A perspective on the effect of climate and local environmental variables on the performance of attic radiant barriers in the United States

ARTICLE IN PRESS

Building and Environment 41 (2006) 1767–1778 www.elsevier.com/locate/buildenv

A perspective on the effect of climate and local environmental variables on the performance of attic radiant barriers in the United States Mario A. Medina, Bryan Young Civil Environmental and Architectural Engineering Department, The University of Kansas, Lawrence, KS 66045, USA Received 29 October 2004; received in revised form 30 November 2004; accepted 15 July 2005

Abstract This paper offers a perspective on how climate and local environmental variables affect the performance of attic radiant barriers across the United States. Transient heat and mass transfer simulations were performed on a vented triangular attic with insulation level of 3:5 m2 K=W (R-19) and the results were based on integrated hourly ceiling heat fluxes over 3-month periods during the cooling season. The ceiling heat transfer percent reductions ranged from 36.8% in the Tropical Savanna climate to 2.3% in the Mediterranean climate. Peak-hour percent reductions in ceiling heat flux ranged from almost 100% in the Marine West Coast climate to 23% in the Desert climate. The results suggested that local ambient air temperature, humidity, cloud cover index, and altitude had first-order effects. The amount of local solar radiation had no effect on the performance of the systems. r 2005 Elsevier Ltd. All rights reserved. Keywords: Radiant barriers; Building energy conservation; Building heat transfer; Climate and environment

1. Introduction The value of radiant barriers, thin aluminum sheets installed in attic spaces, in relation to reducing the ceiling heat transfer to and from conditioned spaces has been documented for years [1–18]. Several conclusions were drawn from the referenced literature. These are summarized below as follows: (a) Radiant barriers do contribute to a reduction of heat transfer rate in attics when compared to attics without radiant barriers. The percent reductions varied from approximately 40–45% if insulation with a resistance value of 1:95 m2 K=W (R-11) was used to approximately 15–20% in the case when the attics had insulation with a resistance value of 5:28 m2 K=W (R-30). Corresponding author. Tel.: +1 785 864 3604; fax: +1 785 864 5631. E-mail address: [email protected] (M.A. Medina).

0360-1323/$ - see front matter r 2005 Elsevier Ltd. All rights reserved. doi:10.1016/j.buildenv.2005.07.018

(b) Radiant barriers installed over the attic ceiling frame (horizontal radiant barrier, HRB) consistently reduced the heat transfer rate by approximately 5% more than a radiant barrier installed against the rafters that support the roof (truss radiant barrier, TRB), even in cases when the end-gables were covered. The HRB also used less material per attic structure, but its use has been discouraged because of dust accumulation. (c) Radiant barrier effectiveness was not increased by it having two low-emissivity surfaces instead of one. (d) The TRB reduced the heat transfer rate as effectively whether the low-emissivity surface was facing down or facing up. (e) Ceiling heat flux reductions as a result of using radiant barriers were larger in nonvented attics when compared to vented attics. However, once a small threshold ventilation rate of 1:27 ðl=sÞ=m2 ð0:25 CFM=ft2 Þ was achieved, radiant barrier effectiveness was independent of airflow.

ARTICLE IN PRESS M.A. Medina, B. Young / Building and Environment 41 (2006) 1767–1778

1768

(f)

(g)

(h)

(i)

Similarly for natural ventilation, increasing the ventilation area from 1/500 (opening/ceiling area) to 1/300 produced no extra reductions in heat flux. Different attic ventilation arrangements produced different percent heat flux reductions. Soffit/ridge combinations produced larger savings than soffit/ soffit or soffit/gable combinations. This was attributed to the degree of stratification of the attic air. Radiant barrier effectiveness was independent of the amount of insolation that impinged on the roof sections. For a daily insolation greater than 1600 kJ/ day (1500 Btu/day), the ceiling heat flux reductions varied slightly with increasing total daily insolation. Radiant barrier performance was independent of the kind of existing insulation (fiberglass, rock wool, cellulose and/or mineral wool) in the attic as long as the thermal resistance values of the insulation were the same. Cooling season ceiling heat flux reductions produced by the radiant barriers were higher than heating season reductions (e.g.,
35% for summer, and approximately 15% for winter if insulation with a resistance value of 3:35 m2 K=W (R-19) was used).

Although the previous research demonstrates that radiant barriers are effective in reducing ceiling heat transfer, the performance of radiant barriers as a function of local climate is still missing. As this technology continues to be adopted by the construction industry and as the number of homeowners accepting and installing radiant barriers in their homes increases, the question remains as to how much this measure can be expected to reduce the ceiling heat transfer in specific areas of the country. This paper is the result of a study, based on computer simulations, of how radiant barriers would perform across the various climates and local environmental conditions found in the Continental United States.

