Heat flux through the timber walls under summer climate conditions in Eastern Europe.

Kairys, Liutauras ; Stankevicius, Vytautas ; Karbauskaite, Jurate 等

Abstract. The systems of air cooling and conditioning become more and more popular in Lithuania because of the reduced costs of equipment and installation. The popularity of the lightweight, heterogeneous and low thermal inertia envelops of buildings is increasing as well. Thus, there is a great demand for accurate methodology for calculation of non-steady state heat transfer that will take into account the solar radiation influences, convection and radiation inside partitions. Generally received building thermal behaviour calculation methods were designed for massive partitions, that is why they have to be re-verified according to the current situation. Thermal receptivity calculation methodology and a common engineering method are analysed in this paper according to experimental results. The research object is a well thermally isolated timber frame wall.

Keywords: thermo-physical parameters, solar radiation, lightweight timber frame partition, ventilated air gap.

1. Introduction

The temperature distribution inside multilayer light-weight partitions is effected not only by the temperature difference on the both surfaces of the partition, but by the directed solar radiation onto the external surface as well. The affect value depends on the thermal inertia of such a partition. It is necessary to compensate the additional cooling load caused by the solar radiation when designing air conditioning and cooling systems [1] for building envelope with a low thermal inertia.

There are fully or partly ventilated air gaps [2] installed in the well isolated (>4,0 [m.sup.2]K/W) walls. The heat leakages through the connections of different materials occur during the service of the walls and depend on the construction quality. So there is an additional heat flux through the multilayer lightweight wall by convection and radiation [3] that have to be assumed into the whole heat exchange balance of the building envelope [4, 5].

Thus, there is a great demand for accurate methodology for calculation of non-steady state heat transfer that will take into account the solar radiation influences, convection and radiation inside partitions. The previous methodologies [6, 7] were designed for calculating thermo-physical processes inside massive partitions, that is why they have to be re-verified according to current situation. The lightweight walls contain the materials with very different physical parameters (thermal conductivity, density, thickness, specific heat capacity etc) [8]. So it is important to estimate precisely the designed values of different materials physical parameters and the effect of their interdependence to the heat storage and heat flux variation inside the wall [9, 10]. The thermo-physical parameters (heat flux density, thermal inertia, time constant, thermal resistance, time lag, decrement factor etc) of the timber frame walls (Fig 1) are analysed in this paper. The calculated parameters are compared with the experimental data. The experiment was hold in the solar energy testing facility "Plataforma Solar de Almeria" in Spain [11] in Oct 2003-Jan 2004. The mentioned parameters were calculated by the methodology [12, 13] that estimates all the conditions discussed above. The methodology was designed in the seventies, but it is still used for calculating the heat transfer through the building envelope. Two types of the timber frame walls (Fig 1) with the common structure in the Eastern Europe region were chosen as the investigation object.

[FIGURE 1 OMITTED]

The aim of this paper is to compare the heat transfer through the discussed walls calculated by different methodologies with the experimental results and to determine the reasons of these disagreements.

Climatic and indoor air parameters were measured. The measurements were performed from 01 Nov 2003 to 25 Jan 2004 with registration every 10 min. The hour's average values were calculated for each measured parameter and database of one-day mean hourly values with the precision of 95 % was created. These values are presented in Table 1.

2. Research results analysis

Heat flux density (Thermopile TNO PU32), temperature (Thermopar TC 1311T), wind speed (hot wire anemometer TSI 8455) and solar radiation density (pyranometer Kipp & Zonen CM 11) meters were installed in the walls for the experiment. The cross-sections of walls and positions of measure devices are shown in Fig 1.

Temperature distribution dynamics across wall section have been analysed first of all. A comparison of calculated and measured temperature wave oscillation amplitudes between walls layers clarified that there is no regular disagreement in the results. The higher temperature oscillation values on inner and outer surfaces and between gypsum plasterboard and thermal isolation were observed in experimental results. Higher temperature oscillation amplitude values on internal surfaces were obtained by calculations. The calculation results differences are shown in Table 2.

