In this chapter, appliance refers to any furnace, boiler, or incinerator (including the burner). Unless the context indicates otherwise, the term chimney includes specialized vent products such as masonry, metal, and factory-built chimneys; single-wall metal pipe; type B gas vents; special gas vents; or masonry chimney liner systems (NFPA Standard 211). Draft is negative static pressure, measured relative to atmospheric pressure; thus, positive draft is negative static pressure. Flue gas is the mixture of gases discharged from the appliance and conveyed by the chimney or vent system.

Appliances can be grouped by draft conditions at the appliance flue gas outlet as follows (Stone 1971):

In the first two configurations, hot flue gas buoyancy, induced-draft chimney fans, or a combination of both produces draft. The third configuration may not require chimney draft, but it should be considered in the design if a chimney is used. If the chimney system is undersized, draft inducers in the connector or chimney may supply draft needs. If the connector or chimney pressure requires control for proper operation, draft control devices must be used.

Available draft D_a is the draft supplied by the vent system, available at the appliance flue gas outlet. It can be shown as

(1)

where

This equation can account for a nonneutral (nonzero) pressure difference between the space surrounding the appliance or fireplace and the atmosphere. If the surrounding space is at a lower pressure than the atmosphere (space depressurized), the pressure difference D_p should also be subtracted from D_t when calculating available draft D_a, and vice versa. This equation applies to all three appliance draft conditions at the vent system inlets; for example, in the second condition with zero draft requirement at the appliance outlet, available draft required is zero, so theoretical draft of the chimney equals the flow resistance, if no depressurization or boost is present.

Satisfactory operation of the appliance and vent system depends on D_a staying within limits. If D_a is too small and the appliance design relies on D_a to draw combustion air into the burner system, incomplete combustion, flame rollout, and/or flue gas spillage can occur at the appliance. Even where combustion air is mechanically supplied to the burner, inadequate D_a can result in portions of the appliance and/or vent system operating under positive pressure. If those sections are not designed to be pressurized, flue gas can leak into the building.

Excessive D_a can cause excessive combustion air to be drawn into the burner system, which not only reduces combustion efficiency, but can also cause flame lifting or other detriments to combustion performance. Many vent systems limit D_a using a draft hood or draft regulator (integral to the appliance or field supplied); these devices limit draft by admitting dilution air into the vent system. The resulting increased flow through the vent system increases ΔP and reduces D_a. Such draft control devices are unsuitable for use in parts of vent systems that may operate with a negative D_a (positive pressure).

Various methods can be used identify the D_a limits within which an appliance can properly operate. Often, the appliance manufacturer publishes a D_a or range of D_as where the appliance is designed to operate. In other cases, acceptable values for D_a are not explicitly specified, but the manufacturer or installation code prescribes a vent system design in terms of length, chimney height, cross-sectional area, etc, that is known to provide the required D_a. An example of the latter approach is that for Category I gas appliances (see the section on Gas Appliance Venting).

Theoretical draft D_t is the natural draft produced by the buoyancy of hot gases in the chimney relative to cooler gases in the surrounding atmosphere. It depends on chimney height, local barometric pressure, and the mean chimney flue gas temperature difference Δt_m, which is the difference in temperature between the flue gas and atmospheric gases. Therefore, cooling by heat transfer through the chimney wall is a key variable in chimney design. Precise evaluation of theoretical draft is not necessary for most design calculations because of the availability of chimney design charts, computer programs, and capacity tables in the references, building codes, and vent and appliance manufacturers’ data sheets.

Chimney temperatures and acceptable combustible material temperatures must be known in order to determine safe clearances between the chimney and combustible materials. Safe clearances for some chimney systems, such as type B gas vents, are determined by standard tests and/or specified in building codes.

Losses from Flow Δp represent the friction losses imposed on the flue gas by flow resistance through the chimney.

Depressurization D_p is negative pressure in the space surrounding the appliance with respect to the atmosphere into which the chimney discharges. D_p can be caused by other appliances and fans operating in the building that remove or add air to the space surrounding the appliance, by building stack effect, by outdoor atmospheric effects such as wind impacting the chimney exit or the side of the building facing or leeward to the wind, and other building phenomena.

Boost D_b is the pressure boost from a mechanical-draft fan. A chimney with an forced-draft fan at the inlet of the chimney would have positive boost (increased static pressure). A chimney with an induced-draft fan at the outlet of the chimney would have negative boost (decreased static pressure).

The following sections cover the basis of chimney design for average steady-state appliance operating conditions, where significant condensation of flue products is not expected. For other appliance operating conditions, a rigorous cyclic evaluation of the flue gas and material surface temperatures in the chimney vent system can be obtained using the VENT-II computer program (GTI 2009). For oil-fired appliances, chimney flue gas and material surface temperature evaluations can be obtained using the OHVAP computer program (Krajewski 1996).

The proper chimney can be selected by evaluating factors such as draft, configuration, size, and operating conditions of the appliance; construction of surroundings; appliance usage classification; residential, low, medium, or high heat (NFPA Standard 211); and building height. The chimney designer should know the applicable codes and standards to ensure acceptable construction.

In addition to chimney draft, the following factors must be considered for safe and reliable operation: adequate air supply for combustion; building depressurization effects; draft control devices; chimney materials (corrosion and temperature resistance); flue gas temperatures, composition, and dew point; wind eddy zones; and particulate dispersion. Chimney materials must resist oxidation and condensation at both high and low fire levels at all design temperatures.

Chimney design balances forces that produce flow against those that retard flow (e.g., friction). Theoretical draft is the pressure that produces flow in gravity or natural-draft chimneys. It is defined as the static pressure resulting from the difference in densities between a stagnant column of hot flue gas and an equal column of ambient air. In the design or balancing process, theoretical draft may not equal friction loss because the appliance is frequently built to operate with some specific pressure (positive or negative) at the appliance flue gas exit. This exit pressure is added to the available draft, which depends on chimney conditions, appliance operating characteristics, fuel, and type of draft control.

Flow losses caused by friction may be estimated by several formulas for flow in pipes or ducts, such as the equivalent length method or the loss coefficient (velocity head) method. Chapter 13 of the 2017 ASHRAE Handbook—Fundamentals covers computation of flow losses. This chapter emphasizes the loss coefficient method because fittings usually cause the greater portion of system pressure drop in chimney systems, and conservative loss coefficients (which are almost independent of piping size) provide an adequate basis for design.

A computer program called VENT-II (GTI 2009) dynamically predicts cyclic flows, temperatures, and pressures in venting systems. Similarly, Krajewski (1996) developed a computer program entitled OHVAP: Oil-Heat Vent Analysis Program.

For large gravity chimneys, steady-state available draft D_a may be calculated from Equation (2) or (3). Both equations use the equivalent length approach, as indicated by the symbol L_e in the flow-loss term (ASHVE 1941). These equations permit consideration of the density difference between chimney gases and ambient air and compensation for shape factors. Mean flue gas temperature T_m must be estimated separately. Starting with Equation (1), the following equations are based on zero depressurization at the appliance flue gas outlet and zero chimney draft boost.

For a cylindrical chimney,

(2)

For a rectangular chimney,

(3)

where

In these equations, the first term determines theoretical draft, based on applicable gas and ambient density. The second term defines draft loss based on the factors for flow in a circular or rectangular duct system. To use these equations, the mass flow w for a variety of fuels and situations and the available draft needs of various types of appliances must be determined.

Equations (2) and (3) may be derived and expressed in a form that is more readily applied to the problems of chimney design, size, and capacity by considering the following factors, which are the steps used to solve the problems in the section on Vent and Chimney Capacity Calculation Examples:

For application to system design, the chimney gas velocity step is eliminated; however, actual velocity can be found readily, if needed.

Table 1 Mass Flow Equations for Common Fuels

Fuel	Ratio M of Mass Flow to Inputa
	b
Natural gas
LPG (propane, butane, or mixture)
No. 2 oil (light)
No. 6 oil (heavy)
Bituminous coal (soft)
Type 0 waste or wood
a Percent CO2 is determined in combustion products with water condensed (dry basis). b Total combustion products include combustion products and excess air.

 Mass Flow of Combustion Products in Chimneys and Vents

Mass flow in a chimney or venting system may differ from that in the appliance, depending on the type of draft control or number of appliances operating in a multiple-appliance system. Mass flow is preferred to volumetric flow because it remains constant in any continuous portion of the system, regardless of changes in temperature or pressure. For the chimney gases resulting from any combustion process, mass flow can be expressed as

(4)

where

w	=	mass flow rate, lb/h
I	=	appliance heat input, Btu/h
M	=	ratio of mass flow to heat input, lb of combustion products per 1000 Btu of fuel burned (M depends on fuel composition and percentage excess air [or CO2] in the chimney)

Volumetric flow rate (ft³/min) Q can be found as follows:

(5)

where ρ_m is gas density (lb/ft³).

Figure 1. Graphical Evaluation of Rate of Vent Gas Flow from Percent CO₂ and Fuel Rate

In chimney system design, the composition and flow rate of the flue gas must be assumed to determine the ratio M of mass flow to input. When combustion conditions are given in terms of excess air, Figure 1 can be used to estimate CO₂. The mass flow equations in Table 1 illustrate the influence of fuel composition; however, additional guidance is needed for system design. The information provided with many heat-producing appliances is limited to whether they have been tested, certified, listed, or approved to comply with applicable standards. From this information and from the type of fuel and draft control, certain inferences can be drawn regarding the flue gas.

Table 2 suggests typical values for the vent or chimney systems for gaseous and liquid fuels when specific outlet conditions for the appliance are not known. If a gas appliance with draft hood is used, Table 2 recommends that dilution air through the draft hood reduce the CO₂ percentage to 5.3%. For appliances using draft regulators, the dilution and temperature reduction is a function of the draft regulator gate opening, which depends on excess draft. If the chimney system produces the exact draft necessary for the appliance, little dilution takes place.

Table 2 Typical Chimney and Vent Design Conditions^a

Fuel	Appliance	% CO2	Temperature Rise, °F	M (Mass Flow/Input), lb Total Flue Gasb per 1000 Btu Fuel Input	Flue Gas Density,c lb/ft3	Flow Rate/Unit Heat Input,ccfm per 1000 Btu/h at Flue Gas Temperature
Natural gas	Draft hood	5.3	300	1.54	0.0483	0.530
Propane gas	Draft hood	6.0	300	1.59	0.0483	0.547
Natural gas
Low-efficiency	No draft hood	8.0	400	1.06	0.0431	0.409
High-efficiency	No draft hood	7.0	240	1.19	0.0522	0.381
No. 2 oil	Residential	9.0	500	1.24	0.0389	0.531
Oil	Forced-draft, over 400,000 Btu/h	13.5	300	0.85	0.0483	0.295
Waste, Type 0	Incinerator	9.0	1340	1.62	0.0213	1.267
a Values are for appliances with flue losses of 17% or more. For appliances with lower flue losses b Total flue gas includes combustion products and excess air. (high-efficiency types), see appliance installation instructions or ask manufacturer for operating data. c At sea level and 60°F ambient temperature.

For manifolded gas appliances that have draft hoods, dilution through draft hoods of inoperative appliances must be considered in precise system design. However, with forced-draft appliances having wind box or inlet air controls, dilution through inoperative appliances may be unimportant, especially if pressure at the outlet of inoperative appliances is neutral (atmospheric level).

Figure 2 can be used to estimate mass and volumetric flow. Flow conditions in the chimney connector, manifold, vent, and chimney vary with configuration and appliance design and are not necessarily the same as boiler or appliance outlet conditions.

Mass flow in incinerator chimneys must account for the probable heat value of the waste, its moisture content, and use of additional fuel to initiate or sustain combustion. Classifications of waste and corresponding values of M in Table 3 are based on recommendations from the Incinerator Institute of America. Combustion data given for types 0, 1, and 2 waste do not include any additional fuel. Where constant burner operation accompanies waste combustion, the additional quantity of products should be considered in the chimney design.

The system designer should obtain exact outlet conditions for the maximum rate operation of the specific appliance. This information can reduce chimney construction costs. For appliances with higher seasonal or steady-state efficiencies, however, special attention should be given to the manufacturer’s venting recommendations because the flue gas may differ in composition and temperature from conventional values.

Table 3 Mass Flow for Incinerator Chimneys

Type of Waste	Heat Value of Waste,a Btu/lb	Auxiliary Fuela per Unit Waste, Btu/lb	Combustion Products
			cfm/lb Waste at 1400°Fb	lb per lb Wastec	M, lb Products per 1000 Btu
0	8500	0	10.74	13.76	1.62
1	6500	0	8.40	10.80	1.66
2	4300	0	5.94	7.68	1.79
3	2500	1500	4.92	6.25	2.50
4	1000	3000	4.14	5.33	5.33
a Auxiliary fuel may be used with any type of waste, depending on incinerator design. b Specialized units may produce higher or lower outlet gas temperatures, which must be considered in chimney sizing, using Equation (14), Equation (23), or Figure 7. c Multiply these values by pounds of waste burned per hour to establish mass flow.

 Mean Chimney Gas Temperature and Density

Chimney gas temperature depends on the fuel, appliance, draft control, chimney size, and configuration.

Density of gas within the chimney and theoretical draft both depend on gas temperature. Although the gases flowing in a chimney system lose heat continuously from entrance to exit, a single mean gas temperature must be used in either the design equation or the chart. Mean chimney gas density is virtually the same as air density at the same temperature. Thus, density may be found as

(6)

where

ρm	=	chimney gas density, lb/ft3
Ts	=	standard temperature = 518.67°R
ρa	=	air density at Ts and Bo = 0.0765 lb/ft3
B	=	local barometric pressure, in. Hg
Tm	=	mean chimney gas temperature at average system conditions, °R
Bo	=	standard pressure = 29.92 in. Hg

The density ρ_a in (Equation 6) is a compromise value for typical humidity. The subscript m for density and temperature requires that these properties be calculated at mean gas temperature or vertical midpoint of a system (inlet conditions can be used where temperature drop is not significant).

