Elsevier

Energy and Buildings

Volume 104, 1 October 2015, Pages 108-121
Energy and Buildings

U-value in situ measurement for energy diagnosis of existing buildings

https://doi.org/10.1016/j.enbuild.2015.06.071Get rights and content

Highlights

  • U-value estimation is crucial for energy diagnosis of existing buildings.

  • Heat flux meter method allows in situ measurements of actual U-value.

  • In situ measurements of U-value can be affected by some relevant metrological issues.

  • Uncertainties of in situ measurements of U-value depend on operative conditions.

Abstract

Energy audits and check of energy performances are more and more urgent for achieving energy saving in existing buildings. In this context, it is very important to estimate U-value of buildings and, to this aim, different approaches are available. U-value is mainly estimated from the knowledge of thermal properties of the single layers of the wall using design data or analogies with similar buildings. To get a more accurate estimation, field analyses should be made like endoscopic investigation or logs. Nowadays, also in situ measurement of U-value using heat flow meters are available. Such measurements involve quite simple operational and calculation issues but, on the other hand, the disadvantage of long duration times and not negligible measurement uncertainties. In this paper, the authors present the results of an experimental campaign aimed both to assess the metrological performance of HFMs and to evaluate the influence of the ambient conditions. Furthermore, in situ U-values have been compared with the estimated ones from design data and field analyses. The results of the test show a good behavior of HFMs when tests are conducted according to ISO 9869. Nevertheless, operative measurement conditions and characteristics of the envelope component under investigation can strongly affect in situ U-value accuracy.

Introduction

Energy audits and diagnosis of existing buildings are becoming urgent for achieving energy saving. Furthermore, recent European Directives [1], [2] impose more and more strict constraints to energy losses of buildings. In this context, the evaluation of thermal transmittance and of the air permeability of building components (thermal bridges included) is a crucial step for the energy diagnosis, in order to achieve effective interventions to improve building energy performances. Anyway it is fundamental to adopt reliable experimental methods, such as heat flow meter method [3], [4], thermography [5], [6] and BDT [7], [8].

Four approaches are available to estimate U-value of existing buildings:

  • 1.

    Estimation based on data obtained by historical analysis of building or analogies with similar and coeval buildings using specific technical databases [9];

  • 2.

    Estimation based on the nominal design data;

  • 3.

    Estimation based on the actual data obtained by structure identification (sampling or endoscope method);

  • 4.

    In situ measurements using HFMs.

The first method is often poorly accurate because of the lack of reliable data about thermal properties of the materials (e.g. thermal conductivity and density) and of the layers constituting the wall. Furthermore, when an existing building is investigated, its actual conditions can differ significantly from the nominal design ones [10].

When geometrical data and thermo-physical characteristics of materials constituting the wall are accurately known from the building design it is possible to adopt second method [11], [12], [13], [14], [15].

Third methods implies specific samples to be taken in situ and subsequent laboratory measurements to obtain accurate values of thickness and thermal conductivity of each layer constituting the wall. This method normally provides a good accuracy. On the other hand, it is often avoided since it reckon on invasive sampling of the building under investigation and it is obviously not applicable for historical buildings or when it is not possible for any reason to take samples of walls.

With the fourth method the actual U-value of envelope components of buildings is directly measured in situ [4], [16]. This method is particularly useful when the structure of the investigated component is unknown and it is impossible to take samples or to perform specific endoscopic analysis.

U-value in situ measurement, very simple in theory, shows a lot of metrological and practical issues, that can lead to significant errors and uncertainties, as evidenced by experimental studies [10], [17], [18]. In particular, one of the critical issues is certainly represented by the variation of climatic parameters and of heat flow and temperature gradients through the investigated component during measurements [19]. In fact, variations of sun radiation, wind velocity, indoor and outdoor temperatures together with thermal inertia of the envelope component under investigation cause the need to get long enough sampling durations. As a consequence, measurement conditions cannot be considered steady.

