SEARCH WITHIN CONTENT
Citation Information : Architecture, Civil Engineering, Environment. Volume 11, Issue 3, Pages 121-130, DOI: https://doi.org/10.21307/ACEE-2018-044
License : (BY-NC-ND-4.0)
Received Date : 24-April-2018 / Accepted: 04-June-2018 / Published Online: 04-April-2019
The widespread thermal improvement in residential buildings involves not only the insulation of outdoor walls but also window replacement. In Poland it is the residents of individual premises who manage the replacement and in order to keep their heating bills low, they seek airtight solutions to minimize the cooling of premises due to air infiltration. In this situation the indoor air quality is not considered at all and no exchange of used air and fresh air occurs. Unawareness on the part of residents and the increased costs of replacing the windows which need additional devices to ensure the inflow of air are the main reason for the deteriorating microclimate conditions in residential dwellings.
The present paper demonstrates the measurement of indoor air quality, the building leakage test and the measurement of air flow in exhaust opening a four-bedroom dwelling located in Gliwice, Poland. In order to evaluate the air exchange within a longer period and in different outdoor climate conditions, the measurements were supplemented with numerical simulation of ventilating airflows. Modifications to improve the indoor air quality in the examined flat were also suggested.
As reported by Central Statistical Office of Poland (GUS), it is estimated that the number of dwellings in Poland amounts to 13.5 millions . Until 1985 the erected buildings had been extremely energy consuming and often without thermal insulation. The time had been characterized by the lack of strict requirements concerning the maximum heat transfer coefficient in outdoor walls: during 1960s – (1.16÷1.42 W/(m2K)) and 1980s (0.75 W/(m2K)). Pursuant to Journal of Laws Dz.U. 2017.2285  from 1st January 2017 the maximum heat transfer coefficient in outdoor walls amounts to 0.23 W/(m2K). The solid brick buildings of 1966 and the pre-fabricated concrete blocks of 1967–1985 are the most energy consuming. The main reason for thermal improvement is the need to meet the new requirements concerning the maximum heat transfer coefficient and to make the buildings energy-efficient. The number of thermal improvement works has been rising since 2002, as demonstrated by Bank Gospodarstwa Krajowego which pays out thermomodernization bonus to finance the part of a loan taken to cover the thermal improvement expenditures. In 2000 the number of the bonuses amounted to 38 (i.e. 482 000 PLN), while in 2015÷2830 bonuses were paid out (the amount of approximately 149 000 000 PLN). The major recipients of thermomodernization bonus are multifamily residential buildings, which account for 94% of the submitted applications .
Thermal improvement involves:
Extra thermal insulation for outdoor parts (both walls and a roof),
Central heating improvement.
Window replacement, being one of the major aspects in thermal refurbishment, frequently results in the deterioration of indoor environment conditions. Modern windows with a low heat transfer coefficient are airtight and lead to insufficient air exchange in buildings  as well as the rise in relative humidity and CO2 concentration.
The air quality in rooms depends on the air parameters, including air temperature and relative humidity. In winter conditions inside the buildings thermal comfort is achieved when temperatures range between 21÷23°C, and relative humidity: 40÷70%. While getting the comfort temperature inside the room is not a problem due to the operation of central heating installation, than maintaining the relative humidity can be troublesome. RH inside the heated rooms very often falls below the recommended value, because the heating results in drying of air. Unless there is insufficient air exchange, with large internal moisture gains (washing, cooking), RH can reach high values. A parameter influencing the air quality in the room is also the concentration of CO2 which is produced by humans. When the air exchange inside a building is not enough the CO2 concentration increases, which can lead to bad mood, sleepiness, headaches, etc. The concentration of carbon dioxide is easily measurable and therefore often used in measurements as an indicator of air quality .