Using a reasonably high ambient (e.g., 60°F) for design ensures improved operation of the chimney when ambient temperatures drop because temperature differentials and draft increase.

Table 4 Mean Chimney Gas Temperature for Various Appliances

Appliance Type	Mean Temperature tm* in Chimney, °F
Natural gas-fired heating appliance	360
with draft hood (low-efficiency)
LP gas-fired heating appliance with	360
draft hood (low-efficiency)
Gas-fired heating appliance, no draft hood
Low-efficiency	460
Mid-efficiency	300
Oil-fired heating appliance (low-efficiency)	560
Conventional incinerator	1400
Controlled air incinerator	1800 to 2400
Pathological incinerator	1800 to 2800
Turbine exhaust	900 to 1400
Diesel exhaust	900 to 1400
Ceramic kiln	1800 to 2400
* Subtract 60°F ambient to obtain temperature rise for use with Figure 7.

A design requires assuming an initial or inlet chimney gas temperature. In the absence of specific data, Table 4 provides a conservative temperature. For appliances that can operate over a range of temperatures, size should be calculated at both extremes to ensure an adequate chimney.

The drop in vent gas temperature from appliance to exit reduces capacity, particularly in sizes of 12 in. or less. In gravity type B gas vents, which may be as small as 3 in. in diameter, and in other systems used for venting gas appliances, capacity is best determined from NFPA Standard 54/AGA Z223.1 or CSA Standard B149.1. In these codes, the tables compensate for the particular characteristics of the chimney material involved, except for very high single-wall metal pipe. Between 12 and 18 in. diameters, the effect of heat loss diminishes greatly because there is greater gas flow relative to system surface area. For 20 in. and greater diameters, cooling has little effect on final size or capacity.

A straight vertical vent or chimney directly off the appliance requires little compensation for cooling effects, even with smaller sizes. However, a horizontal connector running from the appliance to the base of the vent or chimney has enough heat loss to diminish draft and capacity. Figure 3 is a plot of temperature correction C_u, which is a function of connector size, length, and material for either conventional single-wall metal connectors or double-wall metal connectors.

Figure 2. Flue Gas Mass and Volumetric Flow

Figure 3. Temperature Multiplier C_u for Compensation of Heat Losses in Connector

For a more accurate method of calculating temperature loss in vents and chimneys, including effects from condensation, see Rutz et al. (1991).

To use Figure 3, estimate connector size and length and read the temperature multiplier. For example, 16 ft of 7 in. diameter single-wall connector has a multiplier of 0.61. If inlet temperature rise Δt_e above ambient is 300°F, operating mean temperature rise Δt_m will be 0.61 × 300 = 183°F. This factor adequately corrects the temperature at the midpoint of the vertical vent for heights up to 100 ft.

The correction procedure includes the assumption that the overall heat transfer coefficient of a vertical chimney is approximately 0.6 Btu/h·ft²·°F or less (the value for double-wall metal). This procedure does not correct for cooling in very high stacks constructed entirely of single-wall metal, especially those exposed to cold ambient temperatures. For severe exposures or excessive heat loss, a trial calculation assuming a conservative operating temperature shows whether capacity problems will be encountered.

Table 5 Overall Heat Transfer Coefficients of Various Chimneys and Vents

Material		U, Btu/h·ft2·°F*			Remarks
		Observed	Design
Industrial steel stacks		—	1.3		Under wet wind
Clay or iron sewer pipe		1.3 to 1.4	1.3		Used as single-wall material
Asbestos-cement gas vent		0.72 to 1.42	1.2		Tested per UL Standard 441
Black or painted steel stove pipe		—	1.2		Comparable to weathered galvanized steel
Single-wall galvanized steel		0.31 to 1.38	1.0		Depends on surface condition and exposure
Single-wall unpainted pure aluminum		—	1.0		No. 1100 or other bright-surface aluminum alloy
Brick chimney, tile-lined		0.5 to 1.0	1.0		For gas appliances in residential construction per NFPA Standard 211
Double-wall gas vent, 1/4 in. air space Double-wall gas vent, 1/2 in. air space		0.37 to 1.04 0.34 to 0.7	0.6 0.4		Galvanized steel outer pipe, pure aluminum inner pipe; tested per UL Standard 441
Insulated prefabricated chimney		0.34 to 0.7	0.3		Solid insulation meets UL Standard 103 when chimney is fully insulated
* U-factors based on inside area of chimney.

For more precise heat loss calculations, Table 5 suggests overall heat transfer coefficients for various constructions installed in typical environments at usual flue gas flow velocities (Segeler 1965). For masonry, any additional thickness beyond the single course of brick plus tile liner used in residential chimneys decreases the coefficient.

(Equations 7) and (8) describe flow and heat transfer, respectively, in a venting system with D_a = 0.ρ_m

(7)

(8)

where

A	=	area of passage cross section, ft2
cp	=	specific heat
di	=	inside diameter, ft
exp x	=	ex
g	=	gravitational constant = 32.1740 ft/s2
H	=	height of chimney above grade or inlet, ft
k	=	fixed, dimensionless system flow resistance coefficient for pipe and fittings
Lm	=	length from inlet to location (in the vertical) of mean gas temperature, ft
q	=	heat flow rate at vent inlet, Btu/h
qm	=	heat flow rate at midpoint of vent, Btu/h
Δte	=	(T – To) = temperature difference entering system, °F
Δtm	=	(Tm – To) = chimney gas mean temperature rise, °F
T	=	flue gas temperature at vent inlet, °F
Tm	=	chimney mean flue gas temperature, °R
To	=	design ambient temperature, 520°R
Ts	=	518.67°F
	=	heat transfer coefficient, Btu/h·ft2·°F
ρm	=	mean flue gas density from Equation (6)

Table 6 Approximate Theoretical Draft of Chimneys

Vent Gas Temperature Rise, °F	Dt per 100 ft, in. of water
100	0.2
150	0.3
200	0.4
300	0.5
400	0.6
500	0.7
600	0.8
800	0.9
1100	1.0
1600	1.1
2400	1.2
Notes: Ambient temperature = 60°F = 520°R Chimney gas density = air density Sea-level barometric pressure = 29.92 in. Hg Equation (9) may be used to calculate exact values for Dt at any altitude.

Assuming reasonable constancy of , the overall heat transfer coefficient of the venting system material, (Equations 7) and (8) provide a solution for maximum vent gas capacity. They can also be used to develop cooling curves or calculate the length of pipe where internal moisture condenses. Kinkead (1962) details methods of solution and application to both individual and combined gas vents.

 Theoretical Draft

The theoretical draft of a gravity chimney or vent is the difference in weight between a given column of warm (light) chimney gas and an equal column of cold (heavy) ambient air. Chimney gas density or temperature, chimney height, and barometric pressure determine theoretical draft; flow is not a factor. The equation for theoretical draft assumes chimney gas density is the same as that of air at the same temperature and pressure; thus,

(9)

where

B	=	local barometric pressure, in. Hg
Dt	=	theoretical draft, in. of water
H	=	height of chimney above grade or inlet, ft
Tm	=	mean flue gas temperature at average conditions in system, °R
To	=	ambient temperature, °R

Theoretical draft thus increases directly with height and with the difference in density between the hot and cold columns.

Theoretical draft D_t is always positive (unless chimney gases are colder than ambient air). Equation (9) for theoretical draft is the basis for Figure 4, which can be used up to 1000°F and 7000 ft elevation. For example, using Figure 4, ambient air temperature at 40°F, 4000 ft above sea level (25.85 in. Hg), and mean chimney gas temperature at 350°F provides D₁₀₀ draft of 0.5 in. of water per 100 ft of height. For D₁₀₀ draft at 25 in. Hg barometric pressure or 4900 ft above sea level, follow the intersection to the pivot line and read the new D₁₀₀ draft of 0.48 in. of water.

Figure 4. Theoretical Draft Nomograph

Table 7 Altitude Correction

Altitude, ft	Barometric Pressure B, in. Hg	Factor*
Sea level	29.92	1.00
2,000	27.82	1.08
4,000	25.82	1.16
6,000	23.98	1.25
8,000	22.22	1.35
10,000	20.58	1.45
* Multiply operating input by factor to obtain design input.

Theoretical draft should be estimated and included in system calculations, even for appliances producing considerable positive outlet static pressure, to achieve the economy of minimum chimney size. (Equation 9) may be used directly to calculate exact values for theoretical draft at any altitude. For ease of application and consistency with Figure 7, Table 6 lists approximate theoretical draft for typical gas temperature rises above 60°F ambient.

Appliances with fixed fuel beds, such as hand-fired coal stoves and furnaces, require positive available draft (negative gage pressure). Small oil heaters with pot-type burners, as well as residential furnaces with pressure-atomizing oil burners, need positive available draft, which can usually be set by following the manufacturer’s instructions for setting the draft regulator. Available draft requirements for larger packaged boilers or appliances assembled from components may be negative, zero (neutral), or positive.

Compensation of theoretical draft for altitude or barometric pressure is usually necessary for appliances and chimneys functioning at elevations greater than 2000 ft. Depending on the design, one of the following approaches to pressure or altitude compensation is necessary for chimney sizing.

1. Figure 7: Use sea-level theoretical draft.

2. Equation (21): Use local theoretical draft with actual energy input, or use sea-level theoretical draft with energy input multiplied by ratio of sea level to local barometric pressure (Table 7 factor).

3. Equation (23): Use local theoretical draft and barometric pressure with volumetric flow at the local density.

The altitude correction multiplier for input (Table 7) is the only method of correcting to other elevations. Reducing theoretical draft imposes an incorrect compensation on the chart.

Gas appliances with draft hoods, for example, are usually derated 4% per 1000 ft of elevation above sea level when they are operated at 2000 ft altitude or above (see NFPA Standard 54/AGA Z223.1 or CSA Standard B149.1). The altitude correction factor derates the design input so that the vent size at altitude for derated gas appliances is effectively the same as at sea level. For other appliances where burner adjustments or internal changes might be used to adjust for reduced density at altitude, the same factors produce an adequately compensated chimney size. For example, an appliance operating at 6000 ft elevation at 10 × 10⁶ Btu/h input, but requiring the same draft as at sea level, should have a chimney selected on the basis of 1.244 times the operating input, or 12.5 × 10⁶ Btu/h.

 System Pressure Loss Caused by Flow

In any chimney system, flow losses, expressed as pressure drop Δp in inches of water, are the difference between theoretical and available draft:

(10)

where Δp is always positive.

In any chimney or vent system, flow losses resulting from velocity and resistance can be determined from the Bernoulli equation:

(11)

where

k	=	dimensionless system resistance coefficient of piping and fittings
V	=	system gas velocity at mean conditions, fps
g	=	gravitational constant = 32.1740 ft/s2
ρm	=	mean flue gas density, lb/ft3

Pressure losses are thus directly proportional to the resistance factor and to the square of the velocity.

 Available Draft

Starting with Equation (1), the difference between theoretical draft and flow losses without depressurization or boost is

Available draft D_a can therefore be defined as

(12)

See Equations (2) and (3) for geometrically defined versions.

Available draft D_a can be negative, zero, or positive. The pressure difference Δp, or theoretical minus available draft, overcomes the flow losses. Table 8 lists the pressure components for three draft configurations. Table 8 applies to still air (no wind) conditions and a neutral (zero) pressure difference between the space surrounding the appliance or fireplace and the atmosphere. Columns with and without boost are provided.

Table 8 Pressure Equations for Δp

Required Appliance Outlet Pressure or Available Draft Da	Δp Equationa
	Gravity Only	Gravity plus Inducerb
Negative, needs positive draft	Δp = Dt – Da	Δp = Dt – Da + Db
Zero, vent with draft hood or balanced forced draft	Δp = Dt	Δp = Dt + Db
Positive, causes negative draft	Δp = Dt + Da	Δp = Dt + Da + Db
a Equations use absolute pressure for Da. b Db = static pressure boost of inducer at flue gas temperature and rated flow.

The effect of a nonneutral pressure difference on capacity or draft may be included by imposing a static pressure (either positive or negative). One way to circumvent a space-to-atmosphere pressure difference is to use a sealed-combustion system or direct-vent system (i.e., all combustion air is taken directly from the outdoor atmosphere, and all flue gas is discharged to the outdoor atmosphere with no system openings such as draft hoods). The effect of wind on capacity or draft may be included by imposing a static pressure (either positive or negative) or by changing the vent terminal resistance loss. However, a properly designed and located vent terminal should cause little change in Δp at typical wind velocities.

Although small static draft pressures can be measured at the entrance and exit of gas appliance draft hoods, D_a at the appliance is effectively zero. Therefore, all theoretical draft energy produces chimney flow velocity and overcomes chimney flow resistance losses.

 Chimney Gas Velocity

Velocity in a chimney or vent varies inversely with flue gas density ρ_m and directly with mass flow rate. The equation for flue gas velocity at mean gas temperature in the chimney is

(13)

where

V	=	flue gas velocity, fps
di	=	inside diameter, in.
ρm	=	mean flue gas density, lb/ft3
w	=	mass flow rate, lb/h

To express velocity as a function of input and chimney gas composition, w in (Equation 13) is replaced by using (Equation 4):

(14)

Thus, chimney velocity depends on the product of heat input I and the ratio M of mass flow to input.

The input capacity or diameter of a chimney may usually be found without determining flow velocity. Internal or exit velocity must occasionally be known, to ensure effluent dispersal or avoid flow noise. Also, the flow velocity of incinerator chimneys, turbine exhaust systems, and other appliances with high outlet pressures or velocities is needed to estimate piping loss coefficients.

