Thermodynamics

The ladle that makes a sauna hotter without warming it

Throw a ladle of water on the stones of a Finnish sauna and a wave of heat rolls down from the ceiling and stings the skin — while the thermometer on the wall barely moves. The heat is real and it is not in the air's temperature. Skin at 37 °C has its own vapour pressure, 6.3 kilopascals; air with less vapour than that lets sweat evaporate and cools the skin, air with more condenses on it and pays it 2.4 kilojoules for every gram. A ladle carries the room across that line. What skin feels is the wet-bulb temperature of the air, and a thermometer cannot read it.

Assumes: The thermometer that reads low when wet · The heat that changes no temperature, and where it actually goes

A Finnish sauna runs at about eighty degrees Celsius and is bearable for ten or fifteen minutes at a time. A Turkish steam room runs at forty-five and is bearable for about the same. Anyone who has sat in the first knows the moment that marks it: someone ladles water onto the hot stones, there is a hiss, and a few seconds later a wave of heat descends from the ceiling that makes the skin sting and sends bathers on the top bench ducking. The Finns call the steam löyly. The thermometer on the wall, in the middle of all this, moves by a degree or two, and sometimes down.

The heat is real. It is carried not by the air’s temperature but by the water vapour in it, and whether vapour heats skin or cools it is decided by comparing two pressures. The argument is the one the thermometer that reads low when wet made for a wet thermometer in warm air, run the other way: here the air is wetter than the skin, and the vapour goes the other way.

Two ways for heat to reach skin

Hot air heats skin by convection: heat flows across the thin layer of slower air next to the skin at a rate proportional to the temperature difference, h(Ta−Ts)h(T_a - T_s), with a coefficient hh of about eight watts per square metre per kelvin for a seated body in still air. Water vapour crosses the same layer, by diffusion, at a rate proportional to the difference between the vapour pressure in the room and the vapour pressure at the skin. If the skin is wet with sweat at 37 °C, the vapour pressure at its surface is the saturation pressure of water at 37 °C, 6.26 kilopascals. If the room’s vapour pressure is lower, sweat evaporates and takes its latent heat from the skin. If the room’s is higher, vapour condenses on the skin, wet or dry, and gives its latent heat to it.

The same layer governs both transfers, and for air and water vapour the two coefficients are tied together: the mass transfer coefficient is the heat transfer coefficient divided by the air’s heat capacity per volume, to within the small correction of the Lewis number. Put that in, and the latent flux becomes

qL=h (pa−ps)γ,γ=cpP0.622 L≈68 Pa/K.q_L = \frac{h\,(p_a - p_s)}{\gamma}, \qquad \gamma = \frac{c_p P}{0.622\,L} \approx 68\ \text{Pa/K}.

Every 68 pascals by which the room’s vapour pressure exceeds the skin’s acts on the skin like one more degree of air temperature.

Past the skin's vapour pressure, the steam pays its latent heat to the skin. The heat flowing into skin at 37 °C from sauna air at 80 °C, against the air's vapour pressure, for skin wet with sweat and for dry skin, with a convective coefficient of 8 W/m²K. Below 6.26 kPa sweat can evaporate and wet skin is cooled; above it vapour condenses on the skin, wet or dry, and every gram delivers 2.4 kJ. A typical Finnish sauna at 10% humidity, 4.7 kPa, sends 164 W/m² into sweating skin. A 0.2 L ladle on the stones of a 10 m³ room adds 3.3 kPa and the flux becomes 547 W/m² — 3.3 times as much — at the same air temperature.
Fig. 1 The heat flowing into skin at 37 °C from sauna air at 80 °C, against the air’s vapour pressure, for skin wet with sweat and for dry skin. Below 6.26 kPa sweat can evaporate and wet skin is cooled; above it vapour condenses on the skin, wet or dry, and every gram delivers 2.4 kJ. A typical Finnish sauna at 10% humidity, 4.7 kPa, sends 164 W/m² into sweating skin. A 0.2 L ladle on the stones of a 10 m³ room adds 3.3 kPa and the flux becomes 547 W/m², 3.3 times as much.

