The Hearth in the House as a System

Temperature difference

A balloon rises off the ground because of the buoyancy of hot air

Everyone knows that hot air rises. The behavior of a hot air balloon or the pattern of air and gas flow around a camp fire illustrate this natural law. Hot air rises because air expands when heated, becoming lighter or more buoyant than the cooler surrounding air. The tendency of hot air to rise is central to our discussion because it produces both natural draft in chimneys and stack effect in houses.

Draft is the pressure difference that is available to drive the flow of air and/or combustion gases through an appliance and its venting system. All combustion systems that are vented by natural chimney draft depend on the temperature difference that is maintained between the gases in the flue and the outdoor air for proper operation. The greater the temperature difference, the more draft is produced. If the temperature difference is not adequate, such as for example when a smoldering fire produces low flue gas temperature, the upward flow in the chimney becomes weak and unstable, and smoking can result.

Hot gases rise from a camp fire

Conversely, very large temperature differences produce high draft levels. High draft can cause rapid deterioration of the internal components of fireplaces or wood stoves because of the higher temperatures that result from overfiring. Ideally, the chimney-to-outdoor temperature difference and the resulting draft should fall between the low levels that can lead to smoking and the high levels that waste energy and can lead to appliance damage.

Draft can also be developed by mechanical means. For example, forced draft is mechanical draft created by a fan located so that it pushes the flue gases through the chimney. Most pellet stoves have exhaust fans that draw gases out of the combustion chamber and force them into the chimney or vent. Induced draft is mechanical draft created by a fan located so that it pulls the flue gases through the chimney or vent. A chimney top exhaust fan induces draft in the system.

We will concentrate on here on achieving reliable venting by natural chimney draft. The other forms of draft, those developed by mechanical means such as forced draft and induced draft, are less affected by external influences. Note, however, that successful chimney venting is essential, not just when the appliance is operating, but also during standby periods when mechanical draft systems are inactive.

Natural chimney draft is a weak force. To put its relative strength into context, it is necessary to understand how draft occurs. Atmospheric pressure is created by the weight of the blanket of air surrounding the earth. An operating chimney represents a column of gas that is hotter, less dense and therefore lighter than the surrounding air. Draft is the difference between the pressure at the base of the chimney and atmospheric pressure.

Subject to the elevation above sea level, atmospheric pressure is about 100 kilopascals, or 100,000 pascals. High draft developed by a residential chimney full of very hot gases is only about 50 Pa, or 1/2000 of atmospheric pressure. Here is another example: when you are swimming and your ear goes four inches under the surface of the water, your ear drum is exposed to about 1000 Pa of pressure.

The weight of the blanket of air surrounding the earth produces an atmospheric pressure of around 100,000 Pa. By comparison, what we think of as strong chimney draft at only 50 Pa is a relatively weak force.


Taller chimneys produce more draft because the total weight difference between the taller columns of hot and cold air is greater than the weight difference between the shorter columns.

Taller chimneys usually produce stronger draft at a given temperature difference. To visualize why this is so, imagine two columns of air, one hot and one cold. The difference in weight between the hot and cold columns is the draft. Now make both columns taller. The total difference in weight between the two taller columns is greater than the difference in weight of the shorter columns. The taller the chimney, therefore, the more draft it will produce at a given temperature difference.
A rule of thumb suggests that the minimum system height (from hearth to chimney top) should exceed 15 feet in order for it to provide adequate draft. A system installed in a bungalow with a shallow-pitch roof can be less than this height. When diagnosing venting problems, consider total system height as a possible factor.

In practice, increasing the height of an existing chimney may not result in increased draft because the extra length tends to result in greater heat loss. Taller chimneys only produce more draft if temperature difference remains nearly constant.

The reliability of venting through a chimney operating on natural draft is mainly dependent on the temperature difference between the gases in the flue and the outdoor air. There are several other factors that influence chimney venting, but keep in mind that the ability of a chimney to maintain temperature difference establishes its tolerance to the potentially adverse effects of the other factors.


Pascal (Pa) is the metric unit of measurement for pressures. One Pa equals 0.004 inches of water column (wc). We will use pascals in our discussion of pressures. A single pascal is a very small amount of pressure; a single sheet of paper exerts a pressure of about one pascal on a surface. A good way to learn about the pascal pressure measurement is to relate it to the strength of chimney draft, as follows:

  • 5 Pa equals 0.02"wc
    This is lousy draft for a wood fire. At 5 Pa of draft, a wood fire will be almost impossible to kindle and a stove or fireplace will spill smoke when you open the door. On the other hand, this is about the right amount of draft for a gas or oil flame; draft hoods and draft regulators are used to maintain this level.
  • 12 Pa equals 0.05"wc
    This is minimum draft for reasonable operation. A wood stove or fireplace operating on a draft of 12 Pa will probably spill smoke when the door is opened, but it will burn reasonably well when the door is closed.
  • 25 Pa equals 0.1"wc
    This is good draft. A wood stove or fireplace operating on 25 Pa draft will produce a bright hot fire and will probably not spill smoke when the door is opened if the appliance is of good design.

