Contents
The neutral plane is the horizontal level at which hydrostatic pressure is equal on the two sides of an enclosure boundary. It provides the reference for interpreting the pressure differences that drive air and smoke through openings. For a warm enclosure surrounded by cooler air, pressure is generally higher inside above this level and lower inside below it. An opening that crosses the level can therefore carry flow in both directions.[1]
The neutral plane describes pressure, whereas a smoke-layer interface describes gas stratification. Their elevations can differ. Its position also depends on the distribution of openings and leakage, rather than temperature alone. In a fire calculation, the relevant pressure reference, adjoining spaces and opening arrangement must be specified before the neutral plane can be used to interpret the direction and magnitude of gas movement.[2]
Pressure balance and flow direction
Pressure decreases with elevation in a stationary gas column because the column supports the weight of the gas above it. A cooler, denser column has a different vertical pressure gradient from a warmer, less dense column. Comparing the pressures at equal heights on opposite sides of a boundary therefore produces a pressure difference that varies with height. The neutral plane is the elevation at which that difference is zero.[3]
For the usual warm-room case, the internal pressure exceeds the external pressure above the neutral plane. Gas can then flow outward through an opening at that elevation. Below the plane, the external pressure exceeds the internal pressure and fresh air can enter. These directions follow the pressure difference, rather than a general rule that every high opening is an outlet or every low opening is an inlet.[4]
An opening that extends across the neutral plane contains an upper outflow region and a lower inflow region under these conditions. The two streams belong to one opening but experience different pressure differences and gas densities. Their contributions must be treated separately when calculating vent flow. The opening's total area alone does not identify the area available to each direction of movement.[5]
The pressure at the neutral plane is not zero absolute pressure. It is equal pressure across the specified boundary. Likewise, a zero pressure difference at one height does not imply zero flow throughout the complete opening. The pressure difference normally becomes nonzero away from that height, permitting opposing streams above and below. This distinction is essential when reading pressure profiles or model output.[1][5]
The compared spaces also need to remain identifiable. A room may exchange gases with a corridor, a shaft or the external atmosphere. A neutral level defined between a shaft and outside air is not automatically the relevant level for every internal doorway. CFAST handles this by calculating pressure differences between the particular compartments connected by each vent, using the layers on both sides.[5][2]
Neutral plane and smoke-layer interface
The hot gas layer forms through the accumulation of heated combustion products and entrained air. Its lower interface separates the modelled upper and lower gas zones. That boundary represents a change in gas properties, including temperature and density. The neutral plane instead identifies equal pressure between two spaces. These are different physical definitions even when their elevations happen to be close.[6]
The distinction becomes visible in a pressure-based vent model. CFAST divides a wall opening at the layer interfaces of both compartments and at the neutral plane. A layer interface determines which upstream density and temperature apply. A neutral plane determines where flow changes direction. Separate subdivision at these elevations allows the model to represent streams between upper and lower layers without identifying pressure balance with the smoke boundary.[5]
Walton and Thomas describe a limited approximation in which the neutral plane and layer interface are near the middle of a typical opening. That approximation concerns pre-flashover fires after initial smoke filling is complete and exchange flow is established. It should not be applied automatically during the first accumulation of smoke. Nor does it establish a permanent relationship between the two elevations as the fire develops.[7]
Near flashover, their simplified description places the gas-layer interface close to the floor. The vent-flow calculation still requires a pressure balance. Treating the interface height as the neutral height would therefore discard an independently needed quantity. In more complete compartment calculations, layer properties and pressure are found together from mass and energy balances, with the vent model supplying the resulting flows.[7][5]
A visual smoke boundary consequently cannot establish the pressure distribution by itself. It supplies information about stratification, which affects density and therefore the vertical pressure gradients. The remaining pressure reference and connections to adjoining spaces are still required. Conversely, a pressure reversal within a doorway does not establish that the lower stream contains no combustion products under every possible compartment condition.[6][5]
