1、NA-04-6-3 An Analysis of the Effective R-value for Insulation Buried Attic Ducts Dianne Griffiths, P.E. Associate Member ASHRAE ABSTRACT This paper presents the initial analytical work performed to quantify the energy benejts of burying ducts under loosejll attic insulation and assess the sensitivit
2、y of these benejts to diflerent installation conjgurations. Results of the analysis are expressed in terms of an insulation wrap R-value that conven- tional hung ducts would require to achieve un equivalent ther- malperformance to insulation buriedducts. Theperformances of larger trunk-type ducts an
3、d smaller branch-type ducts installed over or between truss chords in attics with diflerent levels of loosejlljbcrglass und cellulose were examined. The various conjgurations simulated were used to developgeneral guidelines for predicting the performance of buried ducts. Thesejndings formed the basi
4、s for a buried duct credit to be incorporated in the 2005 revision to California S Title 24. INTRODUCTION Several studies performed more than a decade ago revealed the significant impact that residential distribution system inefficiencies could have on the overall space condi- tioning system efficie
5、ncy (Cummings et al. 1990; Modera 1989; Palmiter and Bond 1992; Proctor and Pernick 1992; Robison and Lambert 1989). Andrews (2003) summarizes the findings of earlier research by stating that “duct systems in unconditioned spaces typically lose 25% to 40% of the energy output from the space-conditio
6、ning equipment, with leakage and conductive losses contributing comparable amounts.“ More recently, much research has been conducted to support the development of Standard 152 (ASHRAE 2003), a test method for determining residential thermal distribution system efficiencies. Marc Zuluaga Through educ
7、ation and conservation programs, progress has been made in many regions to improve duct system tight- ness and reduce the inefficiency associated with duct leakage (DOE 1999). With greater emphasis on duct air sealing and the use of mastic, low air leakage duct systems are attainable (Hoeschle and C
8、hitwood 2003). Now greater emphasis on reducing conductive losses is warranted. In fact, Andrews (2003) analyzed the even greater impact of conductive duct losses for systems with modulating furnaces or air handlers. During the past three years, the research team has been working with a large Sacram
9、ento area HVAC contractor. When this research was initiated, the standard HVAC system in the region was an attic-mounted horizontal gas furnace attached to round insulated flexible duct with an R-value of 4.2 ft2.h.“F/Btu (0.74 m2.K/W) hung throughout the attic (see Figure i). Duct leakage values of
10、 less than 6% of the air- handler flow were typical. The researchers have encouraged the implementation of a rather simple way to reduce conductive heat gains and losses-burying the ducts under the blown attic insulation rather than hanging them. It should be emphasized that this concept, as present
11、ed here, is only appropriate for dry climates. Covering the ducts with insulation lowers the temperature at the vapor barrier jacket surface. In humid climates, this temperature could be below the dew point of the attic air and condensation could occur. A field study performed in Melbourne, Florida
12、(Griffiths et al. 2002), where the 1% design dew point is 79F (26“C), demonstrated this condition. The authors believe that climates with 1% design dew points below 70F (21C) may be considered dry for the purposes of considering this buried duct concept. Dianne Griffiths and Marc Zuluaga are mechani
13、cal engineers at Steven Winter Associates, Inc., in Nonvalk, Conn. 02004 ASHRAE. 72 1 Figure 1 Attic-mounted furnace with flexible duct distribution. METHODOLOGY To help quantify the energy benefits of the buried duct concept, initial analytical work was performed utilizing a two- dimensional steady
14、-state finite element heat transfer model shown schematically in Figure 2. Several duct sizes and insu- lation depths were modeled, as well as whether the duct rested directly on the ceiling gypsum or on the bottom chord of the roof truss. The model developed is equally valid at assessing the perfor
15、mance of buried ducts in the summer and in the winter. However, since a reduction in duct conduction losses will be most significant during design conditions, and peak load performance is typically more important to cooling than heating, the notations in this work are cooling season oriented. The fo
16、llowing modeling assumptions were made: A 75F (24C) conditioned-space temperature. A constant temperature boundary condition of 55F (13C) was applied to the surface on the inside of the duct to simulate conditioned supply air. 1.25 in. (0.032 m) thick duct insulation wrap has an R-value of 4.2 ft2.h
17、.“F/Btu (0.74 m2.K/W), represent- ing standard California construction practice at the time of this study. A convection boundary condition with a constant heat transfer coefficient of 1.76 ft2.h.“F/Btu (10.0 m2.K/W) was applied to the top surface of the loose fill insulation and to the bottom surfac
18、e of the gypsum board. This value is in the range of commonly accepted convection coefficients for interior building surfaces in the absence of experimental measurement. Radiant heat transfer effects were not accounted for in this analysis. Since hung ducts have greater exposed sur- face areas than
19、buried ducts, the omission of radiant heat gains results in a conservative estimate of the thermal performance benefits of burying the ducts. Loose Fill Attic Air Insulation Insulating Duct WniP Gypsum Board Ceiling Conditioned Space Figure 2 Schematic of conjguration modeled. Duct leakage is zero.