Midsummer temperatures are highest in the southwestern desert and inland close to the Gulf Coast, where afternoon readings surpassing 38  C ð100 FÞ are not uncommon. In the central region of the country, afternoon temperatures range from about 35  C ð95 FÞ north of the Gulf of Mexico to about 27  C ð81 FÞ near the Canadian border. Sea breezes contribute in keeping more moderate temperatures along the Pacific and Atlantic coasts. Mountain peaks in the west are generally the coolest locations in summer; on the East Coast, mountain locations are also cooler than the surrounding Piedmont areas, but not as high as in the west. A detailed distribution of summer Mainland temperatures is shown in Fig. 1, which is a plot of summer mean ambient air temperatures. The temperatures depicted in Fig. 1 represent the most recent summer data archived by the US National Weather Service [19]. Other graphical representations depicting humidity and wind information are shown in Fig. 2 (mean summer relative humidity), and Fig. 3 (mean summer wind speeds). The data of Figs. 2 and 3 are also from [19].

Fig. 1. Summer mean ambient air temperatures in F (source: [19]).

1.1. Climates in the Continental United States The climates in the Mainland of the United States are categorized as Continental, Marine, and Mountain. Table 1 shows the major climates and the factor that controls each climate. An extreme development of the continental climate is the desert climate.

Table 1 Major climates and their controlling factors Climate

Controlling factor

Continental Marine Mountain

Land Ocean Altitude

Fig. 2. Summer mean ambient relative humidity in percent (source: [19]).

ARTICLE IN PRESS M.A. Medina, B. Young / Building and Environment 41 (2006) 1767–1778

By means of temperature, humidity, wind speed, and precipitation regimes the three main climates in the United States are then subdivided into nine climatic regions, four that are located east of the 100  W meridian and five located west of it. The Tropical Savanna (TS), Humid Subtropical (HS), Humid Continental Warm Summer (HCWS), and the Humid Continental Cool Summer (HCCS) climates are found east of the 100  W meridian; the Western High Areas (WHA), Desert (D), Mediterranean (M), Steppe (S), and Marine West Coast (MWC) climates are found to the west of it [17]. The climate subdivision map is shown in Fig. 4. The Tropical Savanna climate is characterized by hot climate with hot and humid summers and with one distinct dry season (winter). The average monthly summer temperature and relative humidity are around 28  C ð82 FÞ and 77%, respectively. The Humid Subtropical climate is a warm, temperate, and rainy climate

1769

with no distinct dry season. Average summer temperature and relative humidity for this region are 29  C ð84 FÞ and 68%, respectively. The Humid Continental Warm Summer region is mainly a cold climate, though summers are warm and humid. The Humid Continental Cool Summer region and Western High Areas are similar climates. They are characterized by cold climates with cool summers and humid winters. The Steppe climate is a mid-latitude cold and dry region in which there is usually a winter drought. Its summers are cold and dry. In the Desert climate precipitation is meager year round and temperatures are high. The months of April through July are very dry. A classic Mediterranean climate has three times as much rain in the wettest month of winter as in the driest month of summer. It has cool and pleasant summers. The Marine West Coast is a warm, temperate, rainy climate. Most of the precipitation occurs in the winter, but there is little during the summer months. Table 2 provides summer weather variables for the nine climates. These are the summer mean monthly values of temperature, relative humidity, and wind speed. It also includes area covered.

2. Description of the computational model

Fig. 3. Summer mean wind speeds in miles per hour (source: [19]).

A transient heat and mass transfer computer model for an attic structure was used in this study. The attic used in the development of the model was composed of two equal-length pitched roof sections, two vertical gable-ends, and one horizontal ceiling frame. The attic geometry and the corresponding thermal and mass exchanges are shown in Fig. 5. The model was based on

Fig. 4. Climatic regions of the Continental United States (source: [17]).

ARTICLE IN PRESS M.A. Medina, B. Young / Building and Environment 41 (2006) 1767–1778

1770

Table 2 Average summer weather data for the nine climates found in the Continental United States Climate

Humid Subtropical Humid Continental Warm Summer Desert Humid Continental Cool Summer Steppe Marine West Coast Mediterranean Western High Areas Tropical Savanna

Summer monthly dry bulb air temperature ð CðFÞÞ

Summer monthly relative humidity (%)

Summer monthly wind speed (km/h (mi/h))

29 (84) 25 (77) 28 (83) 21 (70) 17 (62) 15 (59) 17 (63) 20 (68) 28 (83)

68

13.7 (8.5) 14.1 (8.8) 13.0 (8.1) 14.0 (8.7) 12.7 (7.9) 13.3 (8.3) 16.1 (10.0) 13.7 (8.5) 12.9 (8.0)

70 47 67 43 80 74 50 77

Area covered

Percent area covered

ðkm2 ðmi2 ÞÞ

(%)

1,939,636 (750,430) 1,655,112 (640,350) 1,223,467 (473,350) 905,291 (350,250) 739,043 (285,930) 560,259 (216,760) 508,837 (196,865) 481,581 (186,320) 59,484 (23,014) 8,072,711 (3,123,269)

Total

24.03 20.50 15.16 11.21 9.15 6.94 6.30 5.97 0.74 100.00

Source: [18].