Equations (11), (13), (14), and (22) can be applied to find velocity. The velocity may also be calculated by dividing volumetric flow rate Q by chimney area A. A similar calculation may be performed when the energy input is known. For example, a 34 in. chimney serving a 10 × 10⁶ Btu/h natural gas appliance, at 8.5% CO₂ and 300°F above ambient in the chimney, produces (from Figure 2) 0.36 cfm per 1000 Btu/h. Chimney gas flow rate is

(15)

Dividing by area to obtain velocity, V = 3600/6.305 = 571 fpm

Chimney gas velocity affects the piping friction factor k_L and roughness correction factor. The section on Resistance Coefficients has further information, and Example 3 illustrates how these factors are used in the velocity equations.

Chimney systems can operate over a wide range of velocities, depending on modulation characteristics of the burner system or the number of appliances in operation. Typical velocity in vents and chimneys ranges from 300 to 3000 fpm. A chimney design developed for maximum input and maximum velocity should be satisfactory at reduced input because theoretical draft is roughly proportional to flue gas temperature rise, whereas flow losses are proportional to the square of the velocity. Thus, as input is reduced, flow losses decrease more rapidly than system motive pressures.

Table 9 Resistance Loss Coefficients

Component	Suggested Design Value, Dimensionless*	Estimated Span and Notes
Inlet acceleration (k1)
Gas vent with draft hood	1.5	1.0 to 3.0
Barometric regulator	0.5	0.0 to 0.5
Direct connection	0.0	Also dependent on blocking damper position
Round elbow (k2)
90°	0.75	0.5 to 1.5
45°	0.3	—
Tee or 90° connector (k3)	1.25	1.0 to 4.0
Y connector	0.75	0.5 to 1.5
Cap, top (k4)
Open straight	0.0	—
Low-resistance (UL)	0.5	0.0 to 1.5
Other	—	1.5 to 4.5
Spark screen	0.5	—
Converging exit cone	(di1 / di2)4 – 1	System designed using di1
Tapered reducer (di1 to di2)	1 – (di2 / di1)4 – 1	System designed using di2
Increaser	—	See Chapter 3, 2017 ASHRAE Handbook—Fundamentals.
Piping (kL)		Numerical coefficient (friction factor F) varies from 0.2 to 0.5; see Figure 13, Chapter 3, 2017 ASHRAE Handbook—Fundamentals for size, roughness, and velocity effects.
Note: For combined gravity gas vents serving two or more appliances (draft hoods), multiply total k (components + piping; see Equations [17], [18], and [19]) by 1.5 to obtain gravity system design coefficient. (This rule does not apply to forced- or induced-draft vents or chimneys.) * Initial assumption when size is unknown: k = 5.0 for entire system, for first trial; k = 7.5 for combined gas vents only

Effluent dispersal may occasionally require a minimum upward chimney outlet velocity, such as 3000 fpm. A tapered exit cone can best meet this requirement. For example, to increase the outlet velocity from the 34 in. chimney (A = 6.305 ft² area) from 1600 to 3000 fpm, the cone must have a discharge area of 6.305 × 1600/3000 = 3.36 ft² and a 24.8 in. diameter. An exit cone avoids excessive flow losses because the entire system operates at the lower velocity, and a resistance factor is only added for the cone. In this case, the added resistance for a gradual taper approximates the following (see Table 9):

(16)

Noise in chimneys may be caused by turbulent flow at high velocity or by combustion-induced oscillations or resonance. Noise is seldom encountered in gas vent systems or in systems producing positive available draft, but it may be a problem with forced-draft appliances. Turbulent flow noise can be avoided by designing for lower velocity, which may entail increasing the chimney size above the minimum recommended by the appliance manufacturer. Chapter 49 of the 2019 ASHRAE Handbook—HVAC Applications has more information on noise control.

 System Resistance Coefficient

The velocity head method of determining resistance losses assigns a fixed numerical coefficient (independent of velocity) or k factor to every fitting or turn in the flow circuit, as well as to piping.

The resistance coefficient k that appears in Equations (21) and (23) and Figure 7 summarizes the friction loss of the entire chimney system, including piping, fittings, and configuration or interconnection factors. Capacity of the chimney varies inversely with the square root of k, whereas diameter varies as the fourth root of k. The insensitivity of diameter and input to small variations in k simplifies design. Analyzing details such as pressure regain, increasers and reducers, and gas cooling junction effects is unnecessary if slightly high resistance coefficients are assigned to any draft diverters, elbows, tees, terminations, and, particularly, piping.

For any chimney with fittings, the total flow resistance is a constant plus variable piping resistance (a function of centerline length divided by diameter). Table 9 suggests moderately conservative resistance coefficients for common fittings. Elbow resistance may be lowered by long-radius turns; however, corrugated 90° elbows may have resistance values at the high end of the scale. Table 9 shows resistance as a function of inlet diameter d_i1 and outlet diameter d_i2.

System resistance k may be expressed as follows:

(17)

with

(18)

and

(19)

where

kf	=	fixed fitting loss coefficient
kL	=	piping resistance loss function (Figure 13, Chapter 3 of the 2017 ASHRAE Handbook—Fundamentals, adjusted for units)
k1	=	inlet acceleration coefficient
k2	=	elbow loss coefficient, n2 = number of elbows
k3	=	tee loss coefficient, n3 = number of tees
k4	=	cap, top, or exit cone loss coefficient
F	=	friction factor
L	=	length of all piping in chimney system, ft
di	=	inside diameter, in.

For combined gas vents using appliances with draft hoods, the summation k must be multiplied by a diversity factor of 1.5 (see Table 9 note and Example 5). This multiplier does not apply to forced- or induced-draft vents or chimneys.

When size is unknown, the following k values may be used to run a first trial estimate:

k	=	5.0 for the entire system
k	=	7.5 for combined gas vents only

The resistance coefficient method adapts well to systems in which the fittings cause significant losses. Even for extensive systems, an initial assumption of k = 5.0 gives a tolerably accurate vent or chimney diameter in the first trial solution. Using this diameter with the piping resistance function (Equation [19]) in a second trial normally yields the final answer.

The minimum system resistance coefficient in a gas vent with a draft hood is always 1.0 because all gases must accelerate through the draft hood from almost zero velocity to vent velocity.

For a system connected directly to the outlet of a boiler or other appliance where the capacity is stated as full-rated heat input against a positive static pressure at the chimney connection, minimum system resistance is zero, and no value is added for existing velocity head in the system.

For simplified design, a value of 0.4 for F in (Equation 19) applies for all sizes of vents or chimneys and for all velocities and temperatures. As diameter increases, this function becomes increasingly conservative, which is desirable because larger chimneys are more likely to be made of rough masonry construction or other materials with higher pressure losses. The 0.4 constant also introduces an increasing factor of safety for flow losses at greater lengths and heights.

Figure 5 is a plot of friction factor F versus velocity and diameter for commercial iron and steel pipe at a flue gas temperature of 300°F above ambient (Lapple 1949). The figure shows, for example, that a 48 in. diameter chimney with a flue gas velocity of 80 fps may have a friction factor as low as 0.2. In most cases, k_L = 0.3 L/d_i gives reasonable design results for chimney sizes 18 in. and larger because systems of this size usually operate at flue gas velocities greater than 10 fps.

(Lapple 1949)

Figure 5. Friction Factor for Commercial Iron and Steel Pipe

At 1000°F or over, the factors in Figure 5 should be multiplied by 1.2. Because Figure 5 is for commercial iron and steel pipe, an additional correction for greater or less surface roughness may be imposed. For example, the factor for a very rough 12 in. diameter pipe may be doubled at a velocity as low as 2000 fpm.

For most chimney designs, a friction factor F of 0.4 gives a conservative solution for diameter or input for all sizes, types, and operating conditions of prefabricated and metal chimneys; alternatively, F = 0.3 is reasonable if the diameter is 18 in. or more. Because neither input nor diameter is particularly sensitive to the total friction factor, the overall value of k requires little correction.

Masonry chimneys, including those lined with clay flue tile, may have rough surfaces, tile shape variations that cause misalignment, and joints at frequent intervals with possible mortar protrusions. In addition, the inside cross-sectional area of liner shapes may be less than expected because of local manufacturing variations, as well as differences between claimed and actual size. To account for these characteristics, the estimate for k_L should be on the high side, regardless of chimney size or velocity.

Computations should be made by assuming smooth surfaces and then adding a final size increase to compensate for shape factor and friction loss. Performance or capacity of metal and prefabricated chimneys is generally superior to that of site-constructed masonry.

 Configuration and Manifolding Effects

The most common configuration is the individual vent, stack, or chimney, in which one continuous system carries products from appliance to terminus. Other configurations include the combined vent serving a pair of appliances, the manifold serving several, and branched systems with two or more lateral manifolds connected to a common vertical system. As the number of appliances served by a common vertical vent or chimney increases, the precision of design decreases because of diversity factors (variation in the number of appliances in operation) and the need to allow for maximum and minimum input operation (Stone 1957). For example, the vertical common vent for interconnected gas appliances must be larger than for a single appliance of the same input to allow for operating diversity and draft hood dilution effects. Connector rise, headroom, and configuration in the equipment room must be designed carefully to avoid draft hood spillage and related oxygen depletion problems.

For typical combined vents, the diversity effect must be introduced into Figure 7 and the equations by multiplying system resistance loss coefficient k by 1.5 (see Table 9 note and Example 5). This multiplier compensates for junction effect and part-load operation.

Manifolds for appliances with barometric draft regulators can be designed without allowing for dilution air through inoperative appliances. In this case, because draft regulators remain closed until regulation is needed, dilution under part load is negligible. In addition, airflow through any inoperative appliance is negligible because the combustion air inlet dampers are closed and the multiple-pass heat exchanger has a high internal flow resistance.

Manifold systems of oil-burning appliances, for example, have a lower flow velocity and, hence, lower losses. As a result, they produce reasonable draft at part load or with only one of several appliances in operation. Therefore, diversity of operation has little effect on chimney design. Some installers set each draft regulator at a slightly different setting to avoid oscillations or hunting possibly caused by burner or flow pulsations.

Calculation of the resistance coefficient of any portion of a manifold begins with the appliance most distant from the vertical portion. All coefficients are then summed from its outlet to the vent terminus. The resistance of a series of tee joints to flow passing horizontally straight through them (not making a turn) is the same as that of an equal length of piping (as if all other appliances were off). This assumption holds whether the manifold is tapered (to accommodate increasing input) or of a constant size large enough for the accumulated input.

Coefficients are assigned only to inlet and exit conditions, to fittings causing turns, and to the piping running from the affected appliance to the chimney exit. Initially, piping shape (round, square, or rectangular) and function (for connectors, vertical piping, or both) are irrelevant.

Some high-pressure, high-velocity packaged boilers require special manifold design to avoid turbulent flow noise. Figure 6 shows a typical configuration for this application. The loss coefficients listed in Table 9 for standard tees and elbows are higher than necessary for long-radius elbows or Y entries. Occasionally, on appliances with high chimneys augmenting boiler outlet pressure, it may appear feasible to reduce the diameter of the vertical portion to below the size recommended by the manufacturer. However, any reduction may cause turbulent noise, even though all normal design parameters have been considered.

Figure 6. Typical Connector Design

With the simplifying assumption that the maximum velocity of the flue gas (which exists in the smaller of the two portions) exists throughout the entire system, the design chart (Figure 7) and Equations (20) and (21) can be used to calculate the size of a vertical portion smaller in area than the manifold or of a chimney connector smaller than the vertical. This assumption leads to a conservative design because true losses in the larger area are lower than assumed. Further, if the size change is small, either as a contraction or expansion, the added loss coefficient for this transition fitting (see Table 9) is compensated for by reduced losses in the enlarged part of the system.

These comments on size changes apply more to individual than to combined systems because it is undesirable to reduce the vertical area of the combined type, and, more frequently, it is desirable to enlarge it. If an existing vertical chimney is slightly undersized for the connected load, the complete chart method must be applied to determine whether a pressure boost is needed, because size is no longer a variable.

Sectional gas appliances with two or more draft hoods do not pose any special problems if all sections fire simultaneously. In this case, the designer can treat them as a single appliance. The appliance installation instructions either specify the size of manifold for interconnecting all draft hoods or require a combined area equal to the sum of all attached draft hood outlet areas. Once the manifold has been designed and constructed, it can be connected to a properly sized chimney connector, vent, or chimney. If the connector and chimney size is computed as less than manifold size (as may be the case with a tall chimney), the operating resistance of the manifold will be lower than the sum of the assigned component coefficients because of reduced velocity.

The general rule for conservative system design, in which manifold, chimney connector, vent, or chimney are different sizes, can be stated as follows: Always assign full resistance coefficient values to all portions carrying combined flow, and determine system capacity from the smallest diameter carrying the combined flow. In addition, horizontal chimney connectors or vent connectors should pitch upward toward the stack at a minimum of 1/4 in. per foot of connector length.

 Input, Diameter, and Temperature Relationships

To obtain a design equation in which all terms are readily defined, measured, or predetermined, the gas velocity and density terms must be eliminated. Using (Equation 6) to replace ρ_m and (Equation 14) to replace V in (Equation 11) gives

(20)

Rearranging to solve for I and including the values of π and 2g gives

(21)

Solving for input using (Equation 21) is a one-step process, given the diameter and configuration of the chimney. More frequently, however, input, available draft, and height are given and the diameter d_i must be found. Because system resistance is a function of the chimney diameter, a trial resistance value must be assumed to calculate a trial diameter. This method allows for a second (and usually accurate) solution for the final required diameter.

 Volumetric Flow in Chimney or System

Volumetric flow Q may be calculated in a chimney system for which (Equation 21) can be solved by solving (Equation 11) for velocity at mean density (or temperature) conditions:

(22)

This equation can be expressed in terms of the same variables as (Equation 21) by using the density value ρ_m of (Equation 6) in (Equation 22) and then substituting (Equation 22) for velocity V. Area is expressed in terms of d_i. Multiplying area and velocity and adjusting for units,

(23)

where Q = volumetric flow rate, cfm. The volumetric flow obtained from (Equation 23) is at mean gas temperature T_m and local barometric pressure B. Equations (21) and (23) do not account for the effects of heat transfer or cooling on flow, draft, or capacity.

