Effects of airflow infiltration on the thermal performance of internally insulated ducts
Introduction
Heat gains and losses to duct systems in residential and commercial buildings have been shown to strongly influence the energy efficiency with which conditioned air is delivered to the occupied space. Based both on measurement and simulation, Palmiter and Francisco [5] estimated that heat pump systems in the Pacific Northwest might suffer a 10% increase in seasonal energy requirements from conduction losses from ducts located in crawlspaces. In a very different climate, Parker et al. [6] predicted through a detailed simulation that peak residential duct system heat gains could approach 33% of available cooling system capacity under peak conditions when ducts were located in an attic. Jump et al. [2] performed detailed measurements that determined that supply-duct conduction reduced residential space conditioning efficiency by 16% in California homes tested.
A supply air duct may contain a fiberglass lining for acoustic control and thermal insulation. As conditioned air travels through a supply duct, heat exchange between the air and the duct's surroundings reduces the air's “thermal capacity,” or rate at which it can heat or cool the space to which it is delivered. The magnitude of this thermal gain or loss is inversely proportional to the duct's total thermal resistance, which is the sum of the resistance of the duct's insulated wall and the resistances of the boundary-layer air films inside and outside the duct's wall. Increasing the resistance of the duct's insulation will reduce the thermal gain or loss from the duct, and thereby raise the fraction of the supply air's inlet thermal capacity delivered to the duct's outlet. This ratio of outlet capacity to inlet capacity is the duct's “delivery effectiveness.”
Air flowing through a duct will infiltrate internal fiberglass insulation if the insulation's air-facing surface is pervious. Infiltration induces forced convection within the fiberglass, raising its effective thermal conductivity and lowering its thermal resistance. Encapsulating the insulation's air-facing surface with an impervious barrier prevents infiltration of the insulation and degradation of its thermal performance.
The increase in delivery effectiveness induced by encapsulating the insulation's surface is the duct's “effectiveness gain.” Given the variation with air speed of its insulation's conductivity, a duct's effectiveness gain can be calculated for arbitrary duct geometries, duct air speeds, and exterior ambient conditions.
Only one report of the variation with duct air speed of the total resistance of fiberglass-insulated ductwork was found in the literature. Lauvray [3] reported that the total conductance of a flexible duct with internal fiberglass insulation was invariant at air speeds below 5 m s−1, and rose linearly with air speed at speeds above 5 m s−1. The study did not report the diameter of the duct, the thickness of its insulation, the emissivity of its outer surface, or the speed and temperature of the ambient air. Thus, it is difficult to generalize the reported variation of total thermal resistance, or to calculate the variation with air speed of the insulation's thermal conductivity. No reports of the variation with duct air speed of the conductivity of internal fiberglass insulation were found.
This study determined the air-speed dependence of the conductivity of fiberglass insulation by measuring the inlet-to-outlet temperature drop of heated air as it traveled at various speeds through a long, insulated flexible duct. The results were used to simulate the effectiveness gains obtainable by encapsulating the air-facing surface of the insulation inside ducts in residential and commercial systems. The simulations modeled flexible and rigid ducts, hot and cold air supplies, and duct locations inside and outside the building's thermal envelope.
The temperature-drop conductivity measurement technique requires a long, narrow duct to obtain a good ratio of signal to noise in the observed temperature difference. Flexible branch ducts are manufactured in lengths of up to 15 m, and are typically insulated with low-density fiberglass blankets. Rigid main ducts — e.g., rectangular sheet-metal trunk ducts — are typically insulated with high-density fiberglass blankets, and are not usually manufactured in long lengths. The high-density blankets are less permeable to air than are low-density blankets, and their conductivities are expected to vary less with duct air speed.
Since it was more convenient to obtain a long run of insulated flexible duct than a long run of insulated rigid duct, the conductivity measurements were performed on low-density, flexible-duct insulation. The permeabilities of the low- and high-density blankets were measured, and their ratio used to theoretically extrapolate the air-speed variation of the conductivity of the high-density, rigid-duct insulation from that measured for the low-density blanket.