According to Polish Standard PN-83/B-03430/Az3:2000  in a multifamily building of up to 9 stories, natural or mechanical ventilation can be used. Higher buildings should have mechanical exhaust ventilation or supply-exhaust ventilation. The standard specifies the minimum ventilation air stream in a dwelling (the sum of exhaust air streams from auxiliary rooms):
For kitchen with external window, used gas or coal cookers – 70 m3/h,
For kitchen with external window, used electric cookers, in apartments for up to 3 people – 30 m3/h, – in apartments for more than 3 people – 50 m3/h,
For kitchen without external window, used electric cookers – 50 m3/h,
For bathroom (with or without toilet) – 50 m3/h,
For separate toilet – 30 m3/h,
For auxiliary room without windows – 15 m3/h.
Thus, in case of the most common dwellings in Poland, the total exhaust air stream amounts to 120 m3/h (kitchen with external window, used gas cookers, bathroom) or 150 m3/h (dwellings with separate toilet).
Pursuant to Journal of Laws Dz.U. 2002.75.690  special supply ventilation units or mechanical ventilation should be used if the airtight window frames render the external air infiltration impossible. The above mentioned regulation , as amended, applies to construction or extension of a building. Thermal improvement, window replacement in particular, are renovation works and as such are not governed by the regulation in question. Tilt and turn windows are most often used in the improved buildings. Such windows have the microventing option, which makes the airtightness lower, yet it is hardly ever used in winter so as not to cool the premises.
The research  done in Polish multifamily buildings showed that the exhaust air streams from the dwellings amounted to 10÷105 m3/h, which is only a fraction of the required values determined by PN-83/B-03430/Az3:2000 Standard. The measurements described in  and carried out in winter in a few Polish naturally ventilated dwellings located in a 5 and 11-storey buildings proved that the exhaust air stream was much lower than the one specified by the standard. In case of a 5-storey building the stream was 32÷105 m3/h (the required value was 150 m3/h). It was even worse in the 11-storey building, where the exhaust air stream was 0÷63 m3/h for closed airtight windows and 18÷83 m3/h for widows with the microventing option.
The problem of insufficient air exchange in improved buildings across Poland and the Czech Republic was indicated, inter alia, in . France was fraught with similar problems during 1980s, when trickle vents had been used until mechanical ventilation was introduced.
The air permeability is the air parameter that influences the air exchange in a building. It is characterized by n50 coefficient that specifies the number of air exchanges within 1 hour with the outdoor and indoor pressure difference amounting to 50 Pa. The airtightness of the building envelope (or the airtightness of the dwelling), should be in accordance with PN-EN 12831 standard  (Table 1).
As proved by research , the values are frequently not reached in case of Polish dwellings, especially if they are equipped with airtight windows. Regarding the dwellings with modern windows the n50 coefficient amounted to 1.5 h-1, whereas as regards old and mixed windows the n50 coefficient was 2.7÷3.8 h-1. The results show that the dwellings equipped with old windows are classified as medium airtight while the ones with modern windows – demonstrate high airtightness (Table1).
When assessing air exchange in a building/dwelling, air exchange rate is often applied. According to Polish Standard PN-EN 12831:2006  the minimum air change rate in dwellings should be 0.5 h-1. The same standard applies in many European countries. The research carried out in European countries  shows that the air exchange rate of 0.5 h-1 is not always reached. (Denmark –in 60% of the examined dwellings n<0.5 h-1, Finland – 50%, Norway – 30÷40%). In addition, the data contained in  show that the air exchange rate in Swedish detached houses and dwellings is lower than 0.5 h-1 (0.47 h-1 for dwellings, 0.33 h-1 for detached houses). A similar low air exchange rate is observed in Catalonia, Spain due to heavy structure of the buildings . The situation looks much better in Mediterranean countries – n=1.5 h-1 in Greece [15, 16], n=1.2 h-1 in Portugal . Thus, air exchange rate depends on the outdoor climate, buildings are not so airtight in warm climate. The impact of outdoor climate on opening the windows was observed in Great Britain, where the air exchange rate in 70% of the dwellings was over 0.5 h-1 in the summer and in the winter in 68% of the dwellings, it did not exceed 0.5 h-1 [18, 19]. In a similar way, the research done in four different regions of various climates across the USA showed that the air exchange rate dropped to 0.42 h-1 in winter  whereas in warm climate regions (California, New Jersey, Texas) it amounted to n =0.71 h-1 .
