The percolator
In the percolator, water boils at the bottom of a pot, rises through a central tube and falls back onto a basket of coffee, which it passes through before going back down, and the cycle starts again until the desired color. It is the coffee maker of 20th century America, and one of the most criticized.
Why does the percolator overextract?
Start the scene or choose a stage. The animation simplifies the mechanism; what follows gives the details and the available references.
The pot is filled with water, the tube and the basket filled with ground coffee are set in place, and it goes on the heat. At the bottom of the pot, the water caught under the base plate starts to boil: steam pushes a column of water up the central tube.
Read the stages without animation
- Water boils under the plateThe pot is filled with water, the tube and the basket filled with ground coffee are set in place, and it goes on the heat. At the bottom of the pot, the water caught under the base plate starts to boil: steam pushes a column of water up the central tube.
- The jet falls backWater spurts from the top of the tube, falls on the lid and sprays the coffee basket. The coffee receives water that has just left a boil, the hottest of all methods.
- It passes through and runs back downThe liquid passes through the coffee, then the perforated bottom of the basket, and rejoins the pot. The pot no longer contains water but an ever more loaded coffee; strength is judged by color, often through a small glass dome on the lid.
- The cycle starts againIt is this coffee that rises again through the tube and goes back over the same grounds, as long as the heat is on. Heat and recirculation add up: extraction goes beyond measure. The chemist Charles Trigg added in 1922 that the liquid turns to vapor at the heating point, which carries off the aromas and spoils what is dissolved.
- Filtration, only onceNext to it, a filter passes clean water, only once, through the coffee. In 1913, the American roasters' association measured 2.90 grains of “coffee tannin” per cup for the percolator, 0.29 for the filtration of a pulverized grind: ten times less. This “tannin” is a period name for compounds now linked to chlorogenic acids and their roasting products.
To remember: The percolator sends an already loaded coffee back over the same grounds, at boiling temperature: two causes of overextraction, physical and not a matter of taste.
In history: Hanson Goodrich, of McLeansboro, Illinois, patents the American form in 1889, an accessory that fits into any coffee pot; the apparatus is first French, with Laurens in 1819.
Read the full article and its sources ↓History
The apparatus is first French. Ukers credits Laurens, in 1819, with the original of the device in which boiling water rises through a tube and spreads over the ground coffee; Gaudet perfected it in 1820 with a cloth filter, and by 1825 the pump percolator, driven by steam pressure and a partial vacuum, was widespread in France, Holland, Germany and Austria (source 2). In 1827, Gandais made it a truly practical model, with an outer tube, and Durant, in Châlons-sur-Marne, patented the first inner tube (source 2). In Vienna, in 1922, it is still called the Viennese coffee machine (source 4).
The American form is an accessory. On August 13, 1889, Hanson Goodrich, of McLeansboro, Illinois, obtained patent 408,707 for a device that fits into any coffee pot: a base plate, a telescoping tube adjustable to the water height, and at the top a perforated cup lined with a filtering cloth (source 1). Goodrich praises a continuous circulation, which heats less water at a time, and a coffee free of grounds that no longer needs clarifying agents (source 1), those eggs and shells that importers were still recommending in 1902 (source 3).
The name is misleading. For Ukers, true percolation lets the water run once through the coffee; the French drip is the “true percolator” (source 3). The pump percolator, for its part, sends an already loaded liquid back over the coffee. As early as 1913, the best coffee committee of the American roasters' association, chaired by Edward Aborn, called it into question, and its 1915 report repeats that it is “a very defective utensil” that gives an inefficient infusion (source 3). It nonetheless remained the coffee maker of American kitchens until the electric drip coffee makers, whose arrival the plate recounts.
The motion
The pot is filled with water, the tube and the basket filled with ground coffee are set in place, and it goes on the heat, later on an electric element. When the water boils under the base plate, steam pushes a column of water up the tube; it spurts out at the top, falls on the lid and sprays the coffee, passes through it and rejoins the pot (source 1). The cycle starts again as long as the heat is on, and strength is judged by color, often through a small glass dome on the lid.
