Infrared thermovision technique for the assessment of thermal transmittance value of opaque building elements on site

Energy and Buildings 42 (2010) 2177–2183

Contents lists available at ScienceDirect

Energy and Buildings journal homepage: www.elsevier.com/locate/enbuild

Infrared thermovision technique for the assessment of thermal transmittance value of opaque building elements on site Rossano Albatici a,∗ , Arnaldo M. Tonelli b a b

University of Trento, Department of Civil and Environmental Engineering, Via Mesiano 77, 38123 Trento, Italy Via Miramonti 4, 38068 Rovereto (Tn), Italy

a r t i c l e

i n f o

Article history: Received 12 October 2009 Received in revised form 17 June 2010 Accepted 12 July 2010 Keywords: Infrared thermovision technique Thermal transmittance Heat flowmeter Heat losses Energy requirements

a b s t r a c t Due to new regulations in the field of energy saving, international standards concerning energy requirements of buildings have been developed. In this field, during the design phase, one of the most important parameters to be considered is the value of heat losses coefficient through the envelope. Anyway, very often a great difference is experienced between predicted performance of building elements as calculated using the methods prescribed in technical standards, and the one considering as-build conditions, particularly in the field of building renovation where the envelope thermal transmittance value must be measured on site. Till now, the only method accepted by international standards is the heat flowmeter (HFM) that presents some restriction and some uncertainty in the results. In this study a faster and less invasive method is proposed, the infrared thermovision technique (ITT) whose full potentiality has never been investigated yet, in order to acquire quantitative data of real thermal transmittances of the building envelope in a quasi-steady state condition. The theoretic background is presented together with the application in three case studies. The results indicate that, following a specific methodology, it is possible to record significant data useful to perform a proper assessment of energy performance of existing buildings. © 2010 Elsevier B.V. All rights reserved.

1. Introduction The European Directive 2002/91 on the energy performance of buildings [1], implemented in Italy with the L.D. 192/05 (whose implementing regulation has been recently issued with the D.P.R. 59/2009) later modified by the L.D. 311/06 [2], requires the energy certification in order both to achieve a high energy saving and to guarantee adequate indoor comfort conditions for the users. One of the most important parameters for the calculation of energy requirements of a building during the design phase is the value of heat losses through the envelope. In particular, for what concerns the opaque elements, it is important to calculate the thermal transmittance U-value [W/m2 K]. Very often, however, the difference between the theoretic thermal transmittance (calculated) and the real one (measured) may be very high. The result may be an overestimation of the energy performance of a building. Thus, it is important to define some simple but effective methods in order to estimate the actual U-value under certain condition of use, by means of on site measurement in real situation. This fact is nowadays particularly significant because

∗ Corresponding author. Tel.: +39 0461282622; fax: +39 0461282672. E-mail address: [email protected] (R. Albatici). 0378-7788/$ – see front matter © 2010 Elsevier B.V. All rights reserved. doi:10.1016/j.enbuild.2010.07.010

great attention is put forward the refurbishment of existing buildings and their optimization (new building activity in Europe is assumed to be only 1% of the total living area per year [3]). In order to develop an efficient design procedure it is necessary to start from true data most of all concerning the building envelope, the most energy dispersant element. The European Standard project prEN15203 [4] specifies and describes in detail the methods to be used on site in order to measure the thermal transmittance of the building envelope: the thermal flow meters method is expressly mentioned following the ISO 9869:1994 [5,6]. This method, however, is partially invasive and slow and the results strongly depend on the outer environmental conditions and on the users behavior in the building during the survey (whose collaboration in the management of the living rooms is strictly required). On the other hand, in this field infrared thermal technique has not been fully investigated yet, even if it has been used for more than 30 years in the building sector [7]. A lot of users’ guides have been published by specialized associations where potentialities and limits of the method are described [8,9] but it is always stated as a “qualitative” method [10–12]. Even Italian standard UNI 9252:1988 [13] states that: “The present Standard describes a qualitative test method, based on the infrared thermography, in order to survey and analyze thermal defects of the building envelope. . . The standard