Figure 7. Design Chart for Vents, Chimneys, and Ducts(The dashed-line solution of Equations (21) and (23) applies to combustion products and air.)

Figure 8. Gas Vent with Lateral

(Equation 23) is useful in the design of forced-draft and induced-draft systems because draft fans are usually specified in terms of volumetric flow rate at some standard ambient or selected gas temperature. An induced-draft fan is necessary for chimneys that are undersized, that are too low, or that must be operated with draft in the manifold under all conditions.

Design ambient temperature is 60°F (520°R), and all temperatures given are in terms of rise above this ambient. Thus, the 300°F line indicates a 360°F observed vent gas temperature.

Figure 1: Use sea-level theoretical draft. Theoretical draft may be corrected for altitude or reduced air density by multiplying the operating input by the factor in Table 7.

The resistance coefficient k in Figure 7 summarizes the friction loss of the entire chimney system, including piping, fittings, and configuration or interconnection factors.

Figure 2 can be used with Figure 7 to estimate mass and volumetric flow.

For ease of application and consistency with Figure 7, Table 6 lists approximate theoretical draft for typical gas temperature rises above 60°F ambient.

The equations and design chart (Figure 7) are based on the fuel combustion products and temperatures in the chimney system.

When using a temperature multiplier for inclusion of horizontal sections, it must be applied to grids A and C as follows:

The first trial solution for diameter, using Figure 7 or Equations (21) and (23), need not consider the cooling temperature multiplier, even for small sizes. A first approximate size can be used for the temperature multiplier for all subsequent trials because capacity is insensitive to small changes in temperature.

Neither Figure 7 nor the equations contain the same number or order of steps as the derivation; for example, a step disappears when theoretical and available drafts are combined into Δp. Similarly, the examples selected vary in their sequence of solution, depending on which parameters are known and on the need for differing answers, such as diameter for a given input, diameter versus height, or the amount of pressure boost D_b from a forced-draft fan. To compare the use of equations with using Figure 7, the following sample is provided. First, the sample calculation is made using the provided tables and equations with variations in the original order of steps. It illustrates the direct solution for input, velocity, and volume. Secondly a calculation for input is shown that is derived from Figure 7. It differs from the first solution because of the chart arrangement.

Example 1.

Find the input capacity (Btu/h) of a vertical, double-wall type B gas vent, 24 in. in diameter, 100 ft high at sea level. This vent is used with draft hood natural-gas-burning appliances.

Solution:

1. Mass flow from Table 2. M = 1.54 lb/1000 Btu for natural gas, if no other data are given.

2. Temperature from Table 4. Temperature rise = 300°F and T_m = 360 + 460 = 820°R for natural gas.

3. Theoretical draft from Table 6 or (Equation 9). For 100 ft height at 300°F rise, D_t = 0.5 in. of water.

4. Available draft for draft hood appliances: D_a = 0.

5. Flow losses from Table 8. Δp = D_t = 0.5; flow losses for a gravity gas vent equal theoretical draft at mean gas temperature.

6. Resistance coefficients from Table 9. For a vertical vent,

Draft hood	k1 =		1.5
Vent cap	k4 =		1.0
100 ft piping	kL = 0.4(100/24) =		1.67
System total	k =		4.17

7. Solution for input.

Altitude: Sea level, B = 29.92 from Table 7. d_i = 24 in. These values are substituted into (Equation 21) as follows:

8. A solution for velocity requires a prior solution for input to apply to (Equation 14). First, using (Equation 6),

From (Equation 14),

9. Volume flow can now be found because velocity is known. The flow area of 24 in. diameter is 3.14 ft², so

No heat loss correction is needed to find the new flue gas temperature because the size is greater than 20 in., and this vent is vertical with no horizontal connector.

For the same problem, Figure 7 requires a different sequence of solution. The ratio of mass flow to input for a given fuel (with parameter M) is not used directly; the chart requires selecting a CO₂ percentage in the chimney, either from Table 2 or from operating data on the appliances. Then, the temperatures are entered only as rise above ambient. Finally, calculated or estimated resistance coefficients are used to connect the conditions to the vent diameter. At this intersection, an input or flow can be derived. The solution path is as follows:

1. Enter grid A at 5.3% CO₂, and construct line horizontally to left intersecting natural gas.

2. From natural gas intersection, construct vertical line to Δt = 300°F.

3. From 300°F intersection in grid A, go horizontally left to transfer line in grid B.

4. From transfer line go vertically to k = 4.17.

5. From k = 4.17 run horizontally left to Δt = 300°F in grid C.

6. From 300°F go up to Δp = 0.54 in. of water in grid D.

7. From Δp = 0.54 in. of water in grid D, go horizontally right to intersect d_i = 24 in. in grid E.

8. Read capacity or input at 24 in. intersection in grid E as 10.2 × 10⁶ Btu/h between curved lines.

If input is known and diameter must be found, the procedure is the same as with Equation (21). A preliminary k, usually 5.0 for an individual vent or chimney, must be estimated to find a trial diameter. This diameter is used to find a corrected k, and the chart is solved again for diameter.

Figures 8 to 10 show chimney capacity for individually vented appliances computed by the methods presented. These capacity curves may be used to estimate input or diameter for design chart (Figure 7), (Equation 21) or (Equation 23). These capacity curves apply primarily to individually vented appliances with a lateral chimney connector; systems with two or more appliances or additional fittings require a more detailed analysis. Figures 8 to 11 assume the length of the horizontal connector is (1) at least 10 ft and (2) no longer than 50% of the height or 50 ft, whichever is less. For chimney heights of 10 to 20 ft, a fixed 10 ft long connector is assumed. Between 20 and 100 ft, the connector is 50% of the height. If the chimney height exceeds 100 ft, the connector is fixed at 50 ft long.

For a chimney of similar configuration but with a shorter connector, the size indicated in the figures is slightly larger than necessary. In deriving the data for Figures 8 to 11, additional conservative assumptions were used, including the temperature correction C_u for double-wall laterals (see Figure 3) and a constant friction factor (0.4) for all sizes.

The loss coefficient k₄ for a low resistance cap is included in Figures 8, 9, and 10. If no cap is installed, these figures indicate a larger size than needed.

Figure 9. Draft-Regulated Appliance with 0.10 in. of water Available Draft Required

Figure 8 applies to a gas vent with draft hood and a lateral that runs to the vertical section. Maximum static draft is developed at the base of the vertical, but friction reduces the observed value to less than the theoretical draft. Areas of positive pressure may exist at the elbow above the draft hood and at the inlet to the cap. The height of the system is the vertical distance from the draft hood outlet to the vent cap.

Figure 9 applies to a typical boiler system requiring both negative combustion chamber pressure and negative static outlet pressure. The chimney static pressure is below atmospheric pressure, except for the minor outlet reversal caused by cap resistance. Height of this system is the difference in elevation between the point of draft measurement (or control) and the exit. (Chimney draft should not be based on the height above the boiler room floor.)

Figure 10 shows the use of a negative static pressure connector serving a forced-draft boiler. This system minimizes flue gas leakage in the equipment room. The draft is balanced or neutral, which is similar to a gas vent, with zero draft at the appliance outlet and pressure loss Δp equal to theoretical draft.

Figure 10. Forced-Draft Appliance with Neutral (Zero) Draft (Negative Pressure Lateral)

Figure 11 applies to a forced-draft boiler capable of operating against a positive static outlet pressure of up to 0.50 in. of water. The chimney system has no negative pressure, so outlet pressure may be combined with theoretical draft to get minimum chimney size. For chimney heights or system lengths less than 100 ft, the effect of adding 0.50 in. of water positive pressure to theoretical draft causes all curves to fall into a compressed zone. An appliance that can produce 0.50 in. of water positive forced draft (negative draft) is adequate for venting any simple arrangement with up to 100 ft of flow path and no wind back pressure, for which additional forced draft is required.

The following examples illustrate the use of the design chart (Figure 7) and the corresponding equations.

Figure 11. Forced-Draft Appliance with Positive Outlet Pressure (Negative Draft)

Example 3.

Gravity incinerator chimney (see Figure 13).

Located at 8000 ft elevation, the appliance burns 600 lb/h of type 0 waste with 100% excess air at 1400°F outlet temperature. Ambient temperature T_o is 60°F. Outlet pressure is zero at low fire, and +0.10 in. of water at high fire. The chimney will be a prefabricated medium-heat type with a 60 ft connector and a roughness factor of 1.2. The incinerator outlet is 18 in. in diameter, and it normally uses a 20 ft vertical chimney. Find the diameter of the chimney and the connector and the height required to overcome flow and fitting losses.

Solution:

1. Find mass flow from Table 3 as 13.76 lb combustion products per pound of waste, or w = 600(13.76) = 8256 total lb/h.

2. Find mean chimney gas temperature. Based on 60 ft length of 18 in. diameter double-wall chimney, C_u = 0.83 (see Figure 3). Temperature rise Δt_e = 1400 – 60 = 1340°F; thus, Δt_m = 1340(0.83) = 1112°F rise above 60°F ambient. T_m = 1112 + 60 + 460 = 1632°R. Use this temperature in (Equation 6) to find flue gas density at 8000 ft elevation (from Table 7, B = 22.22 in. Hg):

3. Find the required height by finding theoretical draft per foot from (Equation 9) or Figure 5 (Table 6 applies only to sea level).

4. Find allowable pressure loss Δp in the incinerator chimney for a positive-pressure appliance having an outlet pressure of +0.1 in. of water. From Table 8, Δp = D_t + D_a, where D_t = 0.0074H, D_a = 0.10, and Δp = 0.0074H + 0.1 in. of water.

5. Calculate flow velocity at mean temperature from (Equation 13) to balance flow losses against diameter/height combinations:

This velocity exceeds the capability of a gravity chimney of moderate height and may require a draft inducer if an 18 in. chimney must be used. Verify velocity by calculating resistance and flow losses by the following steps.

6. From Table 9, resistance coefficients for fittings are

1 Tee	k3 = 1.25
1 Elbow	k2 = 0.75
Spark screen	k4 = 0.50
Fitting total	kf = 2.50

The piping resistance, adjusted for length, diameter, and a roughness factor of 1.2, must be added to the total fitting resistance. From Figure 6, find the friction factor F at 18 in. diameter and 72 fps as 0.22. Assuming 20 ft of height with a 60 ft lateral, piping friction loss is

and total k = 2.50 + 1.17 = 3.67

Use (Equation 11) to find Δp, to determine whether this chimney height and diameter are suitable.

For these operating conditions, theoretical draft plus available draft yields

Flow losses of 1.02 in. of water exceed the 0.248 in. of water driving force; thus, the selected diameter, height, or both are incorrect, and this chimney will not work. This can also be shown by comparing draft per foot with flow losses per foot for the 18 in. diameter configuration:

Flow losses per foot of 18 in. chimney = 1.02/80 = 0.0128 in. of water

Regardless of how high the chimney is, losses caused by a 72 fps velocity build up faster than draft.

7. A draft inducer could be selected to make up the difference between losses of 1.02 in. of water and the 0.248 in. of water driving force. Operating requirements are

If the inducer selected (see Figure 27C) injects single or multiple air jets into the gas stream, it should be placed only at the chimney top or outlet. This location requires no compensation for additional air introduced by an enlargement downstream from the inducer.

Because 18 in. is too small, assume that a 24 in. diameter may work at a 20 ft height and recalculate with the new diameter.

1. As before, w = 8256 lb/h.

2. At 24 in. diameter, no temperature correction is needed for the 60 ft connector. Thus, T_m = 1400 + 460 = 1860°R (see Table 4), and density is

3. Theoretical draft per foot of chimney height is

4. Velocity is

5. From Figure 6, the friction factor is 0.225, which, when multiplied by a roughness factor of 1.2 for the piping used, becomes 0.27. For the entire system with k_f = 2.50 and 80 ft of piping, find k = 2.5 + 0.27(80/24) = 3.4. From (Equation 11),

D_t + D_a = (20)(0.00786) + 0.1 = 0.257 in. of water, so driving force is less than losses. The small difference indicates that, although 20 ft of height is insufficient, additional height may solve the problem. The added height must make up for 0.085 in. of water additional draft. As a first approximation,

Added height = Additional draft/draft per foot

= 0.085/0.00786 = 10.8 ft

Total height = 20 + 10.8 = 30.8 ft

This is less than the actual height needed because resistance changes have not been included. For an exact solution for height, the driving force can be equated to flow losses as a function of H. Substituting H = +60 for L in (Equation 17) to find k for (Equation 11),

Checking by substitution, total driving force = 0.357 in. of water and total losses, based on a system with 92.66 ft of piping, equal 0.357 in. of water.

The value of H = 32.66 ft is the minimum necessary for proper system operation. Because of the great variation in fuels and firing rate with incinerators, greater height should be used to ensure adequate draft and combustion control. An acceptable height would be from 40 to 50 ft.

Figure 12. Illustration for Example 2

Figure 13. Illustration for Example 3

Figure 14. Illustration for Example 4

Example 4.

Two forced-draft boilers (see Figure 14).

This example shows how multiple-appliance chimneys can be separated into subsystems. Each boiler is rated 100 boiler horsepower on No. 2 oil. The manufacturer states flue gas operation at 13.5% CO₂ and 300°F temperature rise against 0.50 in. positive static pressure at the outlet. The 50 ft high chimney has a 20 ft single-wall manifold and is at sea level. Find the size of connectors, manifold, and vertical.

Solution: First, find the capacity or size of the piping and fittings from boiler A to the tee over boiler B. Then, size the boiler B tee and all subsequent portions to carry the combined flow of A and B. Also, check the subsystem for boiler B; however, because its shorter length compensates for greater fitting resistance, its connector may be the same size as for boiler A.