Stack effect and fire buoyancy
The same pressure principle applies to stack effect in buildings. When air inside a shaft is warmer than outside air, normal stack effect tends to produce inward flow at lower elevations and outward flow higher up. The pressure difference is negative below the neutral plane and positive above it. The height of the gas columns makes this mechanism particularly relevant to tall shafts.[8]
When the inside air is cooler than outside, the density relationship reverses. Reverse stack effect can then produce positive internal pressure below the neutral plane and negative internal pressure above it. The familiar outward-above and inward-below pattern therefore depends on which column is warmer. The sign convention and temperature relationship should be stated explicitly rather than inferred from the word neutral.[8]
Fire-generated buoyancy produces an analogous pressure difference through the reduced density of hot gases. A hot compartment can drive smoke through wall leakage or into openings above it. As the smoke travels, heat loss and dilution reduce its temperature and usually reduce the buoyancy contribution. The pressure distribution across a connected building can therefore involve both the fire's heating and the building's pre-existing stack conditions.[9]
Klote identifies expansion, wind and ventilation systems as additional driving forces. The simple hydrostatic temperature relation isolates one contribution; it does not combine all of them automatically. For example, hot fire gases can continue upward despite reverse stack conditions affecting cooler building air. A neutral-plane interpretation must therefore be associated with the pressure model and boundary conditions actually being used.[10]
For a building with reasonably uniform leakage distribution, the neutral plane may lie near its middle height. Klote specifically qualifies that approximation: uneven leakage or vented shafts can move the level substantially. Assuming a middle-height plane without examining the distribution of openings can therefore give the wrong pressure difference at a particular floor. The location is part of the flow balance, not a fixed geometric feature.[2]
Pressure relation and symbols
At standard atmospheric pressure, Klote gives the following relation for pressure difference due to the temperature contrast. Here the inside and outside temperatures represent the gas columns being compared, and height is measured relative to their neutral plane. Positive pressure difference means that pressure inside exceeds pressure outside.[11]
| Symbol | Meaning | Unit |
|---|---|---|
| ΔP | Inside pressure minus outside pressure | Pa |
| Toutside | Absolute temperature of the external or adjoining gas column | K |
| Tinside | Absolute temperature of the internal gas column | K |
| h | Signed vertical distance above the neutral plane | m |
| 3460 | Coefficient for the stated units and standard atmospheric pressure | Pa·K/m |
Absolute temperatures are required because the density relationship follows the ideal gas law. Substituting Celsius temperatures into the reciprocal-temperature terms would change the physics and can produce meaningless results. The distinction between height above the floor and height above the neutral plane is equally important. A fixed point in a room changes its value of h if the neutral plane moves.[11]
The relation gives a linear pressure-difference profile when each compared gas column has a uniform temperature. A stratified compartment requires the relevant densities over the different height intervals. In a zone model, changes of layer density lead to corresponding changes in the pressure profile. The neutral level is then located within that piecewise description rather than imposed by one temperature for the complete enclosure.[5]
A pressure difference does not directly specify a total mass flow. CFAST combines it with the upstream density, opening width and discharge coefficient, then integrates the flow over height. Its wall-vent formulation uses a coefficient of 0.7 and separates intervals with different flow directions. This connects the neutral plane to a measurable exchange of mass without assuming a uniform velocity over the full opening.[12]
Worked pressure examples
Consider Klote's fire-compartment example with internal gases at 800 °C and surroundings at 20 °C. Converting to absolute temperatures gives 1,073.15 K and 293.15 K. The temperature-dependent coefficient in the pressure relation is approximately 8.579 Pa per metre above the neutral plane. At a point 1.52 m above that level, the calculated pressure difference is 13.04 Pa, consistent with the source's rounded 13 Pa.[13]
At the neutral plane itself, the calculated pressure difference is zero. Extending the same uniform-column illustration to a point 1.52 m below the plane gives −13.04 Pa. The sign reverses because the signed height reverses. This symmetry belongs to the stated idealised temperature profile and should not be assumed across a real compartment with different upper and lower temperatures.[11][5]