2178

R. Albatici, A.M. Tonelli / Energy and Buildings 42 (2010) 2177–2183

does not apply for the definition of thermal insulation of the building and its air leakage, where other survey methods are required”. Anyway, recently some research groups do use IR thermography as a “quantitative” tool in laboratory in order to measure thermal diffusivity of bricks and plaster [7], but the possibility to measure building envelope parameters on site, as the thermal transmittance values, has never been investigated. Some authors expressly state that “Despite thermography’s many potentials, its application to building materials has not been greatly studied yet. The parameters that may affect measurements are not completely understood and interpreting the result becomes difficult and confusing” [14]. In particular, the primary limitation is the difficulty in measuring some parameters as superficial emissivity of finishing in existing walls on site [15,16] (while some laboratory surveys recently have given some good results [17]) and the proper interpretation of convective heat transfer coefficient on external walls [18–20] (recent researches studied in detail problems of internal surfaces [21,22]). However, some authors have already defined a procedure for the qualitative evaluation of thermal behavior of external walls in existing buildings with IR thermography, and the environmental limits concerning the carrying out of proper survey and the data analysis are known but not completely understood [14,23,24]. 1.1. Aim of the paper In this paper a new experimental methodology using the infrared thermovision technique (ITT) is proposed, so as to acquire quantitative data of real thermal transmittances of the building envelope in a quasi-steady state condition. Even if the Italian standard UNI 9252:1988 [13] states that the infrared thermovision technique is just a “qualitative” method that can be used only to assess and analyze the thermal irregularity of the building envelope but not to determine its thermal insulation level, the authors think that, if the thermographic survey is carried out by a technician with specific knowledge of thermotechnics and building physics and in a proper manner following the procedure described in this paper, the result of the test can be meaningful and complementary to the one made with the thermal flow meter method. Secondly, the ITT is faster (a complete building of medium size as 900 m3 can be surveyed in 1 day) and can be repeated in short time to verify previous results if necessary, even with different boundary conditions. Finally, it is a non-destructive method that has no influence on the normal life of the users during the monitoring campaign. Generally speaking, anyway, it must be considered that in situ measurement of the U-value of a building element is strongly influenced by some factors briefly summarized as follows: (a) Site conditions: - weather conditions during the test or during the previous period of time (with particular reference to wind speed, solar radiation, rain); - typical climate of the site, most of all humidity values that can significantly alter the thermal performance of the building materials. (b) Building conditions: - ageing of the building materials; - proper laying of the elements during the construction. (c) Operating conditions: - managing of the building by the users (heating/cooling and windows opening/closing); - maintenance works (done or not). While preparing the monitoring campaign and even during the evaluation of the ITT results, all these factors must be carefully considered.

2. The heat flowmeter method The heat flowmeter method (HFM) is based on some simplifying hypothesis. In particular, only plane and homogenous elements, uni-directional heat flow and heat transmission due to only conductive phenomena are considered. Convection (due to air temperature and speed) and irradiation (due to the temperature of the surfaces positioned near and around the element) can be simply considered together by means of a so called “environment temperature” that must be properly measured. The measuring equipment consists of: a heat flowmeter that is mounted directly on the face of the element adjacent to the more stable temperature (usually the inner side); at least two surface temperature sensors, one mounted on the inner surface of the element near the flowmeter, the other on the external surface opposite to the flowmeter; two air temperature sensors, one inside the other outside, at least 1.50 m from the floor level. Some practical rules to be followed during the test are: 1. the sensors must be mounted in such a way so as to ensure a result which is representative of the whole element (several flowmeters can be used in order to obtain a representative heat flow average); 2. thermal bridges, particular constructive joints and so on must be avoided; 3. the outer surface of the element should be protected from rain, snow and direct solar radiation; usually, the one with northern exposure is preferred, otherwise artificial screening may be used as well. The electrical data from the flowmeter should be recorded continuously or at fixed intervals over a period of a certain number of days. Generally speaking, the minimum test duration is 72 h if the temperature is stable around the flowmeter, otherwise it may be necessary more than 7 days; to be sure that a sufficient heat flow is present, the minimum difference between inner and outer temperature has to be of 10–15 ◦ C. The maximum time period between two records and the minimum test duration depends on: 1. nature of the element (heavy, light, multi-layers and so on); 2. inner and outer temperature variation (average and variation during and before the test); 3. methods used for data analysis. Usually, the test duration is between 3 and 8 days while data acquisition interval must be at least 15 min. As written before, during the test the minimum difference between inner and outer temperature has to reach 10–15 ◦ C. Data analysis can be made using two methods: average method and dynamic method. The main difference is that with the first one, greater is the number of data recorded higher is the possibility to get a correct final result. With the second method, the test duration can be shorter but higher indoor–outdoor temperature variation is needed. More information is provided in the ISO 9869 itself.