Long-wave Radiation

Solar Radiation

Convection

Transient Conduction Convection Latent Load

Infra-red Radiation Convection

Convection Transient Conduction Radiation Attic Air

Fig. 5. Attic geometry and thermal and mass exchanges (source: [18]).

the first law of thermodynamics. It calculated instantaneous sensible and latent cooling and heating loads based on energy balance equations written for each enclosing surface and for attic air layers. A full

description of the model is found in [13] and its verification against experimental data is found in [14]. The input to the program included attic dimensions, radiation constants (outer surface absorptivities, outer

ARTICLE IN PRESS M.A. Medina, B. Young / Building and Environment 41 (2006) 1767–1778

and inner surface emissivities), percent surface covered by wood, permeance of attic components and wood moisture content, longitude, latitude and time zone of the location. The hourly data included outdoor air temperature, global horizontal solar radiation, wind speed, relative humidity (or dew point), and cloud cover index. A set of linear equations was developed, which set up a matrix of the form (1)

MT ¼ P.

Matrix M contained information about the conduction response factors, convection and radiation coefficients, and attic air stratification. Matrix P contained information about previous values of heat fluxes at each surface and historical values of all surface temperatures, indoor air temperature, outdoor air temperature, solar radiation at each surface, sky temperature, convection and radiation coefficients, conduction response factors and the common ratios. The matrix T was solved for surface and attic air temperatures. The computer model was driven using Typical Meteorological Year 2 (TMY2) weather data for 15 stations within the climatic regions. A ceiling insulation with a thermal resistance of 3:35 m2 K=W (R19) was assumed. The output of the model included hourly ceiling heat fluxes for both the radiant-barrier case and the no-radiant-barrier case. 2.1. Effects of environmental variables on attic thermal exchanges 2.1.1. Conduction heat transfer process in attics Conduction heat transfer in attics occurs at all bounding surfaces. The conduction process is transient because the temperatures on all surfaces vary with time. These surface temperatures are driven mainly by a combination of solar loads, ambient air temperature, and wind currents. Because the temperature difference between the outdoor and indoor environments is significantly larger than temperature differences on the same surface planes, conduction heat transfer in attics was approximated as one-dimensional transient conduction. This approximation simplified the conduction problem and is valid under most circumstances unless metal framing and metal supports are used; however, this is not commonly the case for residential attics. Therefore, the conduction process was handled using conduction response factors as presented in [20]. Under this methodology the heat fluxes were found using q00cond;out ¼

N;S X

Y i;j ðT sii;nDj  T r Þ

j¼0;i¼1



N;S X j¼0;i¼1

X i;j ðT soi;nDj  T r Þ þ CRi q00oði;nD1Þ ð2Þ

1771

for outgoing heat flux, and N;S X

q00cond;in ¼

Z i;j ðT sii;nDj  T r Þ

j¼0;i¼1



N;S X

Y i;j ðT soi;nDj  T r Þ þ CRi q00iði;nD1Þ

ð3Þ

j¼0;i¼1

for incoming heat flux. In Eqs. (2) and (3), X , Y , Z represented conduction response factors, CR was the common ratio, and T si ; T so ; T r were the indoor surface temperature, the outdoor surface temperature, and the reference temperature, respectively. Also, in Eqs. (2) and (3), i; j, and nD denoted surface number, time, and time step, respectively. The q00o and q00i in Eqs. (2) and (3) were previous heat fluxes in the outside and inside surfaces, respectively. 2.1.2. Convection heat transfer in attics Energy flows in residential attics are influenced to a degree by the amount of heat that is transported by means of convection. At every surface, convection heat transfer can be significant. During a typical summer day, heat is transported away from outer surfaces by convection to the wind that flows around the structure. Also, the attic ventilating air carries away heat from the attic structures which otherwise would end up in the conditioned space. Both forced and natural convection are present and make up the combined convective heat transfer coefficient. Convection in attics is mixed, meaning that both the effects of buoyancy forces as well as the effects of forced flows are present. Convection in attics is also both laminar and turbulent, depending on the conditions present at a particular surface in question. For example, laminar-natural and laminar-forced convection dominate convective heat transfer in the triangular end-gables, while laminarforced and turbulent-natural make up the mixed convection coefficient in the attic floor. To account for both laminar and turbulent effects, correlations found in [21] were used. To combine forced and natural effects, the correlation proposed by Churchill [22] was modified according to Chen [23] so it could be used in this case. The resulting equation was n

n

n

Nu ¼ NuF  NuN , n

(4) n

where Nu is the average mixed Nusselt number, NuF n the average Nusselt number, NuN the average natural Nusselt number, and n the constant exponent (a value of 3 is suggested). 2.1.3. Radiation heat transfer in attics Radiation heat transfer is the dominant force of the heat flux which actually enters the conditioned space through the ceiling of the residence. Radiation heat transfer in attic structures was handled by using