A Finnish sauna at eighty degrees and ten per cent relative humidity has a vapour pressure of 4.7 kilopascals. That is below the skin’s 6.26, so sweat can evaporate, and it does: convection brings 344 watts per square metre to the skin and evaporation takes 180 away, leaving 164. Dry skin, before the sweating starts, takes the whole 344. A ladle of water, two decilitres thrown on the stones of a ten-cubic-metre room, adds twenty grams of vapour to each cubic metre — 3.3 kilopascals — and the room’s vapour pressure goes to 8.0. That is above the skin’s. Evaporation stops, condensation begins, and the flux to the skin becomes 547 watts per square metre: more than three times as much, with the air temperature unchanged.

Past the line, every ladle adds the same amount. Each two decilitres adds 3.3 kilopascals, and every kilopascal above the skin’s is worth fifteen degrees of air temperature, so each further ladle adds about 385 watts per square metre for as long as its steam lasts — the heat of nearly fifty degrees more air, delivered by a puff of vapour. Below the line the first part of the ladle’s effect is spent removing the evaporative cooling the bather was enjoying; above it, it is all heating.

The sting that bathers feel is this switch, made in the few seconds it takes the steam to mix down from the ceiling. The air temperature is the same, and the heat has more than tripled.

The skin’s own vapour pressure is not quite that of pure water, and the difference points the same way. Sweat is a dilute salt solution, a few grams of salt per litre, and dissolved salt lowers water’s vapour pressure in proportion to the fraction of its molecules that are ions — the effect the salt that dissolves in the air’s own water followed to its extreme, where a salt crystal pulls water out of air that is not saturated. For sweat the lowering is a fraction of a per cent, tens of pascals, so sweat-wet skin begins to collect condensation slightly before pure water would. As sweat evaporates and its salt is left behind, the skin’s surface grows saltier and the effect grows; it is small beside the ladle’s 3.3 kilopascals, and it never favours the bather.

What skin feels is the wet-bulb temperature

The flux into wet skin is the sum of a temperature difference and a vapour-pressure difference, and the sum can be rewritten as a single temperature. Every state of the air — every pair of air temperature and vapour pressure — that gives the same flux lies on a straight line in the plane of those two quantities, with a slope of minus γ, and those lines are exactly the lines of constant wet-bulb temperature: the temperature a wet thermometer settles at in that air.

What the skin feels is the wet-bulb temperature, not the air's. Air states plotted by temperature and vapour pressure, with lines along which wet skin at 37 °C receives the same heat; each line is a wet-bulb temperature (30, 37, 45 and 55 °C). Air on the 37 °C line gives sweating skin nothing at all, however hot it is. A Finnish sauna at 80 °C and 10% humidity has a wet-bulb temperature of 40 °C; a ladle on the stones moves it to 46 °C without changing the thermometer. A steam room at 45 °C and saturated is at 45 °C — hotter to wet skin than the sauna — while desert air at the same 45 °C is at 21 °C and cools a sweating body.
Fig. 2 Air states plotted by temperature and vapour pressure, with lines along which wet skin at 37 °C receives the same heat; each line is a wet-bulb temperature (30, 37, 45 and 55 °C). Air on the 37 °C line gives sweating skin nothing at all, however hot it is. A Finnish sauna at 80 °C and 10% humidity has a wet-bulb temperature of 40 °C; a ladle on the stones moves it to 46 °C without changing the thermometer. A steam room at 45 °C and saturated is at 45 °C, while desert air at the same 45 °C is at 21 °C and cools a sweating body.