3. Infrared thermovision technique In situ measurement of the element U-value with heat flowmeter method is not always possible due to the great number of limitations. An alternative method is the infrared thermovision technique (ITT).

R. Albatici, A.M. Tonelli / Energy and Buildings 42 (2010) 2177–2183

2179

From (2), thermal transmittance become:

3.1. Equipment

4

The basic equipment consists of a thermograph with a spectral sensitivity of 8–12 ␮m (infrared thermal wavelength range) in order to take infrared thermal images of the building (both outside and inside), a heat source (for example a soldering iron) for the measurement of the surface emissivity, a cardboard box with a little hole or a hosepipe for the external temperature reading, a hot-wire anemometer, if possible, for a proper reading of the wind speed in proximity of the walls. 3.2. Theoretical basis

U =

4

5.67εtot ((Ti /100) − (Tout /100) ) + 3.8054v(Ti − Tout ) Tint − Tout [W/m2 K]

(6)

where Tint is the inner environment temperature [K]. All parameters except for v (that can be measured in the proximity of the wall using a hot-wire anemometer) can be measured using the same thermograph so as to minimize systematic measurement errors. 3.3. The measurement of emissivity ε

Thermal transmittance of a wall is given by the ratio between the rate of heat transfer through the element per unit area (1) and the difference Tint − Tout between inner and outer temperature (2):

Thermal power P, due to heat Q passing through the element, is dissipated from the outer element surface by means of conduction, convection, radiation. Contribution of conduction can be ignored. Considering the Stefan–Boltzman Law for grey body radiation:

The value of emissivity ε that is the emissivity of the outer surface of the element on site (so in real condition of use) in the spectrum range related to the thermograph used, is necessary in order to determine the element surface temperature Ti from the radiant temperature obtained from thermography. While εtot can be deduced from technical manuals, ε must be measured because it depends not only on the surface materials, but on the real surface conditions suffering from pollution, humidity, roughness and so on. The measurement of the emissivity ε value can be achieved with two methods: the comparison with a reference material (for example special adhesive tapes with known emissivity) or the direct measurement of the radiance reflected by the material. In fact, if N is the source power as “seen” within the range  in which the thermograph works, Kirchhoff’s Law states that

E = εT 4 [W/m2 ]

N = ε N +  N + T N

P=

Q [W/m2 ] At

(1)

P [W/m2 K] Tint − Tout

(2)

so: U=

(3)

where ε is the surface integral emissivity,  is the Stefan–Boltzman constant 5.67 × 10−8 [W/m2 K4 ], T is the surface temperature [K], and considering the sensible heat flux for convection as [18]: H = ˛c × (Ts − Tair )

(4)

where ˛c is the convective heat transfer coefficient [W/m2 K], Ts is the surface temperature and Tair is the air temperature, the following formula for the thermal power lost from the walls surface can be used (as the sum of E – thermal power dissipated for radiation, and H – thermal power dissipated for convection): P = 5.67εtot



Ti 100

4

−

T

out

100

4 

+ 3.8054v(Ti − Tout ) [W/m2 ] (5)

where P is the thermal power dissipated through the element [W/m2 ], εtot is the emissivity on the entire spectrum (integral emissivity), Ti is the surface temperature of the element [K], Tout is the outer environment temperature [K], 3.8054v is the convective heat transfer coefficient ˛c with v = wind speed [m/s]. Jurges’ equation was used to calculate ˛c based on the wind velocity near the building element at the time of the measurement.1

It expresses the principle of energy conservation: a radiant power N is partly absorbed (ε N ), partly reflected ( N ) and partly transmitted (T N ). For opaque surfaces, it is N = ε N +  N

Jurges’ equation is given by [25]

˛c = 5.8 + 3.8054v (v < 5 m/s) where ˛c is convective heat transfer coefficient and v is wind velocity near the building element. Being the air speed near the wall lower than 1 m/s and very often near 0 m/s, and being the surface-to-air temperature difference very small (lower than 5 K), the equation has been simplified otherwise convective phenomena would have been too relevant in respect of the radiative ones. Anyway, in this field many uncertainties are still alive and many questions still opened [26], above all for very low air speed and surface-to-air temperature difference [27]. Additional studies are now on going on a real scale experimental building in order to refine the convective heat transfer coefficient for the U measurement of external walls on site using the ITT method proposed.