ARTICLE IN PRESS 1772

M.A. Medina, B. Young / Building and Environment 41 (2006) 1767–1778

radiation enclosure theory. The major assumptions used when modeling attic enclosures were that attic surfaces were isothermal, gray, and diffuse. The graybody assumption was a reasonable assumption to make because both the emitted and the incident radiation were confined to the same wavelength range (infra-red), and in attics the spectral emittance is relatively constant within this range. The diffuse assumption required that all radiation emitted and reflected from any attic surface be diffusively distributed. In attic structures, both the emitted and reflected radiation are indistinguishable and there was no need to separate them. This assumption enabled the calculations to be based on radiosities (the rate at which radiant energy leaves a surface), which are uniform along the surface when the surface is isothermal. Also, the fact that radiosities are uniform along any attic surface made the geometric view factors independent of the magnitude and surface distribution of the radiant energy flux. 2.1.4. Solar radiation on attic outer surfaces The driving force responsible for increasing the temperature of the shingles to 65–75 C ð
150–167 F) is solar radiation. This energy is eventually conducted through the roof sections and later radiated from the attic deck to the attic floor. Therefore, this is one of the major energy sources responsible for increasing the spacecooling load of the residence. Radiant barrier systems have the potential of preventing most of this energy (from the attic deck) from being transferred to the ceiling floor. Consequently, the model was designed to handle the solar loads on the attic outer surfaces. Solar radiation on ‘‘tilted’’ surfaces is seldom measured, but the global radiation on a horizontal surface is measured at most weather stations around the country. Therefore, the solar radiation on a horizontal surface was used to estimate the solar loads on inclined attic surfaces. Both the beam (or direct) component as well as the diffuse component were separated so that the direct component of global radiation could be projected onto the surface via the angles that described the orientation of the surface and its relationship to the sun. The diffuse component remained unchanged regardless of surface orientation. 2.1.5. Moisture transport across attic structures Moisture is an integral part of the environment, and as a result, in humid climates, one of the major loads during the cooling seasons is from moisture exchange. In winter, it is the condensation of moisture on the cold surfaces that creates changes in surface temperature and heat fluxes which cannot be estimated correctly under a ‘‘dry’’ air assumption. Moisture is also important in predicting degradation and deterioration of attic structures as well. Latent effects in attics stem from various sources including moisture migration within structures, moisture transport by the ventilating air, as well as

moisture adsorption/desorption at the interior surfaces. In this model, latent effects were incorporated via a condensation/evaporation theory proposed by Burch and Treado [24] and was supplemented with suggestions from [25–27]. The moisture balance on the attic was written as S X

Ai Permi ðPo;i  Pattic

air Þ

surface i

¼ Q_ air rair ðoattic þ

S X

air

 oo Þ

Ai hw;i ðoattic

air

 ow;i Þ,

ð5Þ

surface i

where Ai is the area of surface i ðm2 ; ft2 Þ, Perm the watervapor permeance of ceiling (kg=Pa s m2 , lbm=h ft2 in Hg), Po;i the water-vapor partial pressure of outer surface of surface i ðPa; in HgÞ, Pattic air the water-vapor partial pressure of attic air (Pa, in Hg), Qair the attic airflow rate ðm3 =s; ft3 = minÞ, rair the ventilating air density ðkg=m3 , lbm=ft3 ), oattic air the humidity ratio of attic air, oo the humidity ratio of outside air, ow;i the humidity ratio of wood surface i, and hw;i the mass transfer coefficient at wood surface i ðkg=m2 s, lbm=ft2 hÞ. 2.2. A measure of radiant barrier performance A measure of the performance of an attic radiant barrier is the percentage reduction in ceiling heat transfer that it produces. Once the ceiling heat fluxes were obtained from running the computer model, the percentage reductions were found using Percent reduction R ¼

Evaluation period

R

q00no

RB

dt 

R

Evaluation period 00 q Evaluation period noRB dt

q00RB dt

, ð6Þ

q00RB

is the ceiling heat flux when a radiant barrier where was present in the attic ðW=m2 ðBtu=h ft2 ÞÞ and q00no RB the ceiling heat flux when there was no radiant barrier present in the attic ðW=m2 ðBtu=h ft2 ÞÞ.