Find the size for combined flow of A and B either by assuming k = 5.0 or by estimating that the size will be twice that found for boiler A operating by itself. Estimate system resistance for the combined portion by including those fittings in the B connector with those in the combined portion.

Data needed for the solution of (Equation 21) for boiler A for No. 2 oil at 13.5% CO₂ include the following:

For a temperature rise of 300°F and an ambient of 60°F,

From Table 6, theoretical draft = 0.5 in. of water for 100 ft of height; for 50 ft height, D_t = (0.5)50/100 = 0.25 in. of water.

Using D_a = 0.5 in. of water, Δp = 0.25 + 0.5 = 0.75 in. of water.

Assume k = 5.0. Assuming 80% efficiency, input is 41,800 times boiler horsepower (see the section on Conversion Factors):

Substitute in (Equation 21):

Solving, d_i = 10.81 in. as a first approximation. From Table 9, find correct k using next largest diameter, or 12 in.:

Inlet acceleration (direct connection)	k1	=	0.0
90° Elbow	k2	=	0.75
Tee	k3	=	1.25
70 ft piping	kL	=	0.4(70/12) = 2.33
System total			k = 4.33

Note: Assume tee over boiler B has k = 0 in subsystem A.

Corrected temperature rise (see Figure 3 for 20 ft single-wall connector) = 0.75(300) = 225°F, and T_m = 225 + 60 + 460 = 745°R. This corrected temperature rise changes D_t to 0.425 per 100 ft (Table 6 or Equation [9]), or 0.21 for 50 ft. Thus, Δp becomes 0.21 + 0.50 = 0.71 in. of water.

So d_i = 10.35 in., or a 12 in. diameter is adequate.

For size of manifold and vertical, starting with the tee over boiler B, assume 16 in. diameter (see also 11).

System k = 3.9 for the 55 ft of piping from B to outlet:

Inlet acceleration	k1 = 0.0
Two tees (boiler B subsystem)	k3 = 2.5
55 ft piping kL = 0.4(55/	16) = 1.4
System total	k = 3.9

Temperature and Δp will be as corrected (a conservative assumption) in the second step for boiler A. Having assumed a size, find input.

A 16 in. diameter manifold and vertical is more than adequate. Solving for the diameter at the combined input, d_i = 14.2 in.; thus, a 15 or 16 in. chimney must be used.

Note: Regardless of calculations, do not use connectors smaller than the appliance outlet size in any combined system. Applying the temperature correction for a single-wall connector has little effect on the result because positive forced draft is the predominant motive force for this system.

Example 5.

Six gas boilers manifolded at 6000 ft elevation (see Figure 15).

Each boiler is fired at 1.6 × 10⁶ Btu/h, with draft hoods and an 80 ft long manifold connecting into a 400 ft high vertical. Each boiler is controlled individually. Find the size of the constant-diameter manifold, vertical, and connectors with a 2 ft rise. All are double-wall.

Solution: Simultaneous operation determines both the vertical and manifold sizes. Assume the same appliance operating conditions as in Example 1: CO₂ = 5.3%, natural gas, flue gas temperature rise = 300°F. Initially assume k = 5.0 is multiplied by 1.5 for combined vent (see note at bottom of Table 9); thus, design k = 7.5. For a gas vent at 400 ft height, Δp = D_t = H × D_t/m = 4 × 0.5 = 2.0 in. of water at rise 300°F in Table 6; D_t = 0.5 in. of water/ft. At 6000 ft elevation, operating input must be multiplied by an altitude correction (Table 7) of 1.25. Total design input is 1.6 × 10⁶(6)(1.25) = 12 × 10⁶ Btu/h. From Table 2, M = 1.54 lb/Btu at operating conditions. T_m = 300 + 60 + 460 = 820°R, and B = 29.92 in. Hg because the 1.25 input multiplier corrects back to sea level. From (Equation 21),

Because the diameter is greater than 20 in., no temperature correction is needed.

Recompute k for 400 + 80 = 480 ft of 22 in. diameter (Table 9):

Draft hood inlet acceleration	k1 = 1.5
Two tees (connector and base of chimney)	2k3 = 2.5
Low-resistance top	k4 = 0.5
480 ft piping	kL = 0.4(480/22) = 8.7
System total	k = 13.2

Combined gas vent design k = multiple vent factor 1.5 × 13.2 = 19.8. Substitute again in (Equation 21):

Substitute in (Equation 21) to obtain the third trial:

The third trial is less than the second and again shows the manifold and vertical chimney diameter to be between 26 and 28 in.

For connector size, see the National Fuel Gas Code (NFPA Standard 54/AGA Standard Z223.1) for double-wall connectors of combined vents. The height limit of the table is 100 ft; do not extrapolate and read the capacity of 18 in. connector as 1,740,000 Btu/h at 2 ft rise. Use 18 in. connector or draft hood outlet, whichever is larger. No altitude correction is needed for connector size; the draft hood outlet size considers this effect.

Note: (Equation 21) can also be solved at local elevation for exact operating conditions. At 6000 ft, the local barometric pressure is 23.98 in. Hg ( Table 7), and assumed theoretical draft must be corrected in proportion to the reduction in pressure:

D_t = 2.0(23.98/29.92) = 1.60 in. of water. Operating input of 6(1.6 × 10⁶) = 9.6 × 10⁶ Btu/h is used to find d_i, again taking final k = 17.1:

This example illustrates the equivalence of the chart method of solution with solution by Equation (21). (Equation 21) gives the correct solution using either method 1, with only the fuel input corrected back to sea-level condition, or method 2, correcting Δp for local barometric pressure and using operating input at altitude. Method 1, correcting input only, is the only choice with Figure 1 because the design chart cannot correct to local barometric pressure.

Figure 15. Illustration for Example 5

Example 6.

Pressure boost for undersized chimney (not illustrated).

A natural gas boiler at sea level (no draft hood) is connected to an existing 12 in. diameter chimney. Input is 4 × 10⁶ Btu/h with flue gas at 10% CO₂ and 300°F temperature rise above ambient. System resistance loss coefficient k = 5.0 with 20 ft vertical chimney. The appliance operates with neutral outlet static draft, so, D_a = 0 in. of water.

a. How much draft boost is needed at operating temperature?

b. What fan rating is required at 60°F ambient temperature?

c. Where in the system should the fan be located?

Solution: Combustion data: from Figure 2, 10% CO₂ at 300°F temperature rise indicates 0.31 cfm per 1000 Btu/h.

Total flow rate Q = (0.31/1000)(4 × 10⁶) = 1240 cfm at chimney gas temperature. Then, (Equation 23) can be solved for the only unknown, Δp:

Δp = 0.50 in. of water needed at 300°F temperature rise. For 20 ft of height at 300°F temperature rise (Table 6 or Equation [9]),

a. Pressure boost D_b supplied by the fan must equal Δp minus theoretical draft (Table 8) when available draft is zero. D_b = Δp – D_t = 0.5 – 0.10 = 0.40 in. of water at operating temperature.

b. Draft fans are usually rated for standard ambient (60°F) air. Pressure is inversely proportional to absolute flue gas temperature. Thus, for ambient air,

This pressure is needed to produce 0.40 in. of water at operating temperature. In specifying power ratings for draft fan motors, a safe policy is to select one that operates at the required flow rate at ambient temperature and pressure (see Example 7).

c. A fan can be located anywhere from boiler outlet to chimney outlet. Regardless of location, the amount of boost is the same; however, chimney pressure relative to atmosphere will change. At boiler outlet, the fan pressurizes the entire connector and chimney. Thus, the system should be gastight to avoid leaks. At the chimney outlet, the system is below atmospheric pressure; any leaks flow into the system and seldom cause problems. With an ordinary sheet metal connector attached to a tight vertical chimney, the fan may be placed close to the vertical chimney inlet. Thus, the connector leaks safely inward, while the vertical chimney is under pressure.

Example 7.

Draft inducer selection (see Figure 16).

A third gas boiler must be added to a two-boiler system at sea level with an 18 in. diameter, 15 ft horizontal, and 75 ft of total height of connector and chimney system. Outlet conditions for natural gas draft hood appliances are 5.3% CO₂ at 300°F temperature rise. Boilers are controlled individually, each with 1.6 × 10⁶ Btu/h, for 4.8 × 10⁶ Btu/h total input. The system is currently undersized for gravity full-load operation. Find capacity, pressure, size, and power rating of a draft inducer fan installed at the outlet.

Solution: Using (Equation 23) requires evaluating two operating conditions: (1) full input at 300°F rise and (2) no input with ambient air. Because the boilers are controlled individually, the system may operate at nearly ambient temperature (100°F flue gas temperature rise or less) when only one boiler operates at part load. Use the system resistance k for boiler 3 as the system value for simultaneous operation. It needs no compensating increased draft, as with gravity multiple venting, because a fan induces flow at all flue gas temperatures. From Table 9, the resistance summation is

Inlet acceleration (draft hoods)	k1 = 1.50
Tee above boiler	k3 = 1.25
Tee at base of vertical	k3 = 1.25
90 ft of 18 in. diameter pipe	kL = 0.4(90/18) = 2.00
System total	k = 6.00

At full load, T_m = 300 + 60 + 460 = 820°R; B = 29.92 in. Hg. Flow rate Q must be found for operating conditions of 1.54 lb per 1000 Btu (Table 2) at density ρ_m and full input (4.8 × 10⁶)/60 Btu/min.

From (Equation 6),

Volumetric flow rate is

For 300°F flue gas temperature rise, D_t = 0.5(75/100) = 0.375 in. of water theoretical draft in the system (Table 6). Solving (Equation 23) for Δp,

Thus, a fan is needed because Δp exceeds D_t. Required static pressure boost (Table 8) is

Fans are rated for ambient or standard air (60 to 70°F) conditions. Flue gas pressure is directly proportional to density or inversely proportional to absolute temperature. Moving 2551 cfm of flue gas at 300°F temperature rise against 0.127 in. of water pressure requires the static pressure boost with standard air to be

Thus, a fan that delivers 2551 cfm of flue gas at 0.20 in. of water at 60°F is required. Figure 17 shows the operating curves of a typical fan that meets this requirement. The exact volume flow rate and pressure developed against a system k of 6.0 can be found for this fan by plotting airflow rate versus Δp from (Equation 23) on the fan curve. The solution, at point C, occurs at 1950 cfm, where both the system Δp and fan static pressure equal 0.46 in. of water.

Although some fan manufacturers’ ratings are given at standard air conditions, the motors selected will be overloaded at temperatures below 300°F air temperature rise. Figure 17 shows that power required for two conditions with ambient air is as follows:

1. 1950 cfm at 0.46 in. of water static pressure requires 0.51 hp

2. 2650 cfm at 0.25 in. of water static pressure requires 0.50 hp

Thus, the minimum size motor will be 0.5 hp and run at 1590 rpm.

Manufacturers’ literature must be analyzed carefully to discover whether the sizing and selection method is consistent with appliance and chimney operating conditions. Final selection requires both a thorough analysis of fan and system interrelationships and consultation with the fan manufacturer to verify the fan and motor capacity and power ratings.

Figure 16. Illustration for Example 7

Figure 17. Typical Fan Operating Data and System Curves

In much of the United States, gas-burning appliances requiring venting of combustion products are installed and vented in accordance with the National Fuel Gas Code (NFPA Standard 54/AGA Z223.1), and in Canada with the Natural Gas and Propane Installation Code (CAN/CSA Standard B149.1). This standard includes capacity data and definitions for commonly used gas vent systems. In addition, these codes group vented gas-fired appliances by draft and flue gas conditions as follows:

Category I appliances are further divided into two broad types: natural-draft appliances rely solely on flue gas buoyancy to draw combustion air through the appliance/vent system and are equipped with draft hood (integral or external to the appliance itself) used to limit draft at the outlet of the appliance. Fan-assisted appliances use a blower to move combustion air through the appliance, but rely on flue gas buoyancy alone to move flue products through the vent system (vent pressure at the outlet of the appliance is zero or negative). From a vent design standpoint, the difference between these two types of Category I appliances is important because the vent systems used with natural-draft Category I appliances must handle not only the products of combustion, but also dilution air drawn into the system through the draft hood. Dilution air increases the total gas volume that must be handled by the vent system, and therefore reduces the capacity of a given size vent system. However, a given vent system handling less dilution air tends to be more prone to forming flue gas condensate, particularly under transient conditions, which can cause corrosion or drainage problems when present in excessive amounts.

Both the National Fuel Gas Code (ANSI/NFPA Standard 54/ANSI/AGA Z223.1) and the Natural Gas and Propane Installation Code (CAN/CSA Standard B149.1) include tables and other information that can be used to design vent systems for one or more Category I appliances. These tables include maximum capacities, representing the largest appliance input rating that could safely be vented using a given vent diameter, height, lateral, and material. For reasons discussed previously, these capacities are larger for fan-assisted appliances than they are for natural draft appliances. At the same time, these tables specify a minimum capacity for a given size chimney used to vent fan assisted appliances to minimize condensate formation. Minimizing condensate formation, as well as the time required to “prime” the chimney (establish draft when the chimney is initially cold) are the reasons for limitations placed by the above codes on the use of single wall vent connectors, and masonry chimneys.

Category I guidelines in these codes were developed using the VENT-II computer program (GTI 2009) to perform transient analysis with the appliance(s) cycling. Gas-fired central furnace cycle times of 3.87 min on, 13.3 min off, 17.17 min total, were determined based on a design outdoor ambient of 42°F with an oversize factor of 1.7. One criterion used in developing these guidelines is that the vent connector dry before the end of the appliance on cycle, whereas the vertical portion of the vent is required to dry out before the end of the total cycle.

Methods for determining the category of a given gas appliance vary. ANSI Standards Z21.13, Z21.47, and Z21.10 include category determination tests for boilers, furnaces, and water heaters and include a requirement that the resulting category be marked on the appliance’s data plate. Appliances not certified to one of these standards may not be categorized at all or may be categorized in a way inconsistent with ANSI test procedures. In such cases, caution should be exercised in attempting to use the venting guidelines mentioned in this section.