The example also illustrates the importance of the reference level. If the neutral plane were 0.50 m higher while the temperatures remained unchanged, the original observation point would lie 1.02 m above it. Its calculated pressure difference would then be 8.75 Pa. This is an illustrative sensitivity calculation, not a prediction that any particular opening change raises the neutral plane by that amount.[11][2]
For a separate building example, Klote considers a 60 m shaft with a middle-height neutral plane. Using the stated rounded Celsius inputs of 21 °C inside and −18 °C outside gives approximately 53.94 Pa at the top. The bottom has the opposite sign under the same assumptions. The source quotes about 55 Pa using its rounded alternative unit inputs, so the values agree at the precision intended.[14]
These two examples show why temperature difference alone is insufficient. The tall shaft develops a substantial pressure difference through its long vertical extent despite a much smaller temperature contrast. The hot-room example develops its pressure difference over a short distance. In both cases, the neutral-plane location determines which part of that vertical extent is relevant at the point being assessed.[13][14]
Application in smoke-flow calculations
Equal mass flow rates do not imply equal volume flow rates when the temperatures differ. Heated outgoing gas has a lower density and can occupy a larger volume than the incoming cooler stream. This is why a mass-balanced opening need not have equal inlet and outlet areas or velocities. The density used for each direction must belong to the gas approaching the opening on that side.[4][12]
A useful vent calculation identifies the spaces on each side, their layer temperatures and densities, and the opening geometry. It then determines the pressure difference over the opening height. Intervals are separated where a layer interface changes the gas properties or where the pressure difference changes sign. This is the basis of the wall-vent method described in the CFAST technical reference.[5]
The destination of a stream matters as well as its direction. Gas flowing from an upper layer into a cooler lower layer of an adjoining space can rise and entrain additional air. CFAST represents that redistribution through a spill-plume treatment. Thus, determining that smoke leaves one compartment is only the first step in predicting where it accumulates in the next compartment.[15]
Opening effectiveness and layer position remain separate considerations. Van Rossum distinguishes the geometric opening area from aerodynamic area and notes that an exhaust can draw relatively clean air when the layer is too shallow. A pressure-based designation as an outlet does not by itself prove efficient removal from the hot layer. The flow path and the gas layer actually supplying that opening must also be identified.[4]
At building scale, leakage between floors affects how the pressure pattern translates into smoke spread. Klote distinguishes cases with negligible inter-floor leakage from those with substantial paths between storeys. The same general stack direction can consequently produce different contamination patterns. A neutral plane describes the driving pressure relationship; the available paths determine which spaces that relationship actually connects.[9]
Finally, the wall-opening picture should not be transferred unchanged to a ceiling opening or a mechanical exhaust. CFAST uses separate treatments for horizontal vents and forced flows, including density-driven exchange through ceiling openings. The neutral plane remains a useful pressure reference, but its interpretation must match the geometry and driving mechanism of the flow being calculated.[16]
See also
Notes
- ↑ 1.0 1.1 Klote (2002), pp. 4-274–4-276; Van Rossum (2027), §9.3.
- ↑ 2.0 2.1 2.2 2.3 Klote (2002), pp. 4-275–4-276; Peacock et al. (2026), pp. 19–20.
- ↑ Klote (2002), pp. 4-274–4-276.
- ↑ 4.0 4.1 4.2 Van Rossum (2027), §9.3.
- ↑ 5.0 5.1 5.2 5.3 5.4 5.5 5.6 5.7 5.8 Peacock et al. (2026), pp. 19–20, §4.1.
- ↑ 6.0 6.1 Peacock et al. (2026), pp. 19–20, Figure 4.1; Walton and Thomas (2002), pp. 3-172–3-174.
- ↑ 7.0 7.1 Walton and Thomas (2002), p. 3-174.
- ↑ 8.0 8.1 Klote (2002), pp. 4-274–4-275.
- ↑ 9.0 9.1 Klote (2002), p. 4-276.
- ↑ Klote (2002), pp. 4-274–4-276.
- ↑ 11.0 11.1 11.2 11.3 Klote (2002), pp. 4-274–4-276, equations 1–2.
- ↑ 12.0 12.1 Peacock et al. (2026), p. 19, equations 4.1–4.5.
- ↑ 13.0 13.1 Klote (2002), p. 4-276.
- ↑ 14.0 14.1 Klote (2002), p. 4-275.
- ↑ Peacock et al. (2026), pp. 20–21.
- ↑ Peacock et al. (2026), pp. 21–22, §§4.2–4.3.
References
- Klote, J. H. (2002). Smoke control. In P. J. DiNenno (Ed.), SFPE handbook of fire protection engineering (3rd ed., pp. 4-274–4-291). National Fire Protection Association.
- Peacock, R. D., McGrattan, K. B., Forney, G. P., & Reneke, P. A. (2026). CFAST: Consolidated fire and smoke transport (version 7). Volume 1: Technical reference guide (NIST Technical Note 1889v1, revision 7.7.7). National Institute of Standards and Technology.
- Van Rossum, E. (2027). Introduction to Fire Safety Engineering.
- Walton, W. D., & Thomas, P. H. (2002). Estimating temperatures in compartment fires. In P. J. DiNenno (Ed.), SFPE handbook of fire protection engineering (3rd ed., pp. 3-171–3-188). National Fire Protection Association.