3. Results and discussion Sample profiles of ceiling heat fluxes for selected stations in the climatic regions are presented, which describe the performance of the radiant barriers. The profiles serve as a useful analytical tool that helps in understanding how radiant barriers work in various climates. Although the performance of attic radiant barriers was evaluated continuously during the summer months (June–August), for reasons of clarity only three typical summer days (July 28 to July 30) are shown in Figs. 6–8.

ARTICLE IN PRESS M.A. Medina, B. Young / Building and Environment 41 (2006) 1767–1778 6.5

Tropical Savanna (July 28-30)

Humid Subtropical (July 28-30)

1773

Desert (July 28-30)

5.5

Ceiling Heat flux (W/m 2)

4.5 3.5 2.5 1.5 0.5 -0.5 0

12

0

12

0

12

0

12

0

12

0

12

0

12

0

12

0

12

-1.5 -2.5 -3.5

Miami, FL

San Antonio, TX

-4.5

Tucson, AZ

Time of Day Standard Attic

Attic w/ Radiant Barrier

Fig. 6. Ceiling heat flux profiles for Tropical Savanna, Humid Subtropical, and Desert climates for July 28–30.

6.5 5.5

Ceiling Heat flux (W/m 2)

4.5

Humid Continental Cool Summers (July 28-30)

Humid Continental Warm Summers (July 28-30)

Western High Areas (July 28-30)

Steppe (July 28-30)

3.5 2.5 1.5 0.5 -0.5 0

12 0

12

0

12

0

12

0

12

0

12

0

12

0

12

0

12

0

12

0

12

0

12

-1.5 -2.5 -3.5

Minneapolis, MN

Topeka, KS

Boulder, CO

Helena, MT

-4.5

Time of Day Standard Attic

Attic w/ Radiant Barrier

Fig. 7. Ceiling heat flux profiles for Humid Continental Cool Summer, Humid Continental Warm Summer, Western High Areas, and Steppe climates for July 28–30.

3.1. Tropical Savanna, Humid Subtropical, and Desert climates Results from these three climates were lumped together because of similarities in radiant barrier performance. Table 3 shows relevant summer average

ambient air temperature, ambient air relative humidity, hourly global horizontal solar radiation, and sky cloud cover index for three cities, each representing a climate. As depicted in Fig. 6, the key similarity is that under these climates heat transfer is mostly always into the conditioned space. That is, attic temperatures are

ARTICLE IN PRESS M.A. Medina, B. Young / Building and Environment 41 (2006) 1767–1778

1774

Fig. 8. Ceiling heat flux profiles for Marine West Coast and Mediterranean climates for July 28–30.

Table 3 Summer weather data for Miami (FL), San Antonio (TX), and Tucson (AZ) Climate

Tropical Savanna Humid Subtropical Desert

Station

Miami, FL San Antonio, TX Tucson, AZ

Average temperature

Average hourly global horizontal solar radiation ðW=m2 ðBtu=h ft2 ÞÞ

Average sky cloud cover index

ð C ðFÞÞ

Average relative humidity (%)

28 (82) 27 (81) 29 (84)

74 66 37

478 (152) 533 (169) 536 (170)

0.57 0.49 0.40

Adapted from [18].

generally always higher than the temperatures in the conditioned space. Radiant barrier performance profiles prove the usefulness of the technology in these climates as it helps in reducing the ceiling heat gains, which increase during periods of high solar activity. During these periods radiant barriers, because of their low emissivity and absorptivity, decrease the infrared radiation from the attic deck to the top of the insulation on the attic floor. It is worthwhile to mention that in the afternoons the interior surfaces of the attic deck reached temperatures in the range of 54  C ð129 FÞ to 60  C ð140 FÞ The profiles in Fig. 6 show that in the Tropical Savanna climate radiant barriers are useful throughout the entire 24 h of the day. In the Humid Subtropical and in the Desert climates radiant barriers are mainly useful during the hours of 10 a.m. through 6 p.m., but still help somewhat in decreasing the heat transfer rate during nights and early mornings. That is, best possible performance is achieved during the periods of high solar

radiation. The largest heat flux reduction during the on-peak hour is produced in the Tropical Savanna climate; 42% compared to 31% and 24% in the Humid Subtropical and Desert climates, respectively. The overall 3-month integrated ceiling reductions in heat transfer in the three climates were 36.8%, 34.3%, and 23.0% for the Tropical Savanna, Humid Subtropical, and Desert climates, respectively. From the weather data of Table 3, these results are not obvious because mean air temperatures are almost identical and mean hourly global horizontal solar radiation and cloud cover fraction are least and highest, respectively, in the Tropical Savanna climate. The reason for these results is the higher humidity of the Tropical Savanna climate. This means that higher solar loads alone do not drive the performance of radiant barriers, but instead a combination of solar load, which drives the ambient air temperatures, and humidity. In humid climates an evaporation process, which result from deposited moisture, takes place in the attic

ARTICLE IN PRESS M.A. Medina, B. Young / Building and Environment 41 (2006) 1767–1778

surfaces. Evaporation produces a cooling effect in the attic surfaces, which seems to be greater in the attic fitted with radiant barriers, thus producing the largest difference in ceiling heat fluxes between the no radiant barrier case and the radiant barrier case. In the Tropical Savanna climate this process continues until the late hours of the day. This translates to largest observed ceiling heat transfer reductions in this climate.