20、All duct gains and losses are due io thermal heat transfer only. R-values are not a function of attic temperature. The thermal conductivity of the loose fill insulation can be varied to represent either loose fill fiberglass or cellulose. The thermal conductivity of loose fill cellulose and fiber- g
21、lass depends on their blown-in density. The R-value of loose fill fiberglass can vary from 2.2 ft2.h.“F/Btu (0.39 m2.K/W) per inch to 2.9 ft2.h.F/Btu (0.51 m2.K/W) per inch, while the R-value of loose fill cellulose can range from 3.1 ft2.h.“F/Btu (0.55 m2.K/W) per inch to 3.7 ft2.h.“F/Btu (0.65 m2.
22、K/W) per inch (ASHRAE 2001). For this study, average values of 2.5 ft2.h.“F/Btu (0.44 m2.K/W) per inch for fiberglass and 3.4 ft*.h.“F/Btu (0.60 m2.WW) per inch for cellulose were used. For a conventional hung duct, the duct UA is expressed as where Q = duct heat gain, Tujjic = attic temperature, an
23、d Tducj = conditioned air temperature within the duct. To evaluate the energy benefits of buried ducts, it would be useful to be able to determine the equivalent R-value with which conventional hung ducts must be wrapped to achieve the same thermal performance as buried ducts. Such an R- value canno
24、t simply be calculated by taking the inverse of buried duct UA. For a buried duct, Q is composed of two components: the detrimental heat gain from the attic (eujjic) 722 ASHRAE Transactions: Symposia and the nondetrimental heat gain from the conditioned space below (Qroom). L? = Qattic + Qroom = uA(
25、attic - duct) + Qroom (2) Duct Inner Diameter 10 in. (0.25 m) 6 in. (O. 15 m) Therefore, when comparing to attic hung ducts, the detri- mental heat gain that we should be concerned with is simply the Qattic portion of the total duct gain Q. Qattic is a function of Tattic, while Qroom is constant for
26、 constant duct and room air temperatures. Although Qroom is typically much smaller than Qu, it cannot be neglected in the buried duct energy balance. This analysis assumes that the space conditioning system is adequately sized to maintain room conditions for any and all attic conditions. The linear
27、relationship expressed in Equation 2 defines the effective duct R-value. This effective value is the R-value with which a conventionally hung duct must be wrapped in order to result in an equivalent amount of detrimental heat gain. Qroom does not affect it even though it is a component of the energy
28、 balance. Fiberglass Insulation Depth Cellulose Insulation Depth 12.5 in. (0.32 m) 16 in. (0.41 m) 12.5 in. (0.32 m) 16 in. (0.41 m) (R-31 attic) (R-40 attic) (R-43 attic) (R-54 attic) 14 28 16 34 25 35 31 43 Deeply Buried + 3+ (0.09m) Fully Buried Conditioned Space Figure 3 Buried duct classificati
29、ons. RESULTS AND DISCUSSION Applying the analytical technique described above, effec- tive duct R-values were calculated for a variety of insulation and duct configurations. More specifically, whether the duct is “deeply,” “fully,” or “partially” buried is examined, as well as duct sizes and whether
30、 the duct is resting on the gypsum board ceiling or run over the attic truss chords. In this study, “deeply buried” indicates that the depth of the loose fill insulation is 3.5 in. (0.09 m) higher than the top of the insulating duct wrap. “Fully buried” indicates that the depth of the loose fill ins
31、ulation is even with the top of the insu- lating duct wrap. “Partially bured” indicates that the depth of the loose fill insulation is 3.5 in. (0.09 m) lower than the top of the insulating duct wrap. A schematic illustrating these classifications is presented in Figure 3. Results are presented as th
32、ermal resistance R-values in Table 1 for different duct sizes. A 10-in. (0.25 m) ID duct was chosen to represent a main supply trunk, while a 6-in. (0.15 m) ID duct was also simulated to represent a branch off a main supply trunk. It is noteworthy that when a 10-in. (0.25 m) duct is fully buried und
33、er 12.5 in. (0.32 m) of insulation, the effective duct R-value with fiberglass is roughly equivalent to the effective duct R-value with cellulose. Yet when a 10-in. (0.25 m) ID duct is buried in 16 in. (0.4 1 m) of fiberglass (attic R-value of 40 ft2.h.”F/Btu), the effective R-value is much higher t
34、han when the same duct is buried in 12.5 in. (0.32 m) of cellulose (attic R-value of 43 ft*.h.”F/Btu). The smaller 6-in. (O. 15 m) ID duct is deeply buried in both 12.5 in. (0.32 m) and 16 in. (0.41 m) of attic insulation, and the effective R-values are correspondingly higher than the 10-in. (0.25 m