(8)

so  N = 1 − ε N

(9)

that is ε = 1 − 

(10)

The method used consists in moving hot elements close to the wall surface, for instance a soldering iron (temperature around 400 K) can be used as heat source. It must be positioned near the surface under investigation at a distance of about 10 cm. A thermographical image is suddenly taken, so as to avoid a possible local heating of the surface due to the long exposition to the heat source. The radiant temperature of the source and of its reflected image visible on the element surface can be measured and compared. In fact, for Eq. (3) the power of the heat source N is proportional to (Ts 4 − Tout 4 ), where Ts is the source radiant temperature, while the power of the reflected image  N is proportional to (Tr 4 − Tout 4 ), where Tr is the reflected image radiant temperature. So, from (9):



1

(7)

ε = 1 −

Tr 4 − Tout 4



Ts 4 − Tout 4

(11)

Finally, the measurement of the surface temperature of the element can be obtained by the thermographical images adjusted with the found emissivity value. 3.4. Other temperature reading Both the outer temperature Tout and the temperature inside the building Tint are measured by means of the same thermograph so as to minimize systematic measurement errors.

2180

R. Albatici, A.M. Tonelli / Energy and Buildings 42 (2010) 2177–2183 Table 1 Materials and thermal characteristics of the layers of external walls in case study A (from outside to inside). Material

s [m]

c [kJ/kg K]

 [W/m K]

 [kg/m3 ]

Plaster Polystyrene Plywood Polystyrene Plywood Rock wool Air Plasterboard

0.02 0.03 0.01 0.035 0.01 0.04 0.06 0.018

0.84 1.380 2.092 1.380 2.092 0.836 1.004 0.837

0.9 0.037 0.144 0.037 0.144 0.04 0.333 0.21

1800 35 800 35 800 40 1 900

s: thickness; c: thermal capacity; : thermal conductivity; : density.

Table 2 U-value of the north fac¸ade using the HFM method in case study A. Fig. 1. Case Study A: Villa dei Girasoli, south fac¸ade.

The method consists in considering some elements acting as black bodies, one for the outer temperature and the other for the inner one. For the outer temperature, a good approximation of the theoretical black body can be a curved hosepipe, even partially coiled up, whose length is at least one order of magnitude more than its diameter. Otherwise, a cardboard box with a little hole can be used. Both of them should be put away from heat sources or from the building itself in order to minimize possible interactions. One hour in an open field in proximity of the building could be sufficient. For the inner temperature, thermographical image of the fac¸ade with a window partially and suddenly opened should be taken in order to record even inner environment. In this case, the black body is represented by the partial window opening, whose dimensions are limited with respect to the room whose temperature is detected. It must be well stressed that with the thermovision technique, data concerning radiation temperature are recorded. Thus, once the emissivity ε value is measured for every surface of the building elements, the data are corrected in such a way to acquire the actual value of the surface temperature Ti . About the recorded values of Tout and Tint , being the emissivity very close to the one of the black body, they can be considered coincident with the real ones. For what concerns the emissivity of the entire spectrum εtot (integral emissivity), values printed on specialized handbook can be used. 4. Case studies Here 3 case studies are presented where the ITT method has been applied. Two of them (case studies A and B) relate to light structured buildings, while the other one (case study C) relates to heavy structured building. In case study A the results of the ITT is compared with the one of the HFM. 4.1. Case study A “Villa dei Girasoli” is a single family house with a timber post and beam structure and light external walls placed in Gonars (45◦ 54 0 N–13◦ 14 0 E, Udine – Italy) near the Adriatic Sea, 21 m asl (Fig. 1). In Table 1 the characteristics of the layers of external walls are summed up. The total width of the wall is d = 0.223 m, the mass is m = 64.135 kg/m2 (very light wall). The calculated U-value is U = 0.29 W/m2 K.