1775

temperatures of the previous three climates, while the solar loads are comparable, the peak-hour heat flux reductions are higher in the four climates of this section. The peak-hour ceiling heat flux reductions are 54%, 46%, 44%, and 36% for the Humid Continental Cool Summer, Humid Continental Warm Summer, Western High Areas, and Steppe climates, respectively. However, the overall 3-month integrated percent reductions for the same climates are 25.7%, 30.0%, 19.7%, and 13.7%, respectively, which are much lower than their equivalent values for the previous three climates. The reason for these lower integrated values, particularly in the Western High Areas and the Steppe climates, is the zero contribution of the radiant barriers during nights and early morning hours.

3.2. Humid Continental Cool Summer, Humid Continental Warm Summer, Western High Areas, and Steppe climates In these climates, heat transfer across the ceiling travels in both directions, to and from the conditioned space, as a result of ambient air temperature, which drops below the indoor air temperature of 24  C (75 F). This is evident from the summer mean air temperatures shown in Table 4. The performance profiles of Fig. 7 show that radiant barriers are effective in reducing heat transfer across the ceiling during the daytime hours, when the solar loads are high, but are not effective during the night and early morning hours. The profiles further indicate that outdoor temperature falls below the indoor temperature of 24  C (75 F) for approximately half of the day. Cool nights and mornings are characteristic of these climates likely because of the enhanced sky radiation, which is further increased by atmospheric dryness. Although the summer mean air temperatures in these climates are approximately 7  C (13 F) lower than the

3.3. Marine West Coast and Mediterranean climate Table 5 shows the mean summer weather data for Marine West Coast and for the Mediterranean climates. Fig. 8 shows that on a typical afternoon the hourly percentage reduction in ceiling heat flux for the Marine West Coast climate is about 100% and approximately 97% for the Mediterranean climate. The radiant barriers block nearly all heat transfer into the conditioned space. The heat loss from the residences across the ceilings is practically the same in both the radiant barrier case and the no radiant barrier case in both the Marine West Coast and Mediterranean climates. This means that if space heating were required during the nighttime and early mornings in these climate, having radiant barriers installed would not translate into more costs due to space heating.

Table 4 Average summer weather data for Minneapolis (MN), Topeka (KS), Boulder (CO), and Helena (MT) Climate

Humid Continental Cool Summer Humid Continental Warm Summer Western High Areas Steppe

Station

Average temperature

Average hourly global horizontal solar radiation ðW=m2 ðBtu=h ft2 ÞÞ

Average sky cloud cover fraction

ð C ðFÞÞ

Average relative humidity (%)

Minneapolis, MN

21 (70)

68

489 (155)

0.52

Topeka, KS

23.3 (74)

67

511 (162)

0.45

Boulder, CO Helena, MT

20.5 (69) 18 (65)

49 48

501 (159) 523 (166)

0.50 0.47

Adapted from [18].

Table 5 Average summer weather data for Astoria (OR) and San Francisco (CA) Climate

Station

Average temperature ð C ðFÞÞ

Average relative humidity (%)

Average hourly global horizontal solar radiation ðW=m2 ðBtu=h ft2 ÞÞ

Average sky cloud cover fraction

Marine West Coast Mediterranean

Astoria, OR San Francisco, CA

15 (59) 15.5 (60)

77 73

394 (125) 542 (172)

0.73 0.34

Adapted from [18].

ARTICLE IN PRESS 1776

M.A. Medina, B. Young / Building and Environment 41 (2006) 1767–1778

Ocean breezes are characteristics of these climates. Thus, as cool ventilation air at temperatures below 24  C (75 F) enters the attic the temperature of the attic, T attic , is reduced and T insulation  T attic is increased. The convected heat transfer between the attic air and the insulation is given by q00conv ¼ hc jðT insulation  T attic Þj

(7)

