Moka pot coffee without the folklore

Pressure: steam, not a pump

Espresso is built around a pump holding roughly 9 bar through a compacted puck of coffee [5]. A moka pot has no pump, no gauge and no lever. The force that moves the water is the vapour pressure of the water in the boiler itself, which grows as that water heats and lifts the liquid up the funnel [1]. It is the same order of magnitude as atmospheric pressure — about one bar — which is why the device carries a safety valve rather than a pressure readout, and why the drink in the cup is far less concentrated than a shot.

The documented description of the process splits it into two phases [1]. The first is a liquid–solid extraction in which both the water temperature and the water flow rate change over time. The second is an intense evaporation phase, and the soluble material it carries out is the material that works against cup quality [1]. Those two phases give you the single timing decision that matters: the steady stream is phase one, and the pale, sputtering tail is phase two. The Specialty Coffee Association publishes machine standards for home brewers, home grinders and espresso machines [6]; a stove-top pot falls outside those documents, so the figures on this page come from the process studies cited below instead.

The boiler load: 15 g into 150 mL

A documented moka protocol fills a 3-cup pot to the safety valve — 150 mL of water — and doses 15 g of ground coffee, which is a 1:10 coffee-to-water ratio [1]. Per gram of coffee that is 10 mL of water, against the 1:15 the filter recipes in a 2023 comparison of four brewers are written at [2].

The narrow ratio is not a style preference. Basket volume caps the dose and boiler volume caps the water, and on a fixed pot the two move together: scale both, keep the 1:10, and the brew repeats. On a 6-cup pot the same logic gives roughly 30 g of coffee and 300 mL of water.

One more piece of arithmetic explains the small cup. A filter bed in the same comparison retained about 2.1 times its coffee mass in water [2]; at that ratio a 15 g bed holds back on the order of 30 g of the 150 mL before a drop has evaporated. Expect the cup to be meaningfully smaller than the fill, and weigh the water in rather than trusting a fill line drawn on the boiler wall.

Temperature and grind

Water leaves the boiler close to its boiling point, because it is that boiling which creates the pressure that lifts it. The coffee bed is therefore wetted at around 100 °C, against the 93 °C used in the filter recipes of the 2023 comparison [2]. That gap is one reason a moka cup of a light-roasted coffee can read flatter or harsher than a pour-over made from the same beans. The chemistry is method-specific as well: moka brews have been analysed alongside espresso, Neapolitan and American preparations in the same study [3], which is part of why two pots of identical beans can taste nothing alike.

Grind sits on the other side of the same coin. The median particle sizes measured in the 2023 comparison were 800 µm for AeroPress, 945 µm for V60, 1030 µm for a low-pressure paper brewer and 1290 µm for a French press [2]. A moka basket is narrower and the pressure behind it is only about one bar, so the bed has to offer some resistance to the flow without stalling it: fine enough that water spreads through the puck instead of finding a channel, coarse enough that the stream does not die halfway up the column.

Water, scale and the aluminium body

Water quality does more work in a moka pot than in most brewers, because the boiler is where scale lands. The specification used in the 2023 study is a dry residue of 75–250 mg/L, ideally 100–150 mg/L; a calcium hardness of 50–175 ppm as CaCO₃; and a pH between 6.5 and 7.5 [2]. Water beyond the hardness ceiling deposits carbonate on the boiler floor and around the valve, and scale on a valve is a maintenance issue before it ever becomes a flavour one.

Aluminium bodies add a second consideration: migration of aluminium from moka pots into the drink has been measured directly in a food-contact study that also covered drinking bottles and other tableware [4]. A stainless-steel pot avoids the question entirely. With an aluminium pot, keep the interior free of abrasive pads and acidic soak solutions, and do not leave spent coffee sitting in the boiler overnight.

Three failure modes and their causes

1. Sputtering and a pale finish. Once the stream thins and hisses, extraction has crossed into the evaporation phase, and what follows is the material the two-phase description identifies as detrimental to quality [1]. Take the pot off the heat at the first sign of lightening rather than waiting for the gurgle to stop.
2. A burnt, hollow cup. Heat applied up the sides of the boiler cooks the metal above the water line and drives the bed temperature up before the extraction has finished. Keep the flame or ring smaller than the base of the pot and centred under it, and start from hot water in the boiler so the bed spends less time at the top of its temperature range.
3. Flow that starts at once and ends weak. A grind that is too coarse, a basket that is not full, or a funnel that is not seated all let water past the coffee rather than through it. The 15 g dose in a 150 mL boiler [1], a level bed, and a funnel pressed home remove all three causes at once.