Therefore, in the case of high instability of measurement conditions and whenever the average method cannot be used, other dynamic data processing methods are available [4], [20], [21], [22], [23]. Such methods, even being more complex, take into account also the thermal mass of components and the heat stored. Thus, they result more accurate in non-stationary conditions and, sometimes, they need shorter sampling times.

In this paper, the authors did not apply dynamic methods since normally they are not adopted because of their higher complexity compared with the average method.

Anyway, it is important to take into account that U-value in situ measurement under real conditions of temperature and moisture can lead to significantly different values if compared with those obtained by means of the estimation methods. This happens not only for the reliability of the measurement, but also because of the different behavior of the measurand at laboratory reference condition (to which the structure identification method refers) [24], [25].

In this paper the results of an experimental campaign aimed to evaluate in situ U-values of seven different building components under different measuring conditions using four commercial HFMs are presented and discussed. Furthermore, main error causes have been analyzed and uncertainty has been estimated for the most common operative conditions. For each building investigated the authors estimated also U-values both from design data (method no. 2) and from core samplings and endoscope analyses (method no. 3) in order to check the compatibility between such estimations and the corresponding in situ U-value measurements (method no. 4).

Section snippets

Theory

Assuming a mono-dimensional heat flow and steady state conditions, the heat flow φ [W m−2] through the envelope component under test is given by the equation:φ=C(θwiθwe)where C [W m−2 K−1] is the thermal conductance, which is only function of the thermo-physical properties of materials, and θwi and θwe [K] are the internal and external wall surface temperatures, respectively.

In the practice, since often indoor and outdoor air temperatures θi and θe [K] are available, heat flow is determined by

Methods

Four HFM dataloggers based on different technologies and measurement principles, shape and dimension of plates and number and type of temperature sensors (TC and RTD) have been assessed, as described in Table 1.

Before the experimental campaign, all temperature sensors have been calibrated at LAMI, the Industrial Measurements Laboratory of the University of Cassino and Lazio Meridionale, accredited by the Italian Accreditation Body.

Seven envelope components belonging to six existing buildings,

U-value uncertainty estimation

Despite the simplicity of the above described estimation methods, the estimation of U-value uncertainty could result very difficult due to the wide variability of materials used in buildings.

Therefore, in order to check the compatibility between U-value in situ measurements and the above described estimation methods, the authors estimated U-values uncertainty using the uncertainty propagation law [27], [28] and also referring to the standard [4].

U-value uncertainty can be very large when the

Results and discussion

Fig. 3 shows the measured internal and external surface temperatures and the heat flow trends together with the ones calculated using the average method. For the sake of representation uniformity, results of the first 72 h for heavy components and the first 12 h for the light ones have been depicted. In Fig. 3, data from HFM1 (for B1, B3, B4 and B5), HFM2 (for B6) and HFM4 (for B2 and B7) are reported.

In Table 5 the results of the in situ U-value measurements are reported for the seven

Conclusions

In this paper the U-value in situ measurements by means of some commercial heat flow meters under different measuring conditions and envelope components have been focused. Moreover, such values have been compared with those obtained by using different structure identification methods through the knowledge of thickness and thermal conductivity values of the materials constituting each layer. To this aim the authors used both design data considering the whole variability of the thermal

References (29)

  • European Parliament. Directive 2002/91/EC of the European Parliament and of the Council of 16 December 2002 on the...
  • European Parliament. Directive 2010/31/EU of the European Parliament and of the Council of 19 May 2010 on the energy...
  • ISO

    Thermal Insulation. Building Elements. In situ Measurement of Thermal Resistance and Thermal Transmittance. ISO Standard 9869

    (1994)
  • ISO

    Thermal Insulation. Qualitative Detection of Thermal Irregularities in Building Envelopes. Infrared method. ISO Standard 6781

    (1983)
  • Cited by (168)

    View all citing articles on Scopus
    View full text