The calculation of temperature oscillation amplitudes is based on thermal receptivity [6] of wall materials and layers' surfaces. Indoor air temperature is constant, but there is no value set. Indoor air temperature during the natural research was kept constant and equal to 23 [degrees]C. The average thermal receptivity of timber frame and thermal isolation layer is calculated according to methodology, but temperature next to the timber stud is 0,5 [degrees]C higher than in the middle of mineral wool by experimental results. We can notify that calculated temperature oscillation values in internal layers of the walls are higher than the measured ones having situation opposite when analysing external layers. It can be concluded that thermal inertia of materials is not evaluated enough by the methodology, even if calculations are based on thermal capacity of materials.

Temperature oscillation amplitudes differences comparing by calculations and measurements are much larger in the wall with a ventilated air gap than in the one with out air gap. The difference is so big (Table 2) that there are doubts about the methodology suitability for calculating of the temperature oscillation amplitudes in the walls with partly or fully ventilated air gaps. So, it is necessary to adjust the coefficients in equations that estimate air gaps influence on the temperature distribution dynamics.

Temperature wave attenuation is another basic parameter describing the wall thermal inertia. It is calculated in proportion to adjacent layers and surfaces thermal receptivity according to the methodology. Temperature oscillation amplitudes between layers were compared to the outer surface temperature oscillation amplitudes in order to get thermal wave attenuation measured during the experiment. The results and their percentage differences are in Table 3.

The results testify that calculated temperature wave attenuation in the wall without air gap does no differ a lot from experimental results (6,1-22,9 %) when in the wall with an air gap the difference is inadmissible (25,0-67,4 %). We can notify also that after a thermal wave penetrates the thermal isolation in both types of the walls, the differences between the calculated and measured values decrease. The influence of thin layers with medium thermal receptivity to the temperature wave attenuation cannot be measured, calculated and estimated easily and properly.

Temperature wave attenuation values calculated by the methodology are lower than those obtained experimentally in every layer except for inner and 4-5 surfaces of the walls. One of the reasons can be wrongly accepted heat transfer coefficient on the internal surface of the wall and inaccurately calculated heat flow from the inner surface to the indoor air.

The measured heat flow densities through the lightweight timber walls were compared with the ones, calculated by thermal receptivity calculation methodology that takes into account the solar radiation and thermal inertia impacts.

The positions of heat flux density meters inside the test-walls are shown in Fig 1. Heat flux densities were measured on inner and outer surfaces of the wall with out air gap, inner and internal air gap surfaces of the wall with air gap. There are measured heat flow densities, solar radiation density and outdoor air temperature curves, calculated maximal heat flow densities from inner surfaces of the walls into the indoor air shown in Fig 2.

[FIGURE 2 OMITTED]

There was almost no heat inflow through the wall with air gap, because the inside air temperature of the test-room was constant and equal to +23 [degrees]C, and the out door air temperature oscillated from +6,9 to +13,8 [degrees]C. There was obtained the relative error 0,17 for calculated heat flow density absolute values comparing with the measured ones for the wall with an air gap. The relative error for the wall without an air gap was just 0,057.

It is evident that maximal absolute heat flow density values on wall inner surface cannot be calculated precisely with the discussed methodology for the wall with an air gap. Meanwhile, the results can be obtained with acceptable precision for the wall without an air gap.

Heat flux density q = 2,5 W/[m.sup.2] through the wall without air gap and q = 2,19 W/[m.sup.2] through the wall with air gap were calculated by common engineering method [14, 15]:

q = ([[THETA].sub.i] - ([[THETA].sub.e] + p x [I.sub.r,max] / [h.sub.se])) / R, (1)

where [[THETA].sub.i]--indoors air temperature [[degrees]C], [[THETA].sub.e]--outdoors air temperature [[degrees]C], [I.sub.r,max]--maximal solar radiation density onto the wall [W/[m.sup.2]], [h.sub.se]--heat transfer coefficient on external surface of the wall [W/([m.sup.2]K)], p--solar radiation absorbance of material [-].

The indoor (+ 23,0 [degrees]C) and outdoor air (+ 9,8 [degrees]C) temperatures from experimental measurements have been chosen. There was obtained the relative error 0,21 for calculated heat flow density absolute values comparing with the measured ones for the wall with an air gap. The relative error for the wall without air gap was 0,28. These differences are much bigger than the ones, evaluated comparing experimental data and calculations by the analysed methodology [6, 7]. The accurate heat flow density through the discussed timber frame walls cannot be calculated precisely with a common engineering method.