The air exchange rate of n=0.5 h-1 is insufficient as shown by the research in European countries  and results in health problems of the part of the residents. Too little amount of fresh air in rooms brings about an increase in indoor pollutants as well as humidity and can result in SBS and allergy (mite build-up). In over 55% of the dwellings and 80% of the detached houses in Sweden the air exchange rate did not exceed 0.5 h-1 [22, 23]. The readings in Denmark are comparable to the ones in Sweden [24, 25]. The studies  carried out in Danish children’s bedrooms (4÷5-year olds) demonstrated that in 57% of the case, the air exchange rate was lower than 0.5 h-1. The average CO2 concentration was below 1000 ppm in 32% of the bedrooms while in 23% of the dwellings it reached 2000 ppm, and in 6% of the rooms exceeded 3000 ppm. Lower CO2 concentrations were recorded in rooms where a parent slepts together with the children, which suggests that the rooms were ventilated more often. It seems that ignorance on the part of the residents over insufficient fresh air amount plays a crucial part. The research done in Polish bedrooms  demonstrated periodical increase in CO2 concentration exceeding 3800 ppm. The research in question showed the average CO2 concentration in the examined bedrooms to range from 535 ppm to 2755 ppm within 8 hours. In two out of three cases CO2 rise did not exceed the recommended Ashrae Standards 62.1-2010  of 700 ppm above CO2 concentration in outdoor air. The same was true for spacious rooms, with the cubic capacity of approximately 25 m3 for one person.
The literature shows the issue of inadequate air exchange rate to be a common problem in many European countries. With energy saving in mind and low awareness of air exchange, residents frequently end up with low quality of indoor air.
Characteristics of monitored rooms
The measurements were performed in a 4-room dwelling located on the fourth floor of the five-storey building. The building is located in the suburb of Gliwice in the south of Poland, in the area there are other five-storey building and detached houses. The apartment was occupied by a 4-person family: two adults and two students.
The total area of the apartment (with the toilet, which was not subject of indoor air quality and airtightness measurement) is 72.5 m2. The apartment is equipped with an airtight windows, without trickle vents. Thermal improvements were carried out in 2011. Table 2 contains information about the rooms where the measurements were carried out.
Temperature, relative humidity and CO2 concentration sensors were carefully located in the space where the conditions were representative, i.e. they were placed away from sources of heat, windows and doors at a height of about 1.8 m above the floor. There was one sensor in each room listed in Table 2. The data were recorded with 5-mins intervals from 7 November until 31 April. During the diagnostics SENSOTRON PS33 indoor air quality monitor for measuring temperature was used (measuring range: 10÷45°C; the accuracy: ±0.5°C), relative humidity (measuring range: 0÷100%; the accuracy: ±3.5%) and CO2 concentration (measuring range: 0÷5000 ppm; the accuracy: ±10 ppm).
The analysis of temperature variation in different areas throughout the period of the measurement showed that there is a relatively high room temperature: 21÷25°C, in the kitchen temperature often increases up to 31°C, due to cooking. The lowest average temperature was recorded in the case of room No. 06, which is only used as a bedroom, as well as in the living room (room No. 02) (Figure 1).
Despite the high temperature in the rooms, the average relative humidity does not drop below 40% (Figure 2), rapid changes in humidity between 20 to 80% are observed periodically in the kitchen.
A high value of maximum relative humidity occurs also in the bathroom, where it reaches approx. 75%. Such conditions remain in the bathroom much longer than in the kitchen, up to 24 h, which is caused by drying laundry. Regarding the rooms, the highest maximum humidity – 60% was observed in the living room 2, where all family gather in the evenings and in room 6, where central heating is often turned off.