No universal venting guidelines exist for Category II, III, or IV appliances. In many cases, guidelines for such an appliance are defined by the appliance manufacturer in terms of an acceptable range of vent length, diameter, material, and terminal design. In such cases, the appliance manufacturer may also specify the means of condensate disposal and vent terminal location. Criteria used to develop these guidelines primarily include maintaining draft (or pressure) at the appliance outlet within acceptable limits, avoiding corrosion damage caused by condensate, and withstanding wind loads. ANSI Standards Z21.13, Z21.47, and Z21.10 include performance tests to establish these guidelines and require the manufacturer to publish them in the installation manual. In some cases the appliance manufacturer may specify an acceptable range of draft (or pressure) at the appliance outlet, rather than defining the vent system geometry.

Oil-fired appliances requiring venting of combustion products must be installed and vented in accordance with ANSI/NFPA Standard 31. The standard offers recommendations for metal relining of masonry chimneys.

Recommendations for minimum chimney areas for oil-fired natural-draft appliances are offered in Tables 3 and 4 in the HYDI (1989).

Implementation of the U.S. National Appliance Energy Conservation Act (NAECA) of 1987 brought attention to heating appliance efficiency and the effect of NAECA on existing chimney systems. Oil-fired appliances maintained a steady growth in efficiency since the advent of the retention-head oil burner and its broad application in both new appliances and the replacement of older burners in existing appliances. Higher appliance efficiencies brought about lower flue gas temperatures. Reduced firing rates became more common as heating appliances were more closely matched to the building heating load. Burner excess air levels also dropped, which resulted in lower mass flows through the chimney and additional reductions in the flue gas temperature. However, the improvements in overall appliance efficiency have not been accompanied by upgrades in existing chimney systems, and upgraded systems are not commonly applied in new construction. An upgrade in a vent system probably involves application of corrosion-resistant materials and/or the reduction in heat loss from the vent system to maintain draft and reduce condensation on interior surfaces of the vent system.

This section is condensed from an ASHRAE Journal article; for more details, please see Stone (1969).

Fireplaces with natural-draft chimneys follow the same gravity fluid flow law as gas vents and thermal flow ventilation systems. All thermal or buoyant energy is converted into flow, and no draft exists over the fire or at the fireplace inlet. Formulas have been developed to study a wide range of fireplace applications, but the material in this section covers general cases only.

Mass flow of hot flue gases through a vertical pipe is a function of rate of heat release and the chimney area, height, and system pressure loss coefficient k. The flow induced in a vertical pipe has a limiting value. A fireplace may be considered as a gravity duct inlet fitting with a characteristic entrance-loss coefficient and an internal heat source. A fireplace functions properly (does not smoke) when adequate intake or face velocity across those critical portions of the frontal opening nullifies external drafts and internal convection effects.

In a fireplace-chimney system, the equations assume that all potential buoyant energy is converted into flow as controlled by various losses. This system is analogous to a gas venting system with a draft hood, thereby allowing use of similar concepts as a starting point for size or capacity analysis. The amount of available draft ahead of the fireplace opening is insignificant and need not be considered. Because chimney efficiency, by one definition, equals available draft divided by theoretical draft, the numerical efficiency value approaches zero. Thus, the flow conversion basis is preferable for design over the efficiency approach.

System parameters for preventing flue gas spillage from a draft hood or similar collection fitting, can be computed with considerable certainty when heat input is constant or cannot exceed a predictable limiting value. Fireplaces, however, can be fired at extremes ranging from smoldering embers to flash fires of newspapers or dry kindling. Normal opening width and length allow greater access of combustion air to fires than a typical chimney can carry away; thus, combustion overloading occasionally leads to some smoking. At low rates of combustion, airflow velocities into the fireplace face are less than the velocity of natural convection currents induced at side walls by heat stored in the brick, allowing wisps of smoke to stray away from the combustion products’ main flow path. Smoking tendencies are compounded at low firing rates by indoor/outdoor pressure differentials caused by winds, thermal forces, or fans, because the accompanying thermal force (buoyancy) of low combustion products temperature is insufficient to overcome strong wind or fan effects.

In the following analysis, note that fireplaces are primarily air-collecting hoods, diluting a small amount of combustion products with large amounts of air. Maximum mass flow of air into any given fireplace chimney is limited, and diminishes past a certain maximum. Thus, as combustion rate increases, combustion product temperatures rise to the point where masonry cracks; metals overexpand, warp, and oxidize; and steady smoking can occur because heated flue gases evolve beyond the limited capacity of the chimney.

An inoperative fireplace is completely at the mercy of indoor/outdoor pressure differences caused by winds, building stack effects, and operation of forced-air heating systems or mechanical ventilation. Thus, smoking during start-up can have complex causes seldom related to the chimney. Increasingly in new homes and especially in high-rise multiple family construction, fireplaces of normal design cannot cope with mechanically induced reverse flow or shortages of combustion air. It is mandatory in these circumstances to treat and design a fireplace as a constantly operating mechanical exhaust system, with induced-draft blowers (mechanical-draft systems) that can overpower other mechanized air-consuming systems, and can develop sufficient flow to avoid smoking and excessive flue temperatures.

The gravity-flow capacity equation (Kinkead 1962) of a fireplace-chimney system equates mass flow with the resultant system driving forces and losses.

(24)

where

From (Equation 24), it is possible to develop a relationship for average frontal velocity, maximum chimney capacity, and variation of gas temperature with changes in fuel input rate.

Using resistance coefficients in these compact systems is preferable to the usual method of equivalent lengths. The summation term k in a vertical chimney is the total of four individual resistance factors:

For a 12 ft high open top chimney, 12 in. in diameter on a typical fireplace, the system resistance is

Note that in a short chimney, the wall friction coefficient k_c is only 0.4 and has a minor effect on system flow. Greater or lesser chimney roughness, or a change from low to high heat loss materials will have little bearing on fireplace effectiveness in short chimneys. In some situations, it may be necessary, for completeness, to include a k term for air supply resistance.

To determine the frontal velocity V_F of ambient air, the term w is replaced using the substitution

(25)

where

Accordingly, mean frontal velocity V_F becomes

(26)

For present purposes, the molecular weight or specific gravity of dilute combustion products is practically the same as that of air, and both can be expressed with adequate accuracy in terms of absolute gas temperature by the same relationship:

(27)

(28)

where

Substitution of the density/temperature relationships (Equations [27] and [28]) into (Equation 26) allows further simplification, leading to the general frontal velocity expression:

(29)

Here, frontal velocity is a function of the product of three terms:

For the 12 ft high example chimney, assume ambient temperature (for calculation purposes) is 70°F, or T_o = 530°R indoors and outdoors, with no air supply resistance. (Equation 29) expresses variation in V_F with gas temperature as shown in Figure 19. Assume also A_c/A_F = 0.10, so that frontal opening is ten times chimney area.

The mean airflow velocity into a fireplace frontal opening is nearly constant from 300°F flue gas temperature rise up to any higher temperature. Local velocities vary within the opening. Depending on design, air typically enters horizontally along the hearth and then flows into the fire and upward, clinging to the back wall (see Figure 18). A recirculating eddy forms just inside the upper front half of the opening, induced by the high velocity of flow along the back. Restrictions or poor construction in the throat area between the lintel and damper also increase the eddy. Because the eddy moves smoke out of the zone of maximum velocity, the tendency of this smoke to escape must be counteracted by some minimum inward air movement over the entire front of the fireplace, particularly under the lintel.

Figure 18. Eddy Formation

Construction of a fireplace, its internal configuration, damper location, height and location of lintel, slope of back and sides, and so on all affect minimum frontal velocity to prevent smoking with ordinary fires. It seems desirable to maintain a smooth, gradual tapering transition between the hearth or flame region up into the damper location. A sudden transition, unless it is well above the lintel, induces velocity components that tend to increase eddying. With a shallow chamber between lintel and damper zone, there is insufficient volume for convection currents, and some flue gases may be diverted horizontally before being captured by the main flow. Masons following the dimensional parameters recommended by Ramsey and Sleeper (1956) or in damper literature can avoid these design flaws. With prefabricated fireplace systems, which often tend to be unconventional, careful testing is essential to ensure safe, smoke-free performance at minimum chimney height.

Figure 19. Effect of Chimney Gas (Combustion Products) Temperature on Fireplace Frontal Opening Velocity

A minimum mean frontal inlet velocity of 0.8 fps, in conjunction with a chimney flue gas temperature of at least 300 to 500°F above ambient, should control smoking in a well-constructed conventional masonry fireplace. As noted in Figure 19, this velocity can be achieved even in low chimneys by system resistance of 2.9 or less, in conjunction with rates of combustion producing flue gas temperatures above a certain minimum level.

Figure 19 also illustrates why increases in flue gas temperature rise greater than 300°F have no perceptible effect on fireplace smoking, because combustion air mass flow and face velocity actually decrease at flue gas temperature rises higher than 500°F. For practical purposes, chimney flue gas temperature has little influence on fireplace performance, if flue gas temperature is at least 300°F above ambient. Fireplace performance analysis thus can be continued assuming a constant flue gas temperature of 500°F rise above ambient:

where T_m = 1030°R and T_o = 530°R.

Assuming 70°F ambient (T_o = 530°R), and factoring out 2g, (Equation 29) becomes roughly

(30)

This expression states that relative fireplace performance is purely a matter of geometry. It allows evaluation of the opening size to maintain minimum frontal velocity as a function of height, as well as quick analysis of effect of frontal area on velocity.

Figure 20 shows that 0.8 fps face velocity requires a relatively small A_F/A_c area ratio at 8 ft chimney height, whereas the fireplace frontal opening area can be nearly twice as large with a 40 ft high chimney. To compute this curve, system resistance is assumed as

Figure 20. Permissible Fireplace Frontal Opening Area for Design Conditions (0.8 fps mean frontal velocity with 12 in. inside diameter round flue)

This expression of resistance assumes the fully open free area of the damper throat to be twice chimney flue area.

A corollary application of (Equation 30) assumes fixed chimney size and height, and explores variation in frontal velocity with changes in area ratio. Figure 21 shows that a 12 in. diameter, 15 ft high chimney cannot produce adequate velocities for frontal area ratios greater than 11. These curves point out the possibility of further simplification to yield a fireplace-chimney design equation for a constant face velocity of 0.8 fps.

Figure 21. Effect of Area Ratio on Frontal Velocity (for chimney height of 15 ft with 12 in. inside diameter round flue)

(31)

This equation allows permissible frontal area A_F to be determined as a function of chimney area, height, and system resistance. It can also be used to determine A_c, if A_F is known. However, in this latter case, Equations (32) and (33) are preferable, because chimney size (D_c, A_c, and R_h) appears twice in the right-hand side of each equation.

For circular flues,

(32)

For other flue shapes,

(33)

where D_c is chimney inside diameter, ft, R_h is inside hydraulic radius, ft, frontal velocity V_F is 0.8 fps at maximum combustion air mass flow rate, and damper opening free area is twice chimney flue area combustion air A_F.

These relationships clearly reveal the origin of rules of thumb specifying opening area as 8, 10, or 12 times chimney area. Some design guides go beyond and classify chimneys by height groups so that short ones serve smaller fireplace openings. Using a mean face velocity of 0.8 fps yields ratios and heights that fall well within the limits of such rules, as well as providing a unifying concept for computing design charts.

The previous relationships primarily apply to masonry single-face fireplaces of conventional construction, but are applicable to other types with considerable validity if face or opening area is properly treated. Corner or double-face designs and many free-standing types embody conventional smokeshelf construction with similar resistance coefficients.

The preceding equations can be applied to illustrate the effect of excessive firing rates on chimney flue gas temperature. Masonry fireplaces are highly inefficient as heating devices, and tests show that, over a wide range of controlled fuel inputs using a drilled-port nonaerated gas burner, 75 to 80% of the gross heating value goes up the chimney. With constant flue heat loss of 80%, 70% of heat loss is sensible heat, which produces the rise in flue gas temperature. This heat input/temperature relationship may be developed by expressing the system flow relationship in terms of heat input:

(34)

where

Equating Equations (34) and (24), and eliminating w,

(35)

Substituting Equations (27) and (28) for density and solving for q,

(36)

For any given system, all terms except (T_m – T_o)^3/2/T_m may be considered fixed; therefore, gas temperature rise for a fireplace can be obtained as a concise function of heat input:

(37)

where is a constant.

The flow-temperature function of a carefully controlled experimental fireplace (Equation [36] or [37]) in which as fixed is plotted in Figure 22 as chimney flue gas temperature versus heat input rate. Data taken on a fireplace-chimney system in this way may readily be evaluated to determine system resistance coefficient k.

Figure 22. Variation of Chimney Flue Gas Temperature with Heat Input Rate of Combustion Products

Figure 23 shows fireplace and chimney dimensions for the specific conditions of circular flues at 0.8 fps frontal velocity. This chart solves readily for maximum frontal opening for a given chimney, as well as for chimney size and height with a predetermined opening. For example, a 30 in. high by 42 in. wide opening for a 12 ft high chimney (measured from the highest point of front opening) requires a 12.5 in. flue. Figure 23 assumes no wind or air supply difficulties. For other face velocities, A_F is found by multiplying frontal area (center scale) by velocity ratio 0.8/V_F. To confirm the example results, calculate from (Equation 32) as follows:

Figure 23. Chimney Sizing Chart for Fireplaces Mean face velocity = 0.8 fps (Stone 2005)

where H is 12 ft and D_c is 14 in.

Although derived specifically for circular flues, A_F applies with negligible sacrifice in performance to chimney flue cross sections such as squared or rounded ovals, because flue area is a much more important factor than friction caused by changes in hydraulic radius. For example, in a 20 ft high chimney, assuming a square flue section equal in area to an 8 in. circle, frontal area is reduced from 4.16 ft² with the round, to 4.09 ft²with the square, or about 2%, a difference that is hardly observable. For some typical constructions, Figure 24 suggests methods of estimating frontal area.

