Measure air leakage with a non-rigid double air-chamber method

ABSTRACT

Air leakage characteristics of a full-size metal curtain wall specimen are estimated by a pressurization technique using double plastic air chambers. The inner chamber was taped on to the mullions of the specimen, while the outer chamber is taped to the frame holding the specimen. The test procedure and data processing routines overcome special limitations of the flexible chamber setup due to the expandability of the chamber membrane and very low air leakage rate of the specimen. The experimental setup and results are presented.

1. Introduction

Air leakage has a great influence on the overall performances of metal curtain walls. It affects the envelope thermal behavior, building energy consumption, surface condensation, and durability of components. Tests have been conducted to assess air leakage of a specimen made up of two aluminum curtain wall systems. This paper presents the results of the depressurization tests on one of the wall systems.

2. Experimental setup

A full-size specimen with two aluminum curtain wall systems has been installed in the specimen frame of an environmental chamber facility [1]. This facility (Figure 1) has been designed for multi-purpose building envelope testing. It consists of the cold box for providing Canadian winter conditions, the hot box for maintaining indoor environments, and the specimen frame for holding wall specimens up-to 13ˇé-62 wide by 23ˇé-62 high (4.1 m by 7.2 m) [2].

Figure 1. Environmental chamber facility

Plastic sheets (10-mil polyethylene) are used to form the pressurization chambers at the outside surface of the specimen. The first chamber is formed between the sheets and the wall surface by taping the plastic sheet directly on to the mullion caps. To avoid the sheets from collapsing on the surface of the curtain wall and sealing the potential air leakage paths, pans of metal meshes are fitted on glazing between mullions. Small blocks of rigid insulation boards are spaced on the mullion caps that are enclosed by the air chamber.

Figure 2. A single chamber test

A second outer chamber is formed over the first inner chamber by taping plastic sheets over the entire specimen frame that holds the curtain wall specimen.

Equipment for the inner chamber includes a high-head air pump, digital pressure transmitters with multi-channel scanning capability, a high-precision laminar flow element, and manually adjustable needle valves for flow rate regulation. The outer chamber pressure is maintained by a shop vacuum and adjusted manually through a relief valve. The relative pressures in each air chamber with respect to the indoor side of the specimen are measured at two heights and the average values are used for calculations. Temperatures are measured at the pressure tap levels to verify thermal uniformity.

A photo of the single chamber during a test is shown in Figure 2.

3. Test results

3.1 Single air chamber

A series of measurements has been obtained from depressurizations of the single air chamber setup (Table 1). The negative signs in pressures and flow rates for indicating depressurization are omitted for simplicity.

A regression analysis [3] is carried out using the "solver" feature of the Excel software (Microsoft, ver 97). The solution is:

(Eq. 1)

Figure 3. Measurement vs. prediction in the single chamber depressurization test

where the hat "^" denotes the estimated flow rates as compared to the measured values. The standard deviations between the measured and the estimated air flow rate is approximately 1.13 LPM. The comparison between the measurements and predications is shown in Figure 3.

3.2. Double chamber

Table 1. Negative pressures in the single chamber and corresponding air flow rates

Pressure

P (Pa) Flow Rate

Q (LPM)

10.6 38.50

19.9 61.38

30.4 83.15

40.6 102.24

50.5 119.29

50.7 119.29

74.7 156.52

78.2 161.41

99.1 189.66

149.2 250.72

200.8 306.89

250.3 355.60

299.2 401.46

The single chamber depressurization test relies on the assumption that there is no perimeter leakage from the inner chamber through paths other than those across the target specimen wall area. However, perimeter leakage has been found to pass through the air spaces inside the covered wall sections to adjacent areas. This perimeter leakage cannot be totally sealed off, neither can its amount be measured directly. The second plastic sheet covers all the possible outlets of the perimeter leakage.

In conducting the depressurization test for the double chamber setup, (negative) pressures are maintained in the second chamber at a level somewhat lower than the pressures in the inner chamber. The pressure differences between the inner and outer chambers are always negative and yet have variations between different depressurization runs. Pressures and flow rates of the inner chamber as well as the pressures in the second chamber are measured. Table 2 lists these data points.

Data from the depressurization test on the double chamber setup are combined with those of the single chamber test. The pressures of the second chamber are considered to be zero for the single chamber data shown in Table 1.

The total air flow depressureized out of the inner chamber equals to the sum of the air leakaege through both the specimen area and the perimeter of the inner chamber during pressure and airflow equilibria. The estimated total amount can be expressed in the sum of two power law terms as:

(Eq. 2)

where C and n are flow coefficient and exponent for the wall leakage, Cp and np are flow parameters for the perimeter leakage of the inner chamber, and P and P2 are pressure of the inner and outer chamber, respectively.

Table 2. Depressurization data

in the double chamber setup

Inner chamber pressure

P (Pa) Outer chamber pressure

P2 (Pa) Pres. diff. across inner chamber

P-P2 (Pa) Flow rate from inner chamber

Qt (LPM)

230.8 221.4 9.4 178.45

93.0 76.7 16.3 115.34

207.8 187.2 20.7 180.66

89.9 68.6 21.3 115.66

79.8 52.5 27.3 116.29

187.3 156.5 30.9 182.24

71.0 38.8 32.2 117.39

60.0 20.7 39.3 117.86

163.2 118.8 44.4 183.82

148.5 94.9 53.6 184.45

Note: Negative signs in pressures and flow rates are omitted.

Using the least squares estimation principle, the parameters can be estimated as:

(Eq. 3)

The measured air flow rates and predicated values by this equation are plotted on separate lines for the wall and perimeter leakage in Figure 4.

Figure 4. Measured vs. estimated flow rates for wall and perimeter leakage

4. Conclusion

The flexible double chamber method is developed on a full-size metal curtain wall specimen in a laboratory setting to measure the air leakage with pressurization techniques. The least squares regression analysis is applied in reducing the experiment data to obtain characteristics of both wall and perimeter leakage.

The principles explored in this paper can be applied for similar experimental settings in future.

5. References

1. Ge, H., P. Fazio, and, J. Rao, "Transfer of heat, moisture and air through metal curtain walls", submitted to Thermal Performance of Exterior Envelopes of Buildings VIII, ASHRAE, 2001.

2. Fazio, P., Athienitis, A., Marsh, C. and Rao, J., "Environmental chamber for investigation of building envelope performance", Journal of Arch. Engineering, 1997, 3(2), pp. 97-102.