Relatively high values of humidity lead to the conclusion that there are large internal gains of moisture, which are not removed by ventilation. This is confirmed by the analysis of CO2 concentration, the value of this parameter increased up to 5000 ppm (Figure 3), while the recommended value should not exceed 1000 ppm. The highest values are observed in the room 1 and in the kitchen.
The monthly averages CO2 concentration time course presented in Figure 4 show a distinct drop of values in the months of March÷April, higher external temperatures are the reason for more frequent airing.
The airtightness test was performed by Minneapolis Blower Door Model 4 (measuring range of air volume: 19÷7200 m3/h with the measurement uncertainty of ±4%; the uncertainty of pressure difference measurement ±1% of reading value or 0.15 Pa). The fan was placed in a door frame and the dwelling was pressurized and depressurized (Figure 5).
The measurement was carried out according to PN-EN 13829:2002 standard  in wind-free conditions. Before the experiment all the windows had been closed and all doors opened. Every leakage had been sealed except for window’s leaks i.e. central heating and water pipes, soil stacks, electrical installation conduits. The bathroom’s and toilet’s doors had been sealed to eliminate the air flow from ventilation outlets or flue ducts. The ventilation outlet in the kitchen had been completely sealed. Figure 6 shows the method of eliminating leaks.
The Blower Door measurements were carried out for airtight windows: for underpressure and overpressure. Results of the measurement are flow characteristic and values of air flow V50 and air change rate n50 for the air pressure difference of 50 Pa. The results are shown in Table 3. According to PN-EN 12831 standard the flat is classified as very tight (n <2 h-1). Blower Door measurements also provide information on flow coefficient C concerning air infiltration through the windows leakages:
a – air infiltration coefficient, m3/(m·h·Pan),
l – total length of the crack, m.
The flow coefficient C is used to describe the airflows:
V – air flow through the single air flow path, m3/h,
C – flow coefficient, m3/(h·Pan),
Δp – pressure difference, Pa,
n – exponent ≌2/3.
On the basis of prior knowledge the value of the flow coefficient C and the total length of the crack l air infiltration coefficient a for the tested windows was determined (Table 3).
Low values of n50 are characteristic for tight windows, also the value of the air infiltration coefficient a within 0.3 is typical for tight, new windows. Similar results were obtained in the publication . For old windows, which are no so tight the value of the air infiltration coefficient a is higher, often and often exceeds 1 m3/(m.h.Pa0.67).
The measurement of exhaust air flow was carried out from 26 February to 12 March. During the measurements the outside air temperature did not exceed 10°C. The air flows were measured in each outlet i.e. in the kitchen, bathroom and toilet by a volume flow hood Swema Flow 233 balometer (measuring range of the volume flow: 2÷65 dm3/s; the accuracy: ±4% for 18÷25°C), (Figure 7).
Based on the measurements, the average air flow for each of the exhaust opening was determined. The results and the required values are shown in Table 4.
The measured values of exhaust air flow represent only a fraction required by the standard PN-83/ B-03430 / Az3: 2000 , although the measurements were carried out in the period when the outside temperature was close to +12°C, i.e. the temperature at which the standard values of air flows should be obtained. In the case of a higher outside temperature, i.e. with a decrease in the internal and external temperature difference, the exhaust air flow will decrease, which will be cause of worse indoor climate conditions in dwelling.
The air change rate calculated on the basis of the measured flow exhaust air volume and knowledge of volume of dwelling is within the range of n = 0.18÷0.3 h-1. This value is lower than the recommended minimum of 0.5 h-1. The low value of the air change rate is the cause of high concentrations of CO2 as described earlier.
Although the assessment of air exchange in an apartment by measuring method is the most reliable, it is only incidental, concerning a specific point in time. The alternative is the numerical simulation of ventilation flow, provided that the necessary data to built a reliable numerical model are available. It is possible then to assess air exchange in the long term for changing outdoor climate.