Period

U-value [W/m2 K] Average method

U-value [W/m2 K] Dynamic method

12–12 to 17–12 26–12 to 3–1 1–1 to 7–1 4–1 to 7–1 23–1 to 27–1

0.68 0.68 0.72 0.80 0.71

0.49 0.54 0.74 0.40 0.48

In order to measure the real thermal transmittance of the wall in situ, the flowmeter method has been applied considering the North fac¸ade, while the infrared thermovision technique has been applied to all the four building sides. 4.1.1. Results with the HFM method The test has been carried on in January, the coldest period of the year in Italy. Data have been acquired every 15 min. Only some sub-periods suitable for the measurement (i.e. meeting the method conditions) have been considered. Data have been analyzed using both the average method and the dynamic one. Main results are summed up in Table 2. It is immediately clear that data analysis gives results strongly dependent on the method used. Moreover, even the same method does not converge, either considering different periods of time or considering different intervals in the same period. This result is probably due to the fact that the difference Tint − Tout between indoor and outdoor temperature during the test period has been often very low, sometimes less than 10 ◦ C in the 24 h. In this particular case the average method is not the most suitable, but only the results obtained with the dynamic one should be considered2 . Rejecting the value 0.74 that is too far from the others, the average U-value of the external wall measured with the HFM method is U = 0.46 W/m2 K. It is a value higher of 59% compared to the theoretic one. 4.1.2. Results with the ITT method The measurement of the emissivity value ε of the surface of the walls on site has been made following the procedure described in Section 3.3.The thermal profile of the soldering iron is highlight in Fig. 2.

2 In case study A, not always the difference between inner and outer temperature was up to 10 ◦ C and moreover inner wall temperature was not very stable (having the wall a lightweight structure). So, average method, even if converging, does not give appropriate results. Dynamic method, on the other hand, is very sophisticated and, considering a measurement time long enough, can give more appropriate results if the value of the index regarding the set of equations to be solved are properly chosen [6]. Secondly, with the dynamic method the test duration can be shorter and this gives the possibility to choose the best period to be considered between the time interval of measurement done.

R. Albatici, A.M. Tonelli / Energy and Buildings 42 (2010) 2177–2183

2181

Fig. 2. Thermal image of the soldering iron with the radiance temperature profile on the right end part.

Recorded data are:

Fig. 3. Case Study B: Veider House, south fac¸ade (a bioclimatic building with a huge sun space and wide window openings facing south).

1. source (soldering iron) temperature: Ts = 140 ◦ C = 413.14 K ; 2. apparent source temperature reflected by the wall: Tr = 30 ◦ C = 303.14 K ; 3. outdoor temperature: Tout = 20 ◦ C = 293.14 K.

ε = 1 −

303.144 − 293.144 413.144 − 293.144

= 0.9512

U-values calculated with (6) are summed up in Table 3 (data acquired from East fac¸ade have not been considered because of disturbing effects on radiation temperature caused by the presence of a grass just watered by the owners). The average U-value of the external wall measured with the ITT method is U = 0.38 W/m2 K. It is a value higher of 31% compared to the calculated one and lower of 21% compared to the one measured with HFM method. 4.1.3. Comments on the results In scientific literature, very few extensive studies have been made on the in situ measurement carried out to determine the asbuild thermal transmittance of building elements by means of HFM method. In [28] it is stated that “results also suggested that as-build U-values of walls are typically around 20% higher than U-values predicted” using technical standards, and in a certain number of cases the difference is up to 100% especially if a layer of insulating material was present. So, a difference of 59% as found in case study A is not surprising. Anyway, concerning the errors in the final U-values from HFM, the contribution of intrinsic calibration, correction for thermal storage effects, correction for in situ use, accuracy of temperature difference and repeatability of the system should be considered [28], for a total random error of 13% in each individual U-value measurement bringing it comparable to the value measured with ITT method (0.4234 W/m2 K vs 0.38 W/m2 K, with a deviation of 0.0434 W/m2 K between the two methods). 4.2. Case study B “Vieider House” is a single family house built in crosslaminated timber panels. It is placed in Valdaora di Sopra (46◦ 46 0 N–12◦ 2 0 E, Bolzano – Italy), 1048 m asl (Fig. 3). Table 3 U-value of the building fac¸ades using the ITT method in case study A. 2