Section snippets
Thermal capacity
To maintain the air in a conditioned room at constant temperature and humidity, the net influx of enthalpy from the inflow of supply air and outflow of room air must equal the room's net thermal load. If the room's airflow is balanced, this net enthalpy influx is the supply air's thermal capacity,Here ṁa is the mass flow rate of the dry-air component of the supply air, and H and HR are the enthalpies/unit mass dry air of the supply air and room air, respectively.
Delivery effectiveness
The effectiveness
Overview
The total resistance of a long, flexible duct with internal fiberglass insulation was measured by blowing hot air through the duct at various speeds, then measuring the air's bulk velocity and the steady-state values of the duct inlet, duct outlet, and ambient air temperatures. These data were used to compute (a) the duct's total resistance R and (b) the insulation's resistance Rf, conductivity kf, reference-temperature conductivity kf*, still-air reference-temperature conductivity kf 0*, and
Overview
The gain in delivery effectiveness achieved by encapsulating the pervious air-facing surface of a supply duct's internal fiberglass insulation depends on
(a) duct properties (length, cross-section, and outer surface's long-wave emissivity);
(b) insulation properties (thickness, still-air conductivity, and sensitivities of conductivity to temperature and velocity);
(c) duct-exterior conditions (air temperature and velocity);
(d) inlet-air conditions (temperature and humidity); and
(e) room-air
Conclusions
The measured conductivity of a flexible duct's low-density internal fiberglass-blanket insulation increased with the square of the duct air speed, rising by 140% as the duct air speed increased from 0 to 15 m s−1. At air speeds recommended for branch ducts, the conductivity of low-density flexible-duct insulation would increase by 6% above its still-air value in a residential system, and by 16% in a commercial system.
The conductivity of a rigid duct's high-density internal fiberglass insulation
Acknowledgements
This work was supported by the Assistant Secretary for Energy Efficiency and Renewable Energy, Office of Building Technology and Community Systems, of the U.S. Department of Energy under Contract No. DE-AC03-76SF00098 and by the California Institute For Energy Efficiency. Thanks to Paul Berdahl, William Fisk, and Mor Duo Wang for their assistance in this study.
References (7)
Handbook of Fundamentals, SI Edition
(1997)- et al.
Field measurements of efficiency and duct retrofit effectiveness in residential forced air systems
Experimental heat transmission coefficients for operating air duct systems
ASHRAE J.
(1978)
Cited by (7)
Impact of ductwork airtightness on fan energy use: Calculation model and test case
2018, Energy and BuildingsCitation Excerpt :There are a number of studies that demonstrate significant energy use impacts of ductwork leakage [1–8].
Porous materials in building energy technologies—A review of the applications, modelling and experiments
2018, Renewable and Sustainable Energy ReviewsCitation Excerpt :The relevant experimental and numerical studies are presented here in two separate sections. Levinson et al. [63] obtained the relationship between air-velocity and the thermal conductivity of fiberglass (a porous material) insulation in an internally insulated duct by obtaining the inlet-to-outlet temperature drop of warmed air passing inside the duct. They found that the measured conductivity of a flexible channel low density inner fiberglass-blanket insulation is intensified by about 140% as the duct air velocity increases from 0 to 15 m/s. Dimoudi et al. [29] conducted experiments on a wall fabricated by two main sub-layers containing the dynamic insulation and ventilated outer envelope sub-layers.
The inaccuracy of heat transfer characteristics of insulated and non-insulated circular duct while neglecting the influence of heat radiation
2009, Energy Conversion and ManagementThe effect of a cool-coating on the thermal effectiveness of exposed exterior ducts
2007, Energy and BuildingsCitation Excerpt :These “fitted” thermal insulations are low compared to the nominal thermal insulations based on observed insulation type and thickness. Thermal insulations degradation due to air intrusion into the insulation would only be expected to be about 10% [5,6]. It should be noted that the fitted thermal insulation is determined by regression of data from daylight hours only, and it include the effects of all simplified assumptions such as neglecting the radiation heat exchange between the ducts and surrounding surfaces.
Assessing the impact of pressure drop and leak distribution on ductwork airtightness measurements
2023, International Journal of Ventilation