To investigate the air flows in the test building, a numerical model was built, imitating the movement of air ventilation by using a simulation program CONTAM . This program is used for numerical analysis of ventilation flows and indoor air quality in buildings. One can also determine the global assessment of the ventilation effectiveness in the whole building, determine the variation of the ventilation air flows and find the influence of the airtightness of building envelope on infiltration [31, 32]. The measurements of the air exchange in the apartment enabled the calibration and validation of the model. Consequently the numerical model which gave reliable simulation results was formed.
CONTAM program was used to balance the air flows through the windows and door and through the ventilating ducts. The air flows were modeled according to the power law equation (2). The pressure difference that causes the air flows allows for both the thermal buoyancy effect induced by the temperature difference between the air inside and outside of the building and the influence of wind. The reliability of the simulation results is greatly dependent on the correct identification of model parameters. In the process of numerical modelling of the building it was assumed that the single zone represents the whole apartment where all the internal doors are open. This approximation does not significantly affect the accuracy of calculations . 22 calculation zones were defined (15 flats, 6 staircases and basement). The internal temperature of 20°C was assumed for all the dwellings, for basement – 12°C, for staircase – 18°C. The case of opening the windows was omitted in the model since the calculation related the period of the measurement during heating season, when windows are rarely opened. Numerical calculations were performed for the weather data from a local weather station for the period from January to March.
The calculation model was tuned and validated by 2-week measurements of exhausted air flow in the flat. This was done by changing value of air infiltration coefficient in the range of 0.1÷0.5 m3/(m·h·Pan)  and testing the compliance of measurement and simulation results. It was found that by determining the air infiltration coefficient value at 0.3 m3/(m·h·Pan), the simulation results would be similar to the obtained measurement results for the whole apartment (shown in Figure 8). The same value was achieved from the blower door measurement, too, then this value was used for further calculations.
The comparison between the measurement and simulation results shows that the results vary by approx. ±10 m3/h and the average relative error is 1.6%.
The air exchange rate (ACH) for the measurement period is 0.24 h-1 (Figure 9), the value is lower than hygiene requirements and the value required by standards.
It should be noted that the assessment of the air exchanges based on the exhaust air stream can be misleading as the inflow of air is not only infiltration air, but also airflow through the flat door from the staircase. The numerical simulation provided data on all streams of air flowing through the apartment (see Figure 10 – positive values – inflow to the apartment, negative values – exhaust air flow). It could be noted that the location of the test apartment – 3rd floor of the building – makes it possible for air flow from the staircase to the apartments to occur. It arises due to the pressure distribution in the stairwell (neutral pressure plane). The air flow from the staircase is not large, it is, however, included in the measurement of the exhaust air.
The simulations for the heating season showed that throughout the heating season, the average air exchange is less than the duration of the experiment and amounts to no more than 0.3 h-1.
The research  showed that to improve air quality in dwellings additional window vents are needed. Other studies  showed that more efficient are wall air valves than natural window ventilators.
Under sill air valves were suggested to improve the effectiveness of the gravity ventilation. The manually operated ventilator, set in the external wall between the window sill and a central heating radiator, delivers fresh air from the outside. The flow characteristics of the diffuser were taken from one of the commercially available products. Calculations were performed for different opening level of the air inlets 20%, 50% and 100% (Figure 11). The required air exchange n = 1 h-1 is achieved in longer period only by using additional fully open air inlets.
The results of this study indicate that the dwelling is very tight (n50 = 1.16 h-1). This is confirmed by measurements of the indoor environment. The carbon dioxide concentration reaches very high values (up to 5000 ppm). The value of the relative humidity is quite high for winter conditions (> 30%), when most commonly RH in homes drops due to heating the air. Such indoor conditions are caused by insufficient air exchange, i.e. lack of removal of internal gains of humidity and moisture and CO2. Also, measurements of exhaust air streams indicate a lack of air flow. The measurements and simulations show that the thermal modernization of residential buildings cannot be limited to the improvement of thermal insulation but should also improve ventilation. One way might be to install window trickle vents or under sill air valves, which significantly improve the exchange of air in houses.