Facade

U-value [W/m K]

South West North

0.37 0.40 0.37

In Table 4 the characteristics of the layers of external walls are summed up. The total width of the wall is d = 0.385 m, the mass is m = 96.100 kg/m2 (light wall). The calculated U-value is U = 0.148 W/m2 K. As for case study A, ITT has been applied on the four fac¸ades of the building. Main results are shown in Table 5, where the U calculated value is compared with the one surveyed with ITT considering the two extreme wind speed conditions: 0 m/s and 1 m/s (over 1 m/s, in fact, convective phenomena become prevalent on radiation ones, so the results with ITT are considered no more valuable). In this case, it is interesting to notice the influence of wind speed on the ITT results. Basically, the difference is of 29.6% with slack air (Ud vs U0 ), while it grows up to 80% if moving air is considered with a speed of v = 1 m/s (Uc vs U0 ) because of the greater thermal dispersion for convective factors. Again, environmental conditions during the test realization are very important for a proper reading of the surveyed data.

4.3. Case study C It is a multistorey building called “Residenza EOS” placed in Rovereto (45◦ 53 0 N–11◦ 02 03 E, Trento – Italy), 204 m asl (Fig. 4). External wall is made up of (from inside to outside) plaster (1.5 cm,  = 0.9 W/m K), brick block (30 cm,  = 0.179 W/m K), insulation (8 cm,  = 0.031 W/m K) and plaster (0.7 cm,  = 0.7 W/m K). The calculated U-value is U = 0.225 W/m2 K. ITT has been applied to the South fac¸ade of the building during February. Two different surveys have been made, with the results shown in Table 6. The average U-value calculated with ITT has been of 0.285 W/m2 K. The difference with the U theoretic value is of +27%.

Table 4 Materials and thermal characteristics of the layers of external walls in case study B (from outside to inside). Material

s [m]

c [kJ/kg K]

 [W/m K]

 [kg/m3 ]

Wooden planks Air Pressed natural fibres Wooden bearing panel Clay panel

0.025 0.03 0.22 0.095 0.015

2.72 1.004 1.70 2.72 1.00

0.13 0.333 0.04 0.13 0.14

500 1 130 500 500

s: thickness; c: thermal capacity; : thermal conductivity; : density.

2182

R. Albatici, A.M. Tonelli / Energy and Buildings 42 (2010) 2177–2183

Table 5 Results of the ITT survey on Vieider House (case study B).

External wall

Tout [◦ C]

Tint [◦ C]

Ti [◦ C]

ε

εtot

P [W/m2 ]

Uc [W/m2 K]

U0 [W/m2 K]

Ud [W/m2 K]

−7.5

+17.5

−6.3

0.955

0.9

5.8143

0.3875

0.1944

0.148

Tout : outside environment temperature; Tint : inner environment temperature; Ti : surface wall temperature; ε : emissivity in the range 8–12 ␮m; εtot : integral emissivity; P: thermal power dissipated through the wall; Uc : thermal transmittance (ITT methodology) with wind speed v = 1 m/s; U0 : thermal transmittance (ITT methodology) with wind speed v = 0 m/s; Ud : calculus thermal transmittance. Table 6 Average U-value of south fac¸ade with ITT in case study C. Survey no.

Tout [◦ C]

Tint [◦ C]

Ti [◦ C]

ε

εtot

v [m/s]

U [W/(m2 K)]

1 2

−2.0 −2.0

+21.0 +20.0

+0.0 +0.1

0.981 0.981

0.5 0.5

0.2 0.1

0.27 0.30

Tout : outside environment temperature; Tint : inner environment temperature; Ti : surface wall temperature; ε : emissivity in the range 8–12 ␮m; εtot : integral emissivity; v : wind speed; U: thermal transmittance (ITT methodology).