Two things set this method apart from all the others. The water touching the coffee has just left a boil, the hottest of all methods; and it is not clean water but an already made coffee that goes back over the grounds. The chemist Charles Trigg added in 1922 that the liquid turns directly to vapor at the heating point, which carries off the aromas and spoils what is dissolved (source 5).
The settings
- Grindfine in the 1913 trial (source 3); not standardized
- Dosenot documented in the sources read
- Brew rationot documented in the sources read
- Temperaturethat of the boil at the bottom of the pot (source 1)
- Timefive minutes in the 1913 trial (source 3); in use, until the desired color
- Pressuresteam, just enough to lift the water in the tube (source 1)
What it has going for it
- It fits any pot: Goodrich's patent is a simple accessory adjustable to the water height (source 1).
- It gives a coffee free of grounds, with no eggs or clarifying agents, which was progress over the boiled coffee of its time (sources 1 and 3).
- It makes large volumes and keeps the coffee hot, which explains its place in kitchens and restaurants; in 1922, the percolation batteries of instant coffee factories hardly differed from restaurant percolators (source 5).
What is held against it
- Boiling water and the recirculation of an already made coffee both push toward overextraction: the flaw is physical, not a matter of taste (sources 1 and 5).
- The 1913 analysis finds in it ten times more “coffee tannin” per cup than in the filtration of a pulverized grind (source 3).
- The liquid that boils in contact with the heat loses its aromas to steam distillation and is spoiled, Trigg noted in 1922 (source 5).
- In the 1917 test, it finished next to last, just ahead of the boiling pot, behind the paper filter, the cloth filter and the glass filter (source 3).
What it gives in the cup
The most cited figure comes from the American roasters' association: 2.90 grains of “coffee tannin” per cup for the percolator, 2.35 for boiling, 2.31 for infusion, 0.29 for the filtration of a pulverized grind (source 3). The 1913 “tannin” is not the commercial tannic acid, the table itself specifies: it is a period name for compounds now linked to chlorogenic acids and their roasting products (source 3).
In 1915, the Mellon Institute in Pittsburgh compared nine apparatuses and found the most caffeotannic acid and caffeine in the boiling method, the least in French drip, and a better infusion between 195 and 200 °F, that is, 90 to 93 °C, than at a boil (source 3). Professor Prescott, at MIT, concluded in 1921 that water at 185 °F, about 85 °C, is better than boiling water, and that filtration is superior to the pump percolator (source 5). The cup one expects from it is thus bitter, astringent, cooked, the more so the longer the cycle lasts.
Anecdotes
- The 1889 patent has no lid to send the jet back onto the coffee: Goodrich points out that the coffee pot's own is enough (source 1).
- The patent places Goodrich in McLeansboro, in Hamilton County; nowhere does it call him a farmer, contrary to what is often read (source 1).
The sources for this sheet
- Hanson Goodrich, US patent 408,707, “Coffee-pot,” McLeansboro (Illinois), filed January 12, 1889, granted August 13, 1889 · online · primary
- William H. Ukers, All About Coffee, New York, 1922, ch. XXXIV (utensils: Laurens, Gaudet, Gandais, Durant) · online · secondary
- William H. Ukers, All About Coffee, New York, 1922, ch. XXXVI (Edward Aborn's committee, 1913-1916; Mellon Institute, 1915; Tea and Coffee Trade Journal test, 1917; boiled coffee of 1902) · online · secondary
- William H. Ukers, All About Coffee, New York, 1922, ch. XXXV (Austrian usages) · online · secondary
- William H. Ukers, All About Coffee, New York, 1922, ch. XVII (Charles W. Trigg, infusion), ch. XXV (instant coffee manufacture) and ch. XXX (S. C. Prescott's report, 1921) · online · secondary