So the quantity a sweating body responds to is the air’s wet-bulb temperature compared with the skin’s own. Air whose wet-bulb temperature is below 37 °C can cool sweating skin, however hot it is; air whose wet-bulb temperature is above 37 °C heats skin whatever the body does, because neither convection nor evaporation can carry heat away. The Finnish sauna, dry and at eighty degrees, has a wet-bulb temperature of only forty — a little above the skin’s, and the reason the room is bearable at all. One ladle moves it to forty-six. A Turkish steam room, saturated at forty-five, is at forty-five, and desert air at the same forty-five degrees but ten per cent humidity is at twenty-one and is a place where a sweating body can shed heat freely, if it can find the water to sweat.

The lines have a second reading. Moist air carries heat in two forms, its sensible heat and the latent heat of its vapour, and their sum per kilogram of dry air, the air’s enthalpy, is almost constant along each wet-bulb line; the two are the same lines to within a few per cent. So what the skin responds to is, very nearly, how much heat the air carries per kilogram, counting the heat hidden in the vapour. A ladle that leaves the temperature alone and adds twenty grams of vapour per cubic metre raises the air’s enthalpy by about as much as heating it by fifty degrees would, which is how the bathers feel it.

The wall thermometer measures the dry-bulb temperature. It is the right instrument for the stove and the wrong one for the bather. A sauna’s real heat is read by wetting the thermometer, which is precisely the instrument the essay on wet thermometers built — and in the steam room, where the two agree, the thermometer for once tells the truth.

Where the heat of the steam comes from

The ladle’s vapour carries heat, and the heat has to come from somewhere. Evaporating two decilitres of water from twenty degrees takes 518 kilojoules: the water must be warmed to a hundred and then paid its latent heat, the heat that changes no temperature, which is most of it.

The steam's heat is the stones' heat, carried to the skin. The energy needed to turn a 0.2 L ladle of water at 20 °C into steam, 518 kJ, expressed as the temperature drop of whatever supplies it. Ten cubic metres of air at 80 °C hold only about 10 kJ per kelvin, so if the air paid it would cool by 52 K. The stones of the stove, some tens of kilograms at several hundred degrees, pay instead and drop by about 16 K, while the air temperature barely moves. The vapour carries that heat across the room as latent heat and gives it up on the coolest surfaces it meets, of which the bathers are some.
Fig. 3 The energy needed to turn a 0.2 L ladle of water at 20 °C into steam, 518 kJ, expressed as the temperature drop of whatever supplies it. Ten cubic metres of air at 80 °C hold only about 10 kJ per kelvin, so if the air paid it would cool by 52 K. The stones of the stove, some tens of kilograms at several hundred degrees, pay instead and drop by about 16 K, while the air temperature barely moves.

The room’s air could not pay it: ten cubic metres of air at eighty degrees hold about ten kilojoules per kelvin, and if the ladle’s heat came from the air the room would cool by fifty-two degrees. The stones pay instead. A sauna stove carries tens of kilograms of stones heated to several hundred degrees, and it is their heat, not the air’s, that the steam carries away; forty kilograms of stone cool by about sixteen degrees per ladle and are reheated by the stove. The vapour then carries that heat across the room as latent heat and releases it on the coolest surfaces it touches — the walls, the benches, and the bathers. It is exactly the mechanism of the pipe that carries heat as vapour, in which a liquid evaporates at the hot end, flows as vapour and condenses at the cold end, carrying far more heat than conduction could; a sauna is a heat pipe with the bathers as its condenser.

The sweat that is not sweat

Condensation is visible, once looked for. The water that beads on skin in the minute after a ladle is partly the room’s water arriving, not the body’s leaving.

Much of the water running down a bather after a ladle is not sweat. The rate at which water vapour condenses on a body with 1.8 m² of skin at 37 °C, against the vapour pressure of the sauna air, once it exceeds the skin's own 6.26 kPa. After a 0.2 L ladle, 8.0 kPa, about 9 grams a minute condense — over half the rate of hard sweating, which is about a litre an hour. A good part of the water that beads on the skin in the minutes after the steam reaches the bench is the room's water, and each gram of it delivers 2.4 kJ to the skin as it forms.
Fig. 4 The rate at which water vapour condenses on a body with 1.8 m² of skin at 37 °C, against the vapour pressure of the sauna air, once it exceeds the skin’s own 6.26 kPa. After a 0.2 L ladle, 8.0 kPa, about 9 grams a minute condense — over half the rate of hard sweating, which is about a litre an hour.