graphic image is taken; in this way the areas with anomalous thermal behavior can be rejected (local thermal bridges, areas with high moisture and so on); 2. the procedure is sufficiently fast: a medium size building (such as a two storey house) can be analyzed in situ in about 2–3 h (plus 15 h of data handling in office). The main limits are: 1. The measurement can be done only during evening so as to avoid direct solar radiation; the best period of time is 3 or 4 a.m. when the difference between inner and outer temperature is maximum. 2. Wind speed must be lower than 1 m/s (best condition is lower than 0.2 m/s) in order to avoid convective phenomena out of control. In particular, the surface-to-air temperature difference plays a secondary role if wind speed is higher than 5 m/s and it is relatively more important for 1 m/s wind speed [27]. For qualitative ITT survey, a wind condition lower than 2.7 m/s is suggested [24] in order to avoid variations created by radiant corner effect and air leakage. ITT quantitative method is based on the measurement of radiation temperature from the building element, so wind speed must be lower as possible and 0.2 m/s has been found to be the theoretical upper limit to have minimum deviation from expected results. 3. The building elements must have stored a sufficient amount of heat during previous days in order to have a dispersed thermal power significantly measurable; that is, for at least 48 h before the measurement the users should have taken inner rooms temperature at a uniform level of 20 ◦ C, while the meteorological situation must have been fair (clear sky, possibly sunny and non-rainy or windy). 4. The difference between inner and outer temperature during the measurement must be of at least 10–15 ◦ C in order to allow a measurable heat exchange through the element [12].

Fig. 4. Case Study C: Residenza EOS, south fac¸ade with a thermal image and a temperature profile.

5. Conclusions The measurement of the U-value by means of infrared thermovision technique has two main and meaningful advantages: 1. first of all, it is not a punctual measurement but it considers all the surfaces of the detected element whose global thermo-

The last condition is probably the most important. For this reason the ITT method described in this paper can be used only during winter time (even if the authors are presently carrying on some studies in order to use the method in dynamic state conditions during spring and autumn). For what concerns the difference between the U planned values and the recorded ones, is not a surprise. In fact, the thermal conductivity  value of an element to be considered during the design phase should not be the one declared by the producer, but a correct value that takes into account the real condition of the elements and the way they have been laid down (that sometimes can be not the proper one). The Italian standard UNI 10351:1994 [29] itself states

R. Albatici, A.M. Tonelli / Energy and Buildings 42 (2010) 2177–2183

that the deviation from the values measured in laboratory and the real ones found in usual production can be of 5% up to 50%. So, the standard gives a modified  value considering the average humidity level in real condition of use, the ageing, the possible tamping of loosed materials, the handling, the installation made by the rule book, the thickness tolerance. The value is increased from 10% up to 50%, sometimes even 100% depending on the material. It is obvious that the estimation of U-value by means of on site measurements, strongly depending most of all on the  value of the material composing the building element, can differ with the same percentage. Being the present research work the first considering the use of ITT method for the survey of in situ U-value of building elements, no other data exist in literature to compare the deviation found in the case study presented. Further research works are necessary on several case studies following the methodology proposed in order to establish an average correlation between the results of the measurement and the theoretical value obtained with technical standards. Acknowledgments The research presented in this paper has been partially supported from the project “Analysis of the efficiency of passive solar devices in timber buildings in the Alpine Space and design proposals” funded by the Italian Ministry of Instruction University and Research, and the project “A new system for timber buildings and its integration with passive and renewable energy sources” funded by Gruppo Polo – Le Ville Plus from Cassacco (Udine – Italy). References [1] European Directive 2002/91 of the European Parliament and of the Council of 16 December 2002 on the Energy Performance of Buildings, 2002. [2] L.D. 311/06, Corrective and Additional Disposal to the Legislative Decree 19 August 2005, no. 192, Concerning the Implementation of the Directive 2002/91/CE on the Energy Performance of Building, 2007 (in Italian). [3] M. Kleemann, R. Heckler, G. Kolb, M. Hille, Die Entwicklung des Energiebedarfs zur Wärmebereitstellung in Gebäuden – Szenarioanalysen mit dem IKARUSRaumwärmemodell (The development of energy demand for building heating – Scenario analysis using IKARUS-thermal model), Bremer Energie-Instituts, Bremen (D), 2000 (in German). [4] prEN15203, Energy performance of buildings – assessment of energy use and definition ratings. [5] S.N. Flanders, Heat flux transducers measure in-situ building thermal performance, Journal of Building Physics 18 (1) (1994) 28–52. [6] ISO 9869, Thermal Insulation – Building Elements – In-situ Measurement of Thermal Resistance and Thermal Transmittance, 1994.

