sirface. Some burned undertones will be present and the acidity is lower. This is a popular roast for making espresso, though lighter roasts are also used and are becoming more popular. Many people believe French roast to be the darkest roast, but they have not had Spanish Roast yet…

  • Very Dark                   Dark French

  • Darkest                       Neapolitan

    Very Dark Roasts (Dark French, Italian, Spanish or Neapolitan) Spanish is the darkest roast of all (Dark French & Italian is a little lighter). The coffee beans are nearly black in color and the flavor tends to be flat with charcoal undertones. Pretty much all of the sugar in the beans has been caramelized and toasted thoroughly. Most Americans would turn their noses up to coffee this overcooked but, to each his own.

Color is what most consumers use to make a decision when purchasing whole bean roasted coffee. Telling a customer that his coffee was roasted to 465 degrees F. means nothing to the average consumer. Referring to it as a Full City roast is more common, yet just as vague. Eventually the consumer makes the connection with local roast names and their favorite flavor characteristics, particularly if they grind their own coffee at home. They begin to appreciate the dry aroma or “bouquet” of the ground coffee, its color and its relationship to taste.

Color is the roasters basic descriptor of his coffees. In order to monitor color and scientifically define color, a number of roast analyzers have been used by the industry over the last fifty years, Electronic measurement of roast color is best accomplished by instruments that take reflectance measurements.

In the 1960s, Dr. Lockhart et al. experimented with half a dozen devices and settled on the Photo-Volt brand reflectance meter as the most user-friendly for coffee roasters. Also included in their overview of the study is a recommended eleven point descriptive scale derived for roast color values from extremely light, medium to extremely dark.

Brown colors are very difficult to accurately discern with the unaided eye. There are, in fact, several pitfalls in this subjective method. Poor lighting, coupled with the bean’s matt finish contributes to the problem. To complicate the matter more, color perception is unique to the individual; we all do not have the same recognition of colors, particularly those so subtle as shades of brown. In addition there is variety in interpretation of the descriptive terms such as light, medium, full city, French, and not to forget, operator fatigue.

In the last several decades we have seen many color reading devices introduced to the food industry, principally for quality control at the postproduction process. Today, even paint stores and interior decorators use them.

In recent years a number of these photo-spectrometers have come into common use for coffee color analysis. Surprisingly the laser, which was invented in 1960, where our story began, has only recently been applied to read roasting and roasted coffee color. With such an instrument affixed to the front of a roaster it is possible to read and record the coffee color throughout the roasting process.


Finally we have caught up to the sixties.


“Roasted Coffee, Color Measurement and Classification”

Presented at the Twentieth Annual Meeting of the Institute of Food Technologists,

By Dr. Ernest E Lockhart,

The Coffee Brewing Institute, Inc.

New York 5, N.Y

Robert Barker