Figure 24. Estimation of Fireplace Frontal Opening Area

These relationships apply to steady-state conditions, which are obtained only after warm-up. Igniting a rapidly flammable charge in a cold system creates pulses of expanding hot combustion products, which frequently escape from the fireplace. In a typical chimney, priming time (time to accelerate from no flow to full upward velocity) is around 5 to 10 s. Thus, initial or intermittent smoking caused by momentarily excessive combustion rates occurs because of system inability to increase flue gas velocity in pace with combustion surges.

Flue or chimney material is of little relevance to fireplace-chimney operation. Materials to which these equations and charts apply include the very hazardous uninsulated single-wall metal, through conventional masonry, as well as the various constructions of lightweight, insulated, factory-built low-heat-space-appliance chimneys. (Safety standards classify fireplaces as low-heat appliances.) Because most fireplace chimneys are short and vertical, neither heat loss nor wall roughness has any important effect on flow. The governing factors in chimney selection for fireplaces are mainly safety, installation, convenience, and esthetics.

Indoor/outdoor pressure differences caused by winds, kitchen or bath exhaust fans, building stack effects, and operation of forced-air heating systems or mechanical ventilation affect the operation of a fireplace. Thus, smoking during start-up can be caused by factors unrelated to the chimney. Frequently, in new homes (especially in high-rise multiple-family construction), fireplaces of normal design cannot cope with mechanically induced reverse flow or shortages of combustion air. In these circumstances, a fireplace should include an induced-draft blower able to overpower other mechanized air-consuming systems. An inducer for this purpose is best located at the chimney outlet and should produce 0.8 to 1.0 fps fireplace face velocity of ambient air in an individual flue or 10 to 12 fps chimney velocity.

In conventional fireplaces, the greater the frontal velocity, the more freedom from smoking. The damper free area, together with its resultant resistance coefficient, are thus major factors in obtaining good masonry fireplace performance, especially with short chimneys. Tests show that damper free area need not exceed twice the required flue area, because little further resistance reduction occurs past this limit.

If the damper selected has a free area equal to or less than the required area, it will be definitely restrictive, despite complete adequacy of other factors. Manufacturers’ literature seldom includes damper free area or opening dimensions, and the dimensions may vary further after installation because of interferences with lintels and other parts. It is expedient to select dampers of adequate free area for best results.

Partially closing a damper during a vigorous fire illustrates this point; what is not so obvious is that greater damper openings may be needed in some cases to control smoke by achieving adequate frontal velocities.

Many free-standing fireplaces are built without the usual smokeshelf/throat damper configuration. The same parameters and relationships apply to free-standing fireplaces as to masonry fireplaces; however, in many designs it is difficult to assign a true frontal area for velocity analysis. Where freestanding fireplaces include back outlets, or require a horizontal run to reach the chimney, compensation is necessary for the flow resistance or pressure losses caused by turns. As a rough approximation, increase the system resistance coefficient k about 1.5 for two 90° elbows, or a back outlet with lateral run into a tee.

Losses caused by lateral connectors and tees generally mean a one-size increase (e.g., moving from 7 to 8) is sufficient. Collar sizes generally determine correct chimney size, and instructions are furnished to cover special situations. More sophisticated prefabricated fireplace designs are provided with matching correctly sized chimneys, and are intended for installation in conventional wood-frame residences.

The equations and design charts presented here assume no wind or air supply difficulties. Lack of replacement air, competing ventilation exhaust fans, and negative interior pressures caused by winds are all obvious causes of smoking or poor fireplace priming. Even when fully primed and hot, thermal forces in a fireplace chimney can be overpowered by a combination of adverse influences. In modern high-rise residential apartments, where an effort has been made to provide all amenities, fireplaces may have to cope simultaneously with all these troubles. Continuous induced draft for the chimney alleviates most of these problems by maintaining a chimney prime at all times. An inducer for this purpose should be able to produce 0.8 to 1.0 fps fireplace frontal velocity of ambient air in any individual flue. Where multiple flues are installed in a chase, a single, larger inducer serving the chase can be sized for the combined fireplace frontal opening area. Even where flues are of different heights and sizes, a draft inducer selection, assuming all flues to be some compromise median height and size, produces far greater user satisfaction than reliance on gravity alone.

In single-family dwellings, fireplace problems are more often caused by reduced interior pressures from wind effects than by poor chimney terminal location or characteristics. Efforts to cure smoking, slow priming, or blowback of ashes usually involve one of the myriad forms of stationary or rotating caps, cowls, or chimney pots. This questionable expedient has contradictory effects. Usually, the added still-air resistance of a cap reduces fireplace frontal velocity, which limits combustion airflow rate, and thus may tend to increase smoking. On the other hand, a cap reduces air dilution of smaller fires, raises chimney temperature, and improves stability of flame, thus tending to mitigate wind impulses that cause momentary flow reduction. The usual fireplace damper can also be used to restrict flow, and thus raise temperature.

Remedies for fireplace malfunctions may be analyzed using (Equation 30). For example, it is apparent that any change of parameters on the right-hand side that might decrease V_F can increase smoking tendencies. If frontal area A_F or k increases, there will be a corresponding decrease in V_F. Similarly, if chimney area A_c or chimney height H are reduced, then V_F decreases. Further, because frontal velocity varies as the square root of the term H/k, it is more effective to reduce frontal area, thereby increasing A_c/A_F, than to increase H or reduce k.

Logical expedients for increasing V_F frontal velocity, and thus improving performance of fireplaces and chimneys, include the following:

CSA Standard P.4.1-02 can be used for measuring annual efficiency in Canada.

Failure to supply outdoor air for combustion may result in erratic or even dangerous operating conditions. A correctly designed gas appliance with a draft hood can function with short vents (5 ft high) using an outdoor air supply opening as small in area as the vent outlet collar. Such an orifice, when equal to vent area, has a resistance coefficient in the range of 2 to 3. If the air supply opening is as much as twice the vent area, however, the coefficient drops to 0.5 or less.

The following rules may be used as a guide:

Factors to be considered when selecting chimney materials include (1) the temperature of flue gases; (2) their composition and propensity for condensation of water vapor from combustion products (dew point); (3) presence of sulfur, halogens, and other fuel and air contaminants that lead to corrosion of the chimney vent system; and (4) the appliance’s operating cycle (condensate dwell time).

Figure 7 covers materials for vents and chimneys in the 4 to 48 in. size range; these include single-wall metal, various multiwall air- and mass-insulated types, and precast and site-constructed masonry. Each has different characteristics (e.g., frequency of joints, roughness, heat loss), but the type of materials used for systems 14 in. and larger is relatively unimportant in determining draft or capacity. This does not preclude selecting a safe product or method of construction that minimizes heat loss and fire hazard in the building.

National codes and standards classify heat-producing appliances as low, medium, and high heat, with appropriate reference to chimney and vent constructions permitted with each. These classifications are mainly based on size, process use, or combustion temperature. Often, the appliance classification gives little information about outlet gas temperature or venting needed. The designer should, wherever possible, obtain gas outlet temperature conditions and properties that apply to the specific appliance, rather than going by code classification only.

Where building codes allow engineered chimney systems, chimney material selection based on gas outlet temperature can save space as well as reduce structural and material costs. For example, in some jurisdictions, approved gas-burning appliances with draft hoods operating at inputs over 400,000 Btu/h may be placed in a heat-producing classification that prohibits use of type B gas vents. An increase in input may not cause an increase in outlet temperature or in venting hazards, and most building codes recommend correct matching appliance and vent.

Single-wall uninsulated steel stacks can be protected from condensation and corrosion internally with refractory firebrick liners or by spraying calcium aluminate cement over a suitable interior expanded metal mesh or other reinforcement. Another form of protection applies proprietary silica or other prepared refractory coatings to pins or a support mesh on the steel. The material must then be suitably cured for moisture and heat resistance.

Moisture condensation on interior surfaces of connectors, vents, stacks, and chimneys is a more serious cause of deterioration than heat. Chimney wall temperature and flue gas velocity, temperature, and dew point affect condensation. Contaminants such as sulfur, chlorides, and fluorides in the fuel and combustion air raise the flue gas dew point. Studies by Beaumont et al. (1970), Mueller (1968), Pray et al. (1942-53), and Yeaw and Schnidman (1943) indicate the variety of analytical methods as well as difficulties in predicting the causes and probability of actual condensation.

Combustion products from any fuel containing hydrogen condense onto cold surfaces or condense in bulk if the main flow of flue gas is cooled sufficiently. Because flue gas loses heat through walls, condensation occurs first on interior wall surfaces cooled to the flue gas dew point, forming a dew and then a liquid film and, with further cooling, flowing down into zones where condensation would not normally occur.

Start-up of cold interior chimney surfaces is accompanied by transient dew formation, which evaporates on heating above the dew point. This phenomenon causes little corrosion when very low sulfur fuels are used. Proper selection of chimney dimensions and materials minimizes condensation and thus corrosion.

Experience shows a correlation between the sulfur content of the fuel and the deterioration of interior chimney surfaces. Figure 5 in Chapter 28 of the 2017 ASHRAE Handbook—Fundamentals illustrates one case, which applies to any fuel gas. The figure shows that the flue gas dew point increases at 40% excess air from 127°F with zero sulfur to 160°F with 15 grains of sulfur per 100 ft³ of fuel having a heat value of 550 Btu/ft³.

The figure can also be used to approximate the effect of fuel oil sulfur content on flue gas dew point. For example, fuel oil with a sulfur content of 0.5% (by mass) contains about 252 grains of sulfur per gallon or 25,200 grains of sulfur per 100 gallons. If the fuel heat value is 140,000 Btu/gal, the ratio that defines the curves in the figure gives a curve value of 18. Estimates for lower percentages of sulfur (0.25, 0.05) can be formed as factors of the value 18.

Because the corrosion mechanism is not completely understood, judicious use of resistant materials, suitably insulated or jacketed to reduce heat loss, is preferable to low-cost single-wall construction. Refractory materials and mortars should be acid-resistant, and steels should be resistant to sulfuric, hydrochloric, and hydrofluoric acids; pitting; and oxidation. Where low flue gas temperatures are expected together with low ambients, an air space jacket or mineral fiber lagging, suitably protected against water entry, helps maintain surface and flue gas temperatures above the dew point. Using low-sulfur fuel, which is required in many localities, reduces both corrosion and air pollution.

Type 1100 aluminum alloy or any other non-copper-bearing aluminum alloy of 99% purity or better provides satisfactory performance in prefabricated metal gas vent products. For chimney service, flue gas temperatures from appliances burning oil or solid fuels may exceed the melting point of aluminum; therefore, steel is required. Stainless steels such as type 430 or 304 give good service in residential construction and are referenced in UL-listed prefabricated chimneys. Where more corrosive substances (e.g., high-sulfur fuel or chlorides from solid fuel, contaminated air, refuse) are anticipated, type AL 29-4CR^® or equivalent stainless steel offers a good match of corrosion resistance and mechanical properties.

As an alternative to stainless steel, porcelain enamel offers good resistance to corrosion if two coats of acid-resistant enamel are used on all surfaces. A single coat, which always has imperfections, allows base metal corrosion, spalling, and early failure.

Prefabricated chimneys and venting products are available that use light corrosion-resistant materials, both in metal and masonry. The standardized, prefabricated, double-wall metal type B gas vent has an aluminum inner pipe and a coated steel outer casing, either galvanized or aluminized. Standard air space from 1/4 to 1/2 in. is adequate for applicable tests and a wide variety of exposures.

Air-insulated all-metal chimneys are available for low-heat use in residential construction. Thermosiphon air circulation or multiple reflective shielding with three or more walls keeps these units cool. Insulated, double-wall residential chimneys are also available. The annulus between metal inner and outer walls is filled with insulation and retained by coupler end structures for rapid assembly.

Figure 25. Building Heating Appliance, Medium-Heat Chimney

Prefabricated, air-insulated, double-wall metal chimneys for multifamily residential and larger buildings, classed as building heating appliance chimneys, are available (Figure 25). Refractory-lined prefabricated chimneys (medium-heat type) are also available for this use.

Commercial and industrial incinerators, as well as heating appliances, may be vented by prefabricated metal-jacketed cast refractory chimneys, which are listed in the medium-heat category and are suitable for intermittent flue gas temperatures to 2000°F. All prefabricated chimneys and vents carrying a listing by a recognized testing laboratory have been evaluated for class of service regarding temperature, strength, clearance to adjacent combustible materials, and suitability of construction in accordance with applicable national standards.

Underwriters Laboratories (UL) standards, listed in Table 10, describe the construction and temperature testing of various classes of prefabricated vent and chimney materials. Standards for some related parts and appliances are also included in Table 10 because a listed factory-built fireplace, for example, must be used with a specified type of factory-built chimney. The temperature given for the steady-state operation of chimneys is the lowest in the test sequence. Factory-built chimneys under UL Standard 103 are also required to demonstrate adequate safety during a 1 h test at 1330°F rise and to withstand a 10 min simulated soot burnout at either 1630 or 2030°F rise.

These product tests determine minimum clearance to combustible surfaces or enclosures, based on allowable temperature rise on combustibles. They also ensure that the supports, spacers, and parts of the product that contact combustible materials remain at safe temperatures during operation. Product markings and installation instructions of listed materials are required to be consistent with test results, refer to types of appliances that may be used, and explain structural and other limitations.

Vent or chimney system design must consider the existence of or need for accessories such as draft diverters, draft regulators, induced-draft fans, blocking dampers, expansion joints, and vent or chimney terminals. Draft regulators include barometric draft regulators and furnace sequence draft controls, which monitor automatic flue dampers during operation. The design, materials, and flow losses of chimney and vent connectors are covered in previous sections.