2183

[7] E. Grinzato, P.G. Bison, S. Marinetti, Monitoring of ancient buildings by the thermal method, Journal of Cultural Heritage 3 (2002) 21–29. [8] J.M. Hart, A Practical Guide to Infra-Red Thermography for Building Surveys, Building Research Establishment, Watford (UK), 1991. [9] N. Walker, Infrared Thermography Handbook – Volume 1 Principles and Practice, The British Institute of Non-Destructive Testing, Northampton (UK), 2004. [10] D.A. Haralambopoulos, G.F. Paparsenos, Assessing the thermal insulation of old buildings – the need of in-situ spot measurements of thermal resistance and planar infrared thermography, Energy Conversion Management 39 (1/26) (1998) 65–79. [11] C.A. Balaras, A.A. Argiriou, Infrared thermography for building diagnostic, Energy and Buildings 34 (2002) 171–183. ˜ I. Canas ˜ Guerrero, I. González Requena, Thermographic survey of [12] S.M. Ocana, two rural buildings in Spain, Energy and Buildings 36 (2004) 515–523. [13] UNI 9252, Thermal Insulation. Survey and Qualitative Analysis of Thermal Irregularities in Buildings Envelopes. Infrared Thermovision Technique, 1988 (in Italian). [14] E. Barreira, V.P. de Freitas, Evaluation of building materials using infrared thermography, Construction and Building Materials 21 (2007) 218–224. [15] N.P. Avdelidis, A. Maropoulou, Emissivity considerations in building thermography, Energy and Buildings 35 (2003) 663–667. [16] R.P. Madding, Emissivity measurement and temperature correction accuracy considerations, in: D.H. LeMieux, J.R. Snell (Eds.), Thermosense XXI, SPIEInternational Society for Optical Engineering, Florida (USA), 1999, pp. 393–401. [17] L. Ibos, M. Marchetti, A. Boudenne, S. Datcu, Y. Candau, J. Livet, Infrared emissivity measurement device: principle and applications, Measurement Science & Technology 17 (2006) 2950–2956. [18] A. Hoyano, K. Asano, T. Kanamaru, Analysis of the sensible heat flux from the exterior surface of building using time-sequential thermography, Atmospheric Environment 33 (1999) 3941–3951. [19] M. Davies, C. Martin, M. Watson, C. Ni Riain, The development of an accurate tool to determine convective heat transfer coefficients in real buildings, Energy and Buildings 37 (2005) 141–145. [20] A. Hagishima, J. Tanimoto, Field measurements for estimating the convective heat transfer coefficient at building surfaces, Building and Environment 38 (2003) 873–881. [21] I. Beausoleil-Morrison, The adaptive simulation of convective heat transfer at internal building surfaces, Building and Environment 37 (2003) 791–806. [22] S. Fohanno, G. Polidori, Modelling of natural convective heat transfer at an internal surface, Energy and Buildings 38 (2006) 548–553. [23] M.Y.L. Chew, Assessing building facades using infra-red thermography, Structural Survey 16 (2) (1998) 81–86. [24] A. Colantonio, Infrared thermographic investigation procedures for four types of generic exterior wall assemblies, in: D.H. LeMieux, J.R. Snell (Eds.), Thermosense XXI, SPIE-International Society for Optical Engineering, Florida (USA), 1999. [25] K. Watanabe, Architectural Planning Fundamentals, 1965. [26] J.A. Palyvos, A survey of wind convection coefficient correlations for building envelope energy systems’ modeling, Applied Thermal Engineering 28 (2008) 801–808. [27] M.G. Emmel, M.O. Abadie, N. Mendes, New external convective heat transfer coefficient correlations for isolated low-rise buildings, Energy and Buildings 39 (2007) 335–342. [28] S. Doran, Safety and Health Business Plan—field investigation of the thermal performance of construction element as built, 36/8/79 cc1637 Final Report, BRE East Kilbride, Building Research Establishment Ltd., UK, 2000. [29] UNI 10351, Building Materials. Thermal Conductivities and Vapor Permeabilities, 1994 (in Italian).