With the room at eight kilopascals, about nine grams of water a minute condense onto a body with 1.8 square metres of skin. Hard sweating, in a hot sauna, runs at about a litre an hour, seventeen grams a minute. So for as long as the steam lasts, a large part of the water running down a bather is the room’s, and each gram of it has delivered 2.4 kilojoules on arrival. It makes a sauna session feel sweatier than it is: weighing before and after, the standard way to measure sweat loss, gives the right answer only because the condensed water drips off or is towelled away along with the sweat.

The vapour does not have to reach the skin to give up its heat. A steam wave descending from the ceiling meets cooler air lower down, and some of it condenses into the visible mist that follows a large ladle, warming that air as it does — the same release of latent heat that, in the cloud that cools more slowly than the air, keeps rising cloudy air warmer than dry air lifted the same distance. A bather on the lower bench may meet air that has been warmed by the steam’s condensation rather than the steam itself, and a little of the ladle’s effect arrives as temperature after all.

The steam’s heat is front-loaded, too. Condensation on the bathers and on the cooler walls, ventilation and the stones’ own reabsorption bring the vapour pressure back down over a minute or two, and the room returns towards its dry state. The sting is a pulse, which is why bathers who like it ask for another ladle.

Some of the heat is returned. Water that condensed on the skin while the room was above 6.26 kilopascals is still there when the room falls back below it, and it then evaporates, taking its latent heat away again — the bather’s skin running the room’s heat pipe in reverse. If every gram that condensed later evaporated from the skin, the ladle’s net effect over a cycle would be only the convective part and the delay; in practice much of the condensate runs off or is wiped away with the sweat, and the heat it delivered stays delivered. That asymmetry, together with the body’s own rising temperature, is why a sauna session leaves a bather hotter than the dry room alone would have done in the same time.

Saturated air at fifty degrees outdoes dry air at a hundred and ten

The same arithmetic explains a fact about bathing traditions that otherwise looks inconsistent: Finnish saunas run at eighty to a hundred degrees and steam rooms at forty to fifty, and people find them roughly equally hot.

Saturated air at fifty degrees outdoes dry air at a hundred and ten. The heat flowing into skin at 37 °C against air temperature, for very dry air (1 kPa of vapour) reaching dry skin and sweating skin, and for saturated air. Dry air heats skin only by convection, 584 W/m² at 110 °C, and sweat that can evaporate fast enough would cancel all of it. Saturated air adds condensation to convection and sweat cannot evaporate into it at all: at 50 °C it delivers 815 W/m². A Finnish sauna at a hundred degrees is bearable for minutes; a steam room is kept below fifty for the same reason. The line for saturated air stops at 62 °C, past which no one sits in it.
Fig. 5 The heat flowing into skin at 37 °C against air temperature, for very dry air (1 kPa of vapour) reaching dry skin and sweating skin, and for saturated air. Dry air heats skin only by convection, 584 W/m² at 110 °C, and sweat that can evaporate fast enough would cancel all of it. Saturated air adds condensation to convection and sweat cannot evaporate into it at all: at 50 °C it delivers 815 W/m².

Very dry air at a hundred and ten degrees heats dry skin at 584 watts per square metre by convection alone, and sweat able to evaporate into it would carry almost all of that away again; the dry sauna’s high temperature is survivable because its air is thirsty. Saturated air at fifty degrees delivers 815 watts per square metre — more than the dry air at a hundred and ten — because it adds condensation to convection and allows no evaporation at all. A steam room is kept below fifty for that reason, and saturated air much above sixty degrees is not a place anyone sits.