The heat flow density oscillations were smaller in the wall with an air gap by measurements and calculation results as it was expected. It is clear from Fig 2 that the heat flow densities on outer surfaces of the walls have the shape of summed outside air temperature and solar radiation density curves and have 3-4 hours time-lag.

An 100 [m.sup.2] area building with 2,7 m height walls was analysed as an example. It has 108 [m.sup.2] area of the walls. The heat flows through 108 [m.sup.2] walls were calculated by three methods mentioned above. The results are presented in Table 4.

We could see in Table 4 that if we calculate the loads for cooling or conditioning systems by methodology, it will give just 5,7 % smaller heat inflow than in reality for the wall without air gap and 16,6 % smaller heat inflow through the wall with air gap. Calculating by equation (1), the system power will be by 67,0 % too low for the building with the walls without an air gap and 64,0 % too low for the building with the walls with a ventilated air gap. So, the simplified equation, which the designers are using (1), cannot assure precise heat flow values.

3. Conclusions

1. More than 50 % of differences have been found while calculating the temperature wave oscillation amplitudes in the wall with the air gap by the methodology from experimental results. The coefficients in the methodology estimating the influence of the ventilated air gap on the results have to be revised.

2. While analysing the temperature wave attenuation, it was observed that thin layers with a moderate thermal receptivity are reducing the precision of the results when the thermal behaviour of timber frame walls is estimated.

3. After the analysis of heat flow it was clarified that it is possible to calculate the accurate absolute heat flow density values for the wall without an air gap only (the relative error with experimental results is 5,7 %). The analysed methodology has to be revised in order to get more precise heat inflow values through the timber frame walls with a ventilated air gap.

4. The accepted building thermal behaviour calculation methods cannot assure high-accuracy results for timber frame walls under the Eastern Europe summer time weather conditions.

References

[1.] Magzoub, E. E.; Osman, E. A. Computer simulation of solar cooled buildings in Khartoum. Renewable energy, 14 (1-4): May-Aug 1998, p. 373-379.

[2.] STR 2.01.03:2003. The declared and designed values of the building materials' thermal parameters (Statybiniu medziagu it gaminiu siluminiu techniniu dydziu deklaruojamosios ir projektines vertes). Vilnius, 2003 (in Lithuanian).

[3.] Endriukaityte, A.; Parasonis, J.; Samajauskas, R.; Bliudzius, R. Estimation of effect of thermal convection on heat transfer through the building partitions. Journal of Civil Engineering and Management Vol 9, Suppl 1, 2003, p. 66-76.

[4.] Cabanillas, R. E.; Estrada, C. A.; Alvarez, G. Combined natural convection and radiation heat transfer in an open tilted cavity. Advanced computational methods in heat transfer, VII 4, 2002, p. 95-106.

[5.] Zuo, Z.; Hua, B.; Ye, G. D. The effect of PV walls on indoor air-conditioning load. In: Proceedings of the 3rd international symposium on heat transfer enhancement and energy conservation, Vols 1 and 2, 2004, p. 1164-1169.

[6.] Bogoslovski, V. Building thermal physics (CTPO[TEXT NOT REPRODUCIBLE IN ASCII]). Moscow, 1970 (in Russian).

[7.] Fokin, K. Building thermal technology (CTPO[TEXT NOT REPRODUCIBLE IN ASCII]). Moscow, 1973 (in Russian).

[8.] Bliudzius, R.; Stankevicius, V. The manual of thermal physics (Statybines silumines fizikos zinynas). Kaunas: Technologija, 2001 (in Lithuanian).

[9.] Koczyk, H. Selection of thermophysical properties of heat storage wall. Journal of Civil Engineering and Management, Vol VIII, No 4, 2002, p. 281-285.

[10.] Heim, D.; Klemm, P. Numerical solution of TIM-PCM solar thermal storage system with ESP-r. Research in building physics, 2003, p. 683-690.