35、) ID duct. Thus, the critical parameter to maximizing the effective R- value of a buried duct is the degree to which it is buried, not the attic R-value. The degree to which ducts are buried is primarily affected by the thickness of the insulation. Since for the same attic R- value, a greater thickn
36、ess of fiberglass is required, it is clear that fiberglass is a better material than cellulose for obtaining the greatest thermal benefits from buried ducts. It is also obvi- ous that when burying ducts, duct thermal distribution e%- ciency is increased when the same attic R-value is obtained Table
37、I. Effective Duct R-values for Fully and Deeply Buried Duct Resting on Gypsum Ceiling (ft*.h.”F/Btu) I l I I Notes: All ducts have an R-4.2 insulating wrap. To convert R-values from I-P units (ft*.h.”F/Btu) to SI units (m2.KIW), multiply by 0.176. ASHRAE Transactions: Symposia 723 using a lower inst
38、allation density e., for fiberglass an R- value of 2.2 ft2.h.“F/Btu per inch 15.4 m.K/W instead of 2.5 ft2.h.“F/Btu per inch 17.3 m.WW). However, using a low installation density of loose fill insulation could also result in decreased attic R-values in areas where there is not much clearance due to
39、sloping roofs. The above simulations were performed with the buried ducts resting on a gypsum ceiling. It was also of interest to determine the effective R-values of buried ducts run over engi- neered truss chords. To this end, simulations were conducted with a 10-in. (0.25 m) ID duct on top of a 2
40、in. x 4 in. (0.05 m x O. 1 O m) truss chord and with a 10 in. (0.25 m) ID duct on top of 3.5 in. (0.09 m) of loose fiil insulation representing an insu- lated cavity between tniss chords. The depth of the loose fill insulation was chosen to be 16 in. (0.41 m) for both cases. Thus, for the same attic
41、 R-value, the duct is now fully buried instead of deeply buried as in the earlier case with the duct rest- ing on the gypsum. Figure 4 presents the situation schemati- cally, and the results of these simulations are presented in Table 2. For this particular example, raising a duct out of the attic i
42、nsulation by resting it on either a 2 in. x 4 in. (0.05 m x O. 10 m) engineered truss chord or 3.5 in. (0.09 m) of loose fill insulation results in a 50% reduction in effective R-value compared to when that duct is more deeply buried when in contact with the gypsum board. This significant penaly in
43、the effective duct R-value underscores the benefits of designing a low profile duct system such that the main supply trunk can be placed between truss chords. In such a design, the branches off from the main supply trunk would run over the truss chords. However, due to their smaller size, these duct
44、s would still be able to be buried under several inches of loose fill insulation in a fiberglass attic with an R-value of 40 ft2.h,“F/Btu. Finally, it was of interest to determine the benefits of only partially burying ducts. To investigate this scenario, a 10 in. (0.25 m) ID duct running over a 2 i
45、n. x 4 in. (0.05 m x O. 10 m) truss chord was placed in 12.5 in. (0.32 m) of loose fill insu- lation. The results of these simulations are also presented in Table 2. The effective R-values for this case are not as low as might be expected. Since heat gains from the conditioned space do not impact th
46、e effective R-value calculation, it is logical that the effective R-values of the two duct configurations are equal-regardless Fiberglass Insulation of the fact that one rests on 3.5 in. (0.09 m) of loose fill insu- lation while the other rests on a truss chord. Simulations were also conducted with
47、partially, fully, and deeply buried 6-in. (O. 15 m) ID ducts in fiberglass and cellu- lose. For a particular buried duct classification and loose fill insulation type, effective R-values calculated for the 6 in. (O. 15 m) ID case were slightly lower than those computed for a 10 in. (0.25 m) ID duct.
48、 The lower effective R-values calcu- lated for the 6 in. (0.15 m) duct are likely due to the fact that the buried duct classifications are expressed in terms of loose fill insulation level relative to the top of a duct. Thus, while the tops of a deeply buried small duct and large duct will be equall
49、y buried, the bottom of the larger duct will be buried in more insulation than that of the smaller duct. In fiberglass, this geometry effect was found to result in an 1 1% to 14% decrease in the effective R-values of a 6 in. (0.15 m) duct compared to a IO in. (0.25 m) duct of the same classification. In cellulose, this geometry effect was found to result in an 1 1 % to 19% decrease in the effective R-values for a 6 in. (0.15 m) duct compared to a 10 in. (0.25 m) duct of the same classification. Cellulose Insulation Attic Air Duct Configuration Over truss chord Loose Fill