Chen et al. [10], proved that hc is greater for upward heat flow. These two factors tend to bring about an increase in convection heat transfer, q00conv , from the top of the insulation to the attic air. Under these circumstances, the radiant barriers are unable to prevent heat loss from the conditioned space. Therefore, there is no significant difference between the ceiling heat flux for the radiant barrier case and no radiant barrier case. 3.4. Summary Table 6 presents the performance of attic radiant barriers in selected cities representative of the nine climatic regions. The performance is given in terms of heat flux comparisons as summer integrated percent reductions (SIPR) and peak hour percent reduction (PHPR). The SIPR represents the reduction in heat transfer over a season while the PHPR represents the instantaneous (or demand) reduction during the hottest hour (peak hour) of day. It is evident that performance of attic radiant barriers depends on the climate in which the building is located. For the climates considered, the summer integrated percent reduction in ceiling heat flux varied from 2.3% in the Mediterranean climate to 36.8% in the Tropical Savanna climate. The peak-hour heat flux reductions varied from 23% in the Desert climate to about 100% in the Marine West climate. The

results reveal that attic radiant barriers will prove to be useful in all the climates. Fig. 9 is a graphical representation of Table 6. That is, it depicts the performance of attic radiant barriers across the Continental United States in terms of SIPR and PHPR. This map represents a tool that could be used to estimate the percentage reduction in ceiling heat flux that would be achieved by installing a radiant barrier in a building located in any of the contiguous 48 states.

4. Conclusions The objective of this study was to determine the influence of climate on the performance of attic radiant barriers in the Continental United States. To achieve this objective, the Continental United States was classified into nine climatic regions. The climatic regions were Tropical Savanna, Humid Subtropical, Humid Continental Warm Summer, Humid Continental Cool Summer, Steppe, Desert, Western High Areas, Mediterranean, and Marine West Coast. Fifteen cities were selected to represent the regions. A standard vented triangular attic design was used for each simulation. The ceiling insulation was assumed to have a resistance value of 3:35 m2 K=W (R-19). A heat and mass transfer computer model for an attic structure was then used to determine the hourly ceiling heat fluxes when there was no radiant barrier and then when there was a radiant barrier. Fifteen weather stations were selected to represent the nine climatic regions. The analysis was done by integrating continuous hourly heat fluxes for the months of June, July and August. The percentage reduction in ceiling heat flux was then found for this period for each station. The results showed the

Table 6 Performance data of attic radiant barriers in the Continental United States Climate

Humid Subtropical

Humid Continental Warm Summer Desert Humid Continental Cool Summer Steppe Marine West Coast Mediterranean Western High Areas Tropical Savanna

Sample station

San Antonio, TX New York-NY Atlanta, GA Topeka, KS Indianapolis, IN Las Vegas, NV Tucson, AZ Minneapolis, MN Detroit, Michigan Pocatello, ID Helena, MT Astoria, OR San Francisco, CA Boulder, CO Miami, FL

Sample summer integrated percent reduction (SIPR) (%) 34.3 32.5 38.5 30.0 30.1 19.2 23.0 25.7 24.3 16.0 13.7 9.6 2.3 19.7 36.8

Average

Peak-hour percent reduction (PHPR) (%)

35.1

31

30.5

46

21.1

23

25.0

54

14.9 9.6 2.3 19.7 36.8

36
100 97 44 42

ARTICLE IN PRESS M.A. Medina, B. Young / Building and Environment 41 (2006) 1767–1778

1777

100

100 50 100

50 0

50

SIPR PHPR

0 SIPR PHPR

0 SIPR PHPR 100 100 50 50 0 100

SIPR PHPR

0 SIPR PHPR

50 0

100

100

50

50

0

SIPR PHPR

SIPR PHPR 0 SIPR PHPR

100 50 0 SIPR PHPR

Fig. 9. Map depicting summer-integrated percent reduction (SIPR) and peak-hour percent reduction (PHPR) performance of attic radiant barriers across the Continental United States.

percentage reduction in ceiling heat flux differed for each climate, which implied that the performance of attic radiant barriers depends on the climate in which the building is located. The lowest summer integrated percent reduction in ceiling heat transfer was 2.3% in the Mediterranean climate and the highest was 36.8% in the Tropical Savanna climate. The performance in the Continental Warm Summer and Humid Subtropical climates were also high. The peak-hour ceiling heat flux reductions ranged between 23% for the Desert climate to almost 100% for the Marine West Coast climate. In summary, the results revealed that attic radiant barriers would prove useful in all of the above climates. Of course, buildings in some climates would benefit more than others in different climates. It was discovered that amongst the local environmental parameters the local ambient air temperature, humidity, cloud cover index, and altitude had first order effects. The amount of local solar radiation had no effect on the performance of the systems.

Acknowledgements The authors acknowledge the financial support provided by the University of Kansas Energy Research Center and the assistance of Mr. Michael Frempong.