 Draft Regulators

Appliances requiring draft at the appliance flue gas outlet generally use barometric regulators for combustion stability. A balanced hinged gate in these devices bleeds air into the chimney automatically when pressure decreases. This action simultaneously increases vent gas flow and reduces temperature. Well-designed barometric regulators provide constant flue gas static pressure over a span of impressed vent gas draft (i.e., the draft that would exist without regulation) of about 0.2 in. of water. A regulator can maintain 0.06 in. of water draft for impressed drafts from 0.06 to 0.26 in. of water. If the chimney system is very high or otherwise capable of generating available draft in excess of the pressure span capability of a single regulator, additional or oversize regulators may be used. Figure 26 shows proper locations for regulators in a chimney manifold.

Barometric regulators are available with double-acting dampers, which also swing out to relieve momentary internal pressures or divert continuing downdrafts. In the case of downdrafts, temperature safety switches actuated by hot gases escaping at the regulator sense and limit malfunctions.

Figure 26. Use of Barometric Draft Regulators

The selection of draft fans, blowers, or inducers must consider (1) types and combinations of appliances, (2) types of venting material, (3) building and safety codes, (4) control circuits, (5) gas temperature, (6) permissible location, (7) noise, and (8) power cost. Besides specially designed fans and blowers, some conventional fans can be used if the wheel and housing materials are heat- and corrosion-resistant and if blower and motor bearings are protected from adverse effects of the flue gas stream.

Figure 27. Draft Inducers

Small draft inducers for residential gas appliance and unit heater use are available with direct-drive blower wheels and an integral device to sense flow (Figure 27A). The control circuit for these applications must provide adequate vent gas flow both before and while fuel flows to the main burner. Other types of small inducers are either saddle-mounted blower wheels (Figure 27B) or venturi ejectors that induce flow by jet action (Figure 27C). An essential safety requirement for inducers serving draft hood gas appliances does not permit appliance interconnections on the discharge or outlet side of the inducer. This requirement prevents backflow through an inoperative appliance.

With prefabricated sheet metal venting products such as type B gas vents, the vent draft inducer should be located at or downstream from the point the vent exits the building. This placement keeps the indoor system below atmospheric pressure and prevents flue gas from escaping through seams and joints. If the inducer cannot be placed on the roof or outdoor wall, metal joints must be reliably sealed in all pressurized parts of the system.

Pressure capability of residential draft inducers is usually less than 1 in. of water at rated flow. Larger inducers of the fan, blower, or ejector type have greater pressure capability and may be used to reduce system size as well as supplement available draft. Figure 27D shows a specialized axial-flow fan capable of higher pressures. This unit is structurally self-supporting and can be mounted in any position in the connector or stack because the motor is in a well, separated from the flue gas stream. A right-angle fan, as shown in Figure 27E, is supported by an external bracket and adapts to several inlet and exit combinations. The unit uses the developed draft and an insulated tube to cool the extended shaft and bearings.

Pressure, volume, and power curves should be obtained to match an inducer to the application. For example, in an individual chimney system (without draft hood) in which a directly connected inducer only handles combustion products, calculation of the power required for continuous operation need only consider volume at operating flue gas temperature. An inducer serving multiple, separately controlled draft hood gas appliances must be powered for ambient temperature operation at full flow volume in the system. At any input, the inducer for a draft hood gas appliance must handle about 50% more standard chimney gas volume than a directly connected inducer. At constant volume with a given size inducer or fan, these demands follow the Fan Laws (see Chapter 21) applicable to power venting as follows:

Centrifugal and propeller draft inducers in vents and chimneys are applied and installed the same as in any heat-carrying duct system. Venturi ejector draft boosters involve some added consideration. An advantage of the ejector is that motor, bearings, and blower blades are outside the contaminated flue gas stream, thus eliminating a major source of deterioration. This advantage causes some loss of efficiency and can lead to reduced capacity because undersized systems, having considerable resistance downstream, may be unable to handle the added volume of the injected airstream without loss of performance. Ejectors are best suited for use at the chimney or vent exit or where there is an adequately sized chimney or vent to carry the combined discharge.

If total pressure defines outlet conditions or is used for fan selection, the relative amounts of static pressure and velocity pressure must be factored out; otherwise, the velocity head method of calculation does not apply. To factor total pressure into its two components, either the discharge velocity in an outlet of known area or the flow rate must be known. For example, if an appliance or blower produces a total pressure of 0.25 in. of water at 1500 fpm discharge velocity , the velocity pressure component can be found from Figure 7. Enter at the horizontal line marked V.H. in grid C, and move horizontally to ambient operating temperature. Read up from the appropriate temperature rise line to the 1500 fpm horizontal velocity line in grid D. Here the velocity pressure reads 0.14 in. of water (calculates to 0.143 at ambient standard conditions), so static pressure is 0.25 – 0.14 = 0.11 in. of water. Because this static pressure is part of the system driving force, it combines with theoretical draft to overcome losses in the system.

Draft inducer fans can be operated either continuously or on demand. In either case, a safety switch that senses flue gas flow or pressure is needed to interrupt burner controls if adequate draft fails. Demand operation links the thermostat with the draft control motor. Once flow starts, as sensed with a flow or pressure switch, the burner is allowed to start. A single draft inducer, operating continuously, can be installed in the common vent of a system serving several separately controlled appliances. This simplifies the circuitry because only one control is needed to sense loss of draft. However, the single fan increases appliance standby loss (especially in boilers) and heat losses via ambient air drawn through inoperative appliances.

The vent or chimney height and method of termination is governed by a variety of considerations, including fire hazard; wind effects; entry of rain, debris, and birds; and operating considerations such as draft and capacity. For example, the 3 ft height required for residential chimneys above a roof is necessary so that small sparks will burn out before they fall on the roof shingles.

Many vent and chimney malfunctions are attributed to interactions of the chimney termination or its cap with winds acting on the roof or with adjoining buildings, trees, or mountains. Because winds fluctuate, no simple method of analysis or reduction to practice exists for this complex situation. Figures 28 to 30 show some of the complexities of wind flow contours around simple structural shapes.

(Clarke 1967)

Figure 28. Wind Eddy and Wake Zones for One- or Two-Story Buildings and Their Effect on Chimney Gas Discharge

Figure 28 shows three zones with differing degrees of flue gas dispersion around a rectangular building: the cavity or eddy zone, the wake zone, and the undisturbed flow zone (Clarke 1967). In addition, a fourth flow zone of intense turbulence is located downwind of the cavity. Chimney flue gases discharged into the wind at a point close to the roof surface in the cavity zone may be recirculated locally. Higher in the cavity zone, wind eddies can carry more dilute flue gas to the lee side of the building. Flue gases discharged into the wind in the wake zone do not recirculate into the immediate vicinity, but may soon descend to ground level. Above the wake zone, dispersal into the undisturbed wind flow carries and dilutes the flue gases over a wider area. The boundaries of these zones vary with building configuration and wind direction and turbulence; they are strongly influenced by surroundings.

Figure 29. Height of Eddy Currents Around Single High-Rise Buildings

The possibility of air pollutants reentering the cavity zone because of plume spread or of air pollution intercepting downwind cavities associated with adjacent structures or downwind buildings should be considered. Thus, the design criterion of elevating the stack discharge above the cavity is not valid for all cases. Consult a meteorologist experienced with dispersion processes near buildings for complex cases and for cases involving air contaminants.

As chimney height increases through the three zones, draft performance improves dispersion, while additional problems of gas cooling, condensation, and structural wind load are created. As building height increases (Figure 29), the eddy forming the cavity zone no longer descends to ground level. For a low, wide building (Figure 30), wind blowing parallel to the long roof dimension can reattach to the surface; thus, the eddy zone becomes flush with the roof surface (Evans 1957). For satisfactory dispersion with low, wide buildings, chimney height must still be determined as if H = W (Figure 28).

Figure 30. Eddy and Wake Zones for Low, Wide Buildings

Chien et al. (1951) and Evans (1957) studied pitched roofs in relation to wind flow and surface pressures. Because the typical residence has a pitched roof and probably uses natural gas or a low-sulfur fossil fuel, dispersion is not important because combustion products are relatively free of pollutants. For example, NFPA Standard 54/AGA Z223.1 requires a minimum distance between the gas vent termination and any air intake, but it does not require penetration above the cavity zone.

Flow of wind over a chimney termination can impede or assist draft. In regions of stagnation on the windward side of a wall or a steep roof, winds create positive static pressures that impede established flow or cause backdrafts in vents and chimneys. Locating a chimney termination near the surface of a low, flat roof can aid draft because the entire roof surface is under negative static pressure. Velocity is low, however, because of the cavity formed as wind sweeps up over the building. With greater chimney height, termination above the low-velocity cavity or negative-pressure zone subjects the chimney exit to greater wind velocity, thereby increasing draft from two causes: (1) height and (2) wind aspiration over an open top. As the termination is moved from the center of the building to the sides, its exposure to winds and pressure also varies.

Terminations on pitched roofs may be exposed to either negative or positive static pressure, as well as to variation in wind velocity and direction. On the windward side, pitched roofs vary from complete to partial negative pressure as pitch increases from approximately flat to 30° (Chien et al. 1951). At a 45° pitch, the windward pitched roof surface is strongly positive; beyond this slope, pressures approach those observed on a vertical wall facing the wind. Wind pressure varies with its horizontal direction on a pitched roof, and on the lee (sheltered) side, wind velocity is very low, and static pressures are usually negative. Wind velocities and pressures vary not only with pitch, but with position between ridge and eaves. Reduction of these observed external wind effects to simple rules of termination for a wide variety of chimney and venting systems requires many compromises.

In the wake zone or any higher location exposed to full wind velocity, an open top can create strong venting updrafts. The updraft effect relative to wind dynamic pressure is related to the Reynolds number. Open tops, however, are sensitive to the wind angle as well as to rain (Clarke 1967), and many proprietary tops have been designed to stabilize wind effects and improve the performance. Because of the many compromises made in vent termination design, this stability is usually achieved by sacrificing some of the updraft created by the wind. Further, locating a vent cap in a cavity region frequently removes it from the zone where wind velocity could have a significant effect.

Performance optimization studies of residential vent cap design indicate that the following performance features are important: (1) still-air resistance, (2) updraft ability with no flow, and (3) discharge resistance when vent gases are carried at low velocity in a typical wind (10 fps vent velocity in a 20 mph wind). Tests in UL Standard 441 for proprietary gas vent caps consider these three aspects of performance to ensure adequate vent capacity.

Frequently, air supply to an appliance room is difficult to orient to eliminate wind effects. Therefore, the vent outlet must have a certain updraft capability, which can help balance a possible adverse wind. When wind flows across an inoperative vent termination, a strong updraft develops. Appliance start-up reduces this updraft, and in typical winds, the vent cap may develop greater resistance than it would have in still air. Certain vent caps can be made with very low still-air resistance, yet exhibit excessive wind resistance, which reduces capacity. Finally, because the appliance operates whether there is a wind or not, still-air resistance must be low.

Some proprietary air ventilators have excessive still-air resistance and should be avoided on vent and chimney systems unless a considerably oversized vent is specified. Vertical-slot ventilators, for example, have still-air resistance coefficients of about 4.5. To achieve low still-air resistance on vents and chimneys, the vertical-slot ventilator must be 50% larger than the diameter of the chimney or vent unless it has been specifically listed for such use.

Freestanding chimneys high enough to project above the cavity zone require structurally adequate materials or guying and bracing for prefabricated products. The prefabricated metal building heating appliance chimney places little load on the roof structure, but guying is required at 8 to 12 ft intervals to resist both overturning and oscillating wind forces. Various other expedients, such as spiral baffles on heavy-gage freestanding chimneys, have been used to reduce oscillation.

The chimney height needed to carry the effluent into the undisturbed flow stream above the wake zone can be reduced by increasing the effluent discharge velocity. A 3000 fpm discharge velocity avoids downward eddying along the chimney and expels the effluent free of the wake zone. Velocity this high can be achieved only with forced or induced draft.

Rain entry is a problem for open, low-velocity, or inoperative systems. Good results have been obtained with drains that divert the water onto a roof or into a collection system leading to a sump. Figure 31 shows several configurations (Clarke 1967; Hama and Downing 1963). Runoff from stack drains contains acids, soot, and metallic corrosion products, which can cause roof staining. Therefore, these methods are not recommended for residential use. An alternative procedure is to allow all water to drain to the base of the chimney, where it is piped from a capped tee to a sump.

Rain caps prevent vertical discharge of high-velocity flue gases. However, caps are preferred for residential gas-burning equipment because it is easier to exclude rain than to risk rainwater leakage at horizontal joints or to drain it. Also, caps keep out debris and bird nests, which can block the chimney. Satisfactory vent cap performance can be achieved in the wind by using one of a variety of standard configurations, including the A cap and the wind band ventilator, or one of the proprietary designs shown in Figure 31. Where partial rain protection without excessive flow resistance is desired, and either wind characteristics are unimportant or windflow is horizontal, a flat disk or cone cap 1.7 to 2.0 diameters across located 0.5 diameter above the end of the pipe has a still-air resistance loss coefficient of about 0.5.

See Chapter 46 of the 2019 ASHRAE Handbook—HVAC Applications for additional information on vent and chimney termination and wind effects.

Building and installation codes and standards prescribe the installation and safety requirements of heat-producing appliances and their vents and chimneys. Chapter 52 lists the major national building codes, one of which may be in effect in a given area. Some jurisdictions either adopt a national building code with varying degrees of revisions to suit local custom or, as in many major metropolitan areas, develop a local code that agrees in principle but shares little common text with the national codes. Familiarity with applicable building codes is essential because of the great variation in local codes and adoption of modern chimney design practice.

The national standards listed in Table 11 give greater detail on the mechanical aspects of fuel systems and chimney or vent construction. Although these standards emphasize safety aspects, especially clearances to combustibles for various venting materials, they also recognize the importance of proper flow, draft, and capacity.

The following conversion factors have been simplified for chimney design

Figure 31. Vent and Chimney Rain Protection