In cold weather stack effect creates a pressure greater than atmospheric pressure at high levels of the house and a pressure lower than atmospheric pressure at low levels of the house. As is the case with draft in chimneys, the greater the temperature difference, the more stack effect is produced; the taller the building, the more powerful is the stack effect. The neutral pressure plane (NPP) is the level between the high pressure zone at upper levels and the low pressure at lower levels in a house at which the pressure is equal to atmospheric pressure.  If the leaks in the building envelope were evenly distributed, the neutral pressure plane would be at the vertical midpoint of the building.

A house built over a vented crawl space may have leaks evenly distributed around the envelope, so the neutral pressure plane can be at about the vertical midpoint, like the one above.

A house built on a concrete slab (which has no leaks) will have most of its leaks higher, and, because the NPP follows the leaks, will tend to have a higher neutral pressure plane, like the one above.

A house built with a basement below grade will tend to have most of its leaks concentrated higher in the heated space and so its neutral pressure plane will usually be significantly higher than the vertical midpoint, as in the example above.

In practice, the majority of leaks in most houses are above the vertical midpoint. Since the neutral pressure plane follows the leaks, it is normally higher in the building than the midpoint.


The fact that the neutral pressure plane follows the leaks is significant. When an upstairs window is opened (in effect, a very large leak), it causes the NPP to rise to its level. This creates a greater level of negative pressure low in the house, and in extreme cases, can cause spillage or backdrafting in a basement hearth.

When a basement window is opened, the NPP goes down to its level, reducing the negative pressure there. The lowering of the NPP is the reason that the flow in backdrafting chimney serving a fireplace or stove located low in the house can be corrected by opening the nearest window.

Stack effect is produced by temperature difference and is not significantly influenced by the leakiness of the house envelope. At a given outdoor temperature in winter, a pressure difference is created from bottom to top of all houses, but a greater volume of air flows into and out of a leaky house than a tight house.

Weatherizing older houses by installing new windows and caulking leaks often has the effect of raising the neutral pressure plane. This occurs because the most noticeable leaks are those low in the building where cold air leaks in. These leaks are considered the highest priority for weatherizing because they directly affect comfort. Leaks high in the house where air flows out are not noticed by the householder and so they tend to be dealt with last. Since the NPP follows the leaks, it rises toward the majority of the leaks high in the house when lower leaks are sealed. This explains why weatherizing procedures can lead to venting failure, particularly of appliances located low in the house.

Table 1 below shows how temperature difference and stack height affect pressure. The numbers in the middle are the pascals of pressure difference created by various stack heights at various temperature differences. Using the table, if difference between the average flue gas temperature and the outdoor temperature is 400°F in a 25 foot chimney, the draft would be 44 Pa. Or, if the outdoor temperature is 20°F (producing a 50°F temperature difference), the stack effect in a house with a total envelope height of 20 feet would be 7 Pa.

Temperature difference, stack height and pressure difference
Pressure in pascales for various stack heights and temperature differences

Average temperature difference
°F (°C)

Note: this is the average temperature between the gas in the stack and the outside air; it is not a conversion.

1000 (555) 26 39 52 65 78 92 105
800 (444) 24 36 48 60 73 85 97
600 (333) 21  32 43 54 64 75 86
400 (222) 18 26 35 44 52 61 70
200 (111) 11 17 23 28 34 39 45
100 (56) 7 10 13 16 20 23 26
50 (28) 4 5 7 9 11 13 14
20 (11) 2 2 3 4 5 5 6

Height of stack in feet (metres)


The figures in the table are calculated projections of total draft. If you were to measure the temperature in a chimney of known height, then measure the draft using a manometer, you would see a pressure considerably lower than the figure in the table. There are at least three reasons for this. First and most important, the temperature difference figures on the vertical axis of the table are based on the average temperature in the stack, less the outdoor temperature. Your thermometer is measuring the gas temperature at the base of the chimney, which can be hundreds of degrees higher than the exit temperature at the top of the chimney, particularly if the chimney is air-cooled and/or runs outside the building envelope. The second reason for a lower measured pressure is that this table does not account for friction losses in a chimney. And thirdly, your manometer probe is measuring only static pressure. Without more equipment, the manometer will not read the velocity pressure in the stack, a factor which is added to static pressure.

The table is presented as an aid to visualizing the relationships involved in chimney draft and stack effect in buildings. You can consider the lower part of the table as referring to stack effect in houses or the draft developed in a chimney at standby. In fact, you might find that the lower part of the table more closely matches actual measurements of stack effect because the indoor temperature will be consistent throughout the house.


Back to the Table of Contents