References [1] Fairey PW. The measured side-by-side performance of attic radiant barrier systems in hot, humid climates. Proceedings of the 19th international thermal conductivity conference. 1985. [2] Katipamula S, O’Neal DL. An evaluation of the placement of radiant barriers on their effectiveness in reducing heat transfer in attics. Proceedings of the third annual symposium on improving building energy efficiency in hot and humid climates. 1986. p. 68–77. [3] Levins WP, Karnitz MA, Knight DK. Cooling energy measurements of unoccupied single-family houses with attics containing radiant barriers. Proceedings of the third annual symposium on improving building energy efficiency in hot and humid climates. 1986. p. 78–87. [4] Lear WE, Barrup TE, Davis KE. Preliminary study of a vented attic radiant barrier system in hot, humid climates using side-byside, full-scale test houses. Proceedings of the fourth annual symposium in improving building energy efficiency in hot and humid climates. 1987. p. 195–8. [5] Ober DG, Volckhausen TW. Radiant barrier insulation performance in full-scale attics with soffit and ridge venting. Proceedings of the fifth annual symposium on improving building energy efficiency in hot and humid climates. 1988. p. 174–85. [6] Goss WP, Miller RG. Literature review and predictions of reflective building insulation system performance, 1900–1989. ASHRAE Transaction 1989. [7] Hall JA. Performance testing of radiant barriers (RB) with R-11, R-19, and R-30 cellulose and rock wool insulation. Proceedings of the fifth annual symposium on improving building energy efficiency in hot and humid climates. 1989. p. 174–85.

ARTICLE IN PRESS 1778

M.A. Medina, B. Young / Building and Environment 41 (2006) 1767–1778

[8] Yarbrough D, Cook D, Wilkes KE. Contamination of reflective foils in horizontal applications and its effect on thermal performance. ASHRAE Conference. 1989. [9] Bourne MA, Hoeschele R. Climate influence on attic radiant barriers and demand savings. American council for an energyefficient economy summer study on energy efficiency in buildings. 1990. p. 1.5–.13. [10] Chen H, Larson T, Erikson RW. The performance of an attic radiant barrier for a simulated Minnesota winter. Proceedings of the thermal performance of the exterior envelopes of buildings conference. 1992. [11] Medina MA, O’Neal DL, Turner WD. Effect of attic ventilation on the performance of radiant barriers. Journal of Solar Energy Engineering 1992;114(1):234–9. [12] Medina MA, O’Neal DL, Turner WD. Effects of radiant barrier systems on ventilated attics in a hot and humid climate. Proceedings of the eighth annual symposium on improving building energy efficiency in hot and humid climates. 1992. p. 47–52. [13] Medina MA, O’Neal DL, Turner WD. A transient heat and mass transfer model of residential attics used to simulate radiant barrier retrofits—part I: development. Journal of Solar Energy Engineering 1998;120(1):32–8. [14] Medina MA, O’Neal DL, Turner WD. A transient heat and mass transfer model of residential attics used to simulate radiant barrier retrofits—part II: validation and simulations. Journal of Solar Energy Engineering 1998;120(1):39–44. [15] Medina MA. Effects of shingle absorptivity radiant, barrier emissivity, attic ventilation flowrate, and roof slope on the performance of radiant barriers. International Journal of Energy Research 2000;24(8):665–78. [16] Medina MA. On the performance of radiant barriers in combination with different attic insulation levels. Energy and Buildings 2000;33(1):31–40.

[17] Frempong M. Influence of climate on the performance of attic radiant barriers in the Continental United States. Master’s thesis, Architectural Engineering Department, The University of Kansas; 2001 [18] Medina MA, Frempong M. Evaluation of ceiling heat fluxes in residential buildings with attic radiant barriers in prevalent climates across the United States. Proceedings of the architectural engineering conference. 2003. [19] US National Weather Service US ‘‘summary of day data’’. Online. 2005 [20] Mitalas GP, Stephenson DG. Thermal response factors. ASHRAE Transactions 1967;73(PartI). [21] ASHRAE. Handbook of fundamentals. Atlanta, GA: American Society of Heating, Refrigeration, and Air Conditioning Engineers. 1989 [22] Churchill SW. A comprehensive correlating equation for laminar assisting. forced and free convection. AIChE Journal 1977;23:10–6. [23] Chen TS, Armaly BF, Ramachandran N. Correlations for laminar mixed convection flows on vertical, inclined, and horizontal flat plates. Journal of Heat Transfer 1986;108: 835–40. [24] Burch DM, Treado SJ. Ventilating residences and their attics for energy conservation–experimental study. In: Reppart MH, editor. Summer attic and whole-house ventilation, vol. 548. NBS Special Publication; 1978. p. 73–104. [25] Burch DM, Lemay MR, Rian BJ, Parker EJ. Experimental validation of an attic condensation model. ASHRAE Transactions 1984;20(Part 2A):59–77. [26] Cleary PG. Moisture control by attic ventilation—an in-situ study. ASHRAE Transactions 1985;91(Part 1):227–39. [27] Wilkes KE. Analysis of annual thermal and moisture performance of radiant barrier systems. ORNL/CON-319. Oak Ridge, TN: Oak Ridge National Laboratory; 1991.