[11.] www.psa.es

[12.] STR 2.09.04:2002. The power of building's heating system. Energy costs for heating. Lithuanian Department of the Environment (Pastato sildymo sistemos galia. Energijos sanaudos sildymui). Vilnius, 2002 (in Lithuanian).

[13.] TCH23-314-2000-Ka[??]. O. The standard of the energy accumulated thermal protection of dwellings and public buildings (Hop[TEXT NOT REPRODUCIBLE IN ASCII]. Kaliningrad, 2000 (in Russian).

[14.] STR 2.05.01:1999. Thermal technology of building partitions. Environment Department of Lithuanian Republic (Pastatu atitvaru silumine technika). Vilnius, 1999 (in Lithuanian).

[15.] Parasonis, J.; Stankevicius, V. Thermal physics of buildings. Workbook (Pastatu silumine fizika. Pratybos ir laboratoriniai darbai). Vilnius: Technika, 2000. 104 p. (in Lithuanian).

SILUMOS MAINAI PER MEDZIO KARKASO SIENA RYTU EUROPOS VASAROS KLIMATO SALYGOMIS

L. Kairys, V. Stankevicius, J. Karbauskaite

Santrauka

Oro vesinimo ir kondicionavimo sistemos sparciai populiareja Lietuvoje del vis mazejanciu irenginiu kainu ir eksploatacijos islaidu. Naujai statomu pastatu konstrukcijoms vis dazniau naudojamos lengvos, daugiasluoksnes, taciau mazai inertiskos atitvaros. Taigi atsiranda poreikis tiksliai apskaiciuoti nestacionariuosius siluminius procesus, vykstancius minetose atitvarose, ivertinant saules spinduliuotes itaka, silumos mainu dydi konvekcija ir spinduliuote. Dabartines metodikos buvo kuriamos skaiciuoti vidutiniu ir dideliu atitvaru siluminei busenai, todel jos turi buti patikrintos atsizvelgiant i dabartinius pokycius.

Siame straipsnyje analizuojama siluminio imlumo skaiciavimo metodika ir paprastasis inzinerinis silumos srauto tekejimo per atitvaras skaiciavimo metodas palyginamas su eksperimento rezultatais. Tyrimo objektas--gerai apsiltinta (R>4,5 [m.sup.2]K/W) medine karkasine siena.

Raktazodziai: silumines fizikos parametrai, saules spinduliuote, medine karkasine atitvara, vedinamasis oro tarpas.

Liutauras Kairys (1), Vytautas Stankevicius (2), Jurate Karbauskaite (3)

(1) Dept of Civil Engineering and Architecture, Kaunas University of Technology, Studenut g. 48, LT-51367 Kaunas, Lithuania. E-mail: liutauras.kairys@one.lt

(2) Institute of Architecture and Construction, Laboratory of Thermal Building Physics, Tunelio g. 60, LT-44405 Kaunas, Lithuania. E-mail: 2silfiz@asi.lt, 3jukarb@asi.lt

Liutauras KAIRYS. PhD student at the Dept of Civil Engineering and Management, Kaunas University of Technology, Kaunas, Lithuania. Author of several papers on thermal inertia problems in lightweight building envelope.

Vytautas STANKEVICIUS. Prof Dr Habil. Head of the Building thermal laboratory at the Institute of Architecture and Construction of Kaunas University of Technology. Professor of the Dept of Civil Engineering and Architecture of Kaunas University of Technology. Member of Heating, Cooling and Air Conditioning Association.

Author of more than 200 articles and 4 inventions. Scientific interests: energy savings in buildings, the durability of outer surfaces of building envelope.

Jurate KARBAUSKAITE. Doctor (techn sciences). Researcher at the Institute of Architecture and Construction of Kaunas University of Technology. The expert of Lithuanian Studies and Research Foundation. Research interests: energy savings in buildings, heat and mass transfer, heat losses through the building envelope.

