The siphon

History

The oldest description read is by Professor Nörrenberg, published in 1827 in the Zeitschrift für Physik und Mathematik, a journal of physics and mathematics, under the title “Description of a coffee machine” (source 1). He presents it as a physics apparatus put together the previous winter to show his students the use of steam and of Hero's globe, that vessel in which compressed air drives water through a tube; several of his listeners and friends had built one in turn and used it with satisfaction (source 1). The equipment is laboratory equipment: a bottle with its bottom cut off, a glass tube, a pewter strainer, a pharmacy flask and an alcohol lamp (source 1).

The patents follow. Ukers notes in 1839 an English patent granted to James Vardy and Moritz Platow for a vacuum-operated urn with a glass upper vessel, then the first French patent for a glass apparatus, granted to Madame Vassieux, of Lyon, in 1842 (source 2); the Japanese maker Hario dates this patent to 1841 (source 3). These apparatuses herald the double glass bulbs, very popular in Europe until the end of the 19th century, then taken up again in the United States in 1914-1916: five models made a name there, two survived in 1922, heated by alcohol, gas or electricity (source 2). Loeff's patent, in Berlin, often dated 1830, was found in no primary source; Hario speaks only of a design of the 1830s (source 3).

In Great Britain, Napier's machine, around 1840, takes another form: a silver globe, an infuser where the coffee is poured in with the boiling water, and a tube ending in a cloth-covered strainer that brings the clear coffee back into the globe when the gas is lowered (source 2). Ukers attributes it to Robert Napier (1791-1876), engineer of the Clyde shipyards, who never patented it; thirty years later, it was made in Glasgow by Thomas Smith & Son under his direction, and a patent by List, in 1891, adapted it to hotels and ships (source 2). The Wikipedia encyclopedia names rather James Robert Napier (source 6).

Japan has kept the object. According to Hario, the company has made heat-resistant glass siphons since the 1940s, developed in 1946 an all-glass prototype modeled on the Cory, and the country created in 2003 a national siphon championship, followed in 2009 by a world championship (source 3).

The motion

The water is heated in the bottom bulb; the trapped air and steam drive it through the tube into the top bulb, where the grind is poured in; it is stirred, left to steep, then the heat is removed (sources 1 and 4). As it cools, the bottom bulb goes into negative pressure, and the outside air pushes the coffee through the filter, strongly enough to pass through a grind that would clog a filter cone (source 4). Nörrenberg already found the filtration much shorter than in machines where the coffee passes by its own weight alone, and the grounds so well pressed that they came out in one breath in a block molded on the neck (source 1).

Jonathan Gagné, an astrophysicist who studies the physics of coffee, published in 2019 a recipe that exploits this suction: 23.5 g of coffee for 400 g of water, an infusion at around 92 °C, the heat cut thirty-five seconds after the start, a cloth wet with cold water around the bottom bulb to speed the descent, all in under three minutes, with a paper filter rather than cloth (source 4).

The settings

• GrindFrom medium to very fine: with a paper filter, Gagné goes down to a Turkish coffee grind (source 4).
• Dose23.5 g for 400 g of water in Gagné's recipe (source 4); no standard.
• Brew ratio1:17 for Gagné (source 4); 1:15 to 1:17 in plate XV (source 7).
• TemperatureAbout 92 °C in the top bulb for Gagné (source 4); the heat stays on during steeping (source 7).
• TimeA few seconds of boiling for Nörrenberg (source 1); thirty-five seconds of steeping and under three minutes in all for Gagné (source 4).
• PressureNot quantified in the sources read: the rise comes from air and steam, the descent from a negative pressure born of cooling (sources 1 and 4).

What it has going for it

• Descent by suction allows a finer grind than in gravity percolation, and Gagné gets extraction yields of 23.2 to 24.6% from it (source 4).
• Heat stays applied during steeping, which makes the method hot and stable (source 7).
• With a paper filter, the cup is clean, with no oil or fine particles (source 4).

What is held against it

• It is a demanding method, longer to carry out and to clean than the others, by Gagné's own admission (source 4).
• The traditional cloth filter holds oils poorly, and goes rancid and moldy easily (source 4).
• Hario's paper filter holder lets fines through its central screw as soon as the grind is a little finer than for a cone (source 4).
• Drawing an already loaded infusion through the bed can make the cup flat and astringent (source 8).
• Forced cooling does not solve everything: according to Nörrenberg, cold compresses packed the strainer instead of speeding filtration (source 1).

What it gives in the cup

The siphon is a mixed method, but its cup is that of an immersion: when the coffee comes back down, the grind has bathed for almost all of the contact time (source 5). The cup is hot: Nörrenberg wrote in 1827 that, even cut with cold milk in the usual proportions, it remained too hot for most tongues (source 1).

Clarity depends on the filter. With paper and a very fine grind, Gagné gets a cup with no oil or fines, at yields that he recommends only with coffees well developed in the roast, otherwise woody or astringent notes appear (source 4). No published measurement of dissolved matter specific to the siphon was found.

Anecdotes

• Shortly before his death, at the end of a quarrel in the Glasgow workshop where his machine was made, old Napier is said to have thrown at old Smith: “You may be a good goldsmith, but I am a better engineer.” Ukers reports the scene as a tale that is told (source 2).
• Nörrenberg describes the moment when, for lack of pressure, the coffee in the tube turns into a host of small steam bubbles streaming downward, almost the same beautiful spectacle, he writes, as the circulation of blood in young fish under a good microscope (source 1).

The sources for this sheet

1. Professor Nörrenberg, “Beschreibung einer Kaffehmaschine,” Zeitschrift für Physik und Mathematik, vol. 3, 1827, pp. 269-272 (Bayerische Staatsbibliothek, text recognized from the digitization) · online · primary
2. William H. Ukers, All About Coffee, New York, The Tea and Coffee Trade Journal Company, 1922, chapters XXXIV and XXXVI and chronology (UTF-8 transcription from the local library): Vardy and Platow 1839, Vassieux 1842, Napier 1840, List 1891, American revival 1914-1916 · online · primary
3. Hario UK, “History of the syphon,” September 17, 2026 (manufacturer) · online · manufacturer
4. Jonathan Gagné, “A Recipe to Brew High Extraction Coffee with the Siphon,” Coffee ad Astra, March 25, 2019 · online · secondary
5. Jonathan Gagné, “Why do Percolation and Immersion Coffee Taste so Different?” Coffee ad Astra, July 16, 2019 · online · secondary
6. Wikipedia (en), “Vacuum coffee maker,” accessed October 21, 2026, which cites John W. List, “The Napier coffeemaker,” 2007 · online · secondary
7. PXL Roasters, plate XV “Preparations,” section 08, The siphon (preparations-08.html), and table of section 04 (preparations-04.html), revision of September 24, 2026 · internal
8. Jonathan Gagné, “Reaching Fuller Flavor Profiles with the AeroPress,” Coffee ad Astra, September 7, 2021 (remark on the siphon) · online · secondary