The same comparison sets the danger in hot climates. A wet-bulb temperature above about 35 °C means that even a resting, sweating, shaded body cannot lose the heat its own metabolism makes, and the essay on wet thermometers ended on how close parts of the tropics now come to that limit. A sauna deliberately exceeds it for a few minutes, with the bather’s body temperature rising by a degree or so before they leave; the stone, the ladle and the clock are the controls.

The same exchange at a smaller scale

Condensation on a cooler surface is the ordinary way water moves when vapour pressures differ. The ice that grows by stealing from the droplets found ice crystals in a cloud growing at the expense of supercooled droplets because ice’s vapour pressure is lower than water’s, so vapour leaves the droplets and arrives on the ice; here vapour leaves the hot stones’ steam and arrives on the skin, whose vapour pressure is lower than the room’s. The bench under the bather plays a different part. The metal that feels colder than the wood used the sauna bench as its example of a surface that can be touched at ninety degrees because pine barely conducts; the steam, by contrast, does not need to conduct anything — it carries its heat in its latent form and gives it up exactly where it condenses.

What the single coefficient leaves out

The fluxes here use one convective coefficient for the whole body, a uniform skin temperature of 37 °C, and the Lewis analogy with a factor of one. In a real sauna the skin warms to about forty degrees within minutes, which raises the skin’s vapour pressure to 7.4 kilopascals and moves the line between evaporation and condensation; the coefficient varies over the body and grows when the steam wave arrives, because the steam brings moving air with it; and the room is strongly stratified, often twenty degrees hotter at the ceiling than at the floor, so a bather on the top bench meets a different air from one on the bottom. Radiation from the hot walls and stove, which adds heat independently of the air, is ignored, and it is not small: walls and a ceiling near eighty degrees exchange heat with skin at a rate equivalent to a coefficient of six or seven watts per square metre per kelvin, close to convection’s eight, and the stove itself, far hotter, adds a strong beam on the side facing it. A room’s warmth to the body is a combination of its air and its surfaces, the operative temperature that the room that is cold at twenty-one degrees found making a room with cold walls feel cold at a comfortable air temperature; a sauna is that effect run hot, and radiation roughly doubles the dry part of the flux without touching the ladle’s contribution. The ladle’s effect is computed for perfect mixing into the whole room; in fact the steam arrives as a pulse much stronger at first and near the stove. The numbers describe the mechanism and its scale, not a particular sauna.

Still open: how much heat the body can take, and how fast

The physiology at the end of this physics is less settled than the physics. How a body’s core temperature responds to a given heat flux into the skin depends on blood flow to the skin, sweat rate, hydration and acclimatisation, and the long-term health associations of regular sauna bathing reported from Finnish cohort studies — lower cardiovascular mortality among frequent bathers — are associations, not established effects. Whether the brief heat pulses of löyly matter differently from steady heat, and what limits of wet-bulb temperature and duration are safe for whom, are questions that the same arithmetic frames but cannot answer.

Skin wet with sweat at 37 °C has a vapour pressure of 6.26 kPa: air with less vapour lets sweat evaporate and cools it, air with more condenses on it and pays it 2.4 kJ a gram, and every 68 Pa of excess counts as one more degree of air temperature. A 0.2 L ladle in a 10 m³ sauna lifts the vapour pressure from 4.7 to 8.0 kPa, past the skin’s, and the heat into sweating skin goes from 164 to 547 W/m² at an unchanged air temperature — the stones, not the air, paying the 518 kJ. What the skin feels is the air’s wet-bulb temperature, 40 °C before the ladle and 46 °C after, which is why saturated air at 50 °C outdoes dry air at 110 and why the thermometer on the sauna wall is the wrong instrument for the bather.

Part 14 of 14

This essay is one argument about Chemical potential. The others:

The objects named here

The third axis, after the field and the reading path: the things themselves, and every essay that touches each one.

CondensationHeat-transfer coefficientLatent heatLewis numberVapour pressureWet-bulb temperature