Received 04 Nov 2005; accepted 05 Dec 2005

Table 1. Results of experiment at "Plataforma Solar de Almeria"
in Spain

Parameter Average value
 of the day

 Temperature,
 [degrees]C
Wall without air gap, Measurement position--next to studs

 A1 12,1
 A2 12,2
 A3 12,6
 A4 15,1
 A5 17,9
 A6 22,6
 A7 22,8
Fall with air gap, Measurement position--next to studs

 DI 12,1
 D2 11,9
 D3 11,8
 D4 12,2
 D5 12,5
 D6 15,6
 D7 19,1
 D8 22,7
 D9 22,9
Wall without air gap, Measurement position--between studs

 B1 11,8
 B2 11,5
 B3 12,1
 B4 17,3
 B5 19,1
 B6 22,6
 B7 22,8
Wall with air gap. Measurement position - between studs

 C1 11,7
 C2 11,6
 C3 11,9
 C4 12,5
 C5 12,8
 C6 15,3
 C7 18,1
 C8 22,3
 C9 22,9

 Heat flow, W/m.sup.2]
 Wall without air gap

W1 Heat flux meter on inside surface -2,3
WE Heat flux meter on outside surface 2,3
 Other parameters

A T11 Inside air temperature 30 cm above the 22,8
floor, [degrees]C

A T12 Inside air temperature 130 cm above the 22,9
floor, [degrees]C

A T13 Inside air temperature 250 cm above the 23,0
floor, [degrees]C

WS_E air velocity in ventilated air gap, m/s 0,069

PIRE Solar radiation into the surface of the 160,8
wall, W/[m.sup.2]

EI Heal flux meter on inside surface -2,1

EE Heat flux meter on outside surface 2,5

AVE1 Wind speed in 10 m height, m/s 2,0

ADE1 Wind direction in 10 m height, [degrees] 193,8

A TE01 Outside air temperature in 10 m 9,5
height, [degrees]C

RHE1 Relative humidity in 2 m height, %. 70,9

GHE1 Solar radiation into horizontal surface, 94,7
W/[m.sup.2]

DHE1 Diffused solar radiation into horizontal
surface, W/[m.sup.2] 46,4

Table 2. Temperature oscillation amplitudes between wall layers

 Temperature oscillation amplitude, [degrees]C

 Wall without air gap
Position Experimental Calculation Calculation results
 results results differences from
 experimental data, %

Outer 22,5 18,4 18,2
surface
 1-2 20,3 20,7 1,9
 2-7 -- -- --
 7-3 -- -- --
3 6(2-6) 21,4 22,3 4,2
 6-4 6,8 5,8 14,7
 4-5 0,9 0,8 11,1
 Inner 0,8 0,6 25,0
surface
 Wall with air gap

Position Experimental Calculation Calculation results
 results results differences from
 experimental data, %
Outer
surface 20,9 15,6 25,3
 1-2 21,1 17,6 16,6
 2-7 17,1 19,1 11,7
 7-3 13,2 14,2 7,5
3 6(2-6) 12,2 15,7 26,7
 6-4 3,3 4,1 24,2
 4-5 1,1 0,5 54,5
 Inner 0,9 0,4 55,5
Surface

Table 3. Temperature wave attenuation during penetration through
the walls
 Temperature wave attenuation, times

 Wall without air gap
Position Experimental Calculation Calculation results
 results results differences from
 experimental data, %

Outer
surface -- 1,1 --
 1-2 1,0 0,9 10,0
 2-7 -- -- --
 7-3 -- -- --
3 6(2-6) 1,0 0,9 10,0
 6-4 3,3 3,5 6,1
 4-5 23,6 25,9 9,7
 Inner 27,4 33,7 22,9
surface
 Wall with air gap

Position Experimental Calculation Calculation results
 results results differences from
 experimental data, %
Outer
surface -- 1,11 --
 1-2 1,0 1,0 0
 2-7 1,2 0,9 25,0
 7-3 1,6 1,2 25,0
3 6(2-6) 1,7 1,1 35,3
 6-4 6,3 4,2 33,3
 4-5 19,9 31,2 56,8
 Inner 24,2 40,5 67,4
Surface

Table 4. Comparison example. The heat flow through 108 [m.sup.2] walls

 Heat flow, W (relative error with experimental results, %)

 Wall without air gap
Experimental Calculation by Calculation by
 results methodology equation (1)

 376,9 355,3 123,1
 (0%) (5,7%) (67,0%)

 Wall with air gap

Experimental Calculation by Calculation by
 results methodology equation (1)

 299,2 249,5 108,0
 (0%) (16,6%) (64,0%)