*3.2. Flow Test Value Judgment*

Kuo et al. [17] compiled the basic volume leakage requirements of test chambers from various countries. For instance, the previous version of ISO 5925-1 [22] and ISO 5925-2 [23] required the basic volume leakage of test chambers to be less than 1 m3/h, and the previous version of CNS 15038 [21] required the basic volume leakage of test chambers to be less than 2 m3/h. DIN 18095-1 [7] and DIN 18095-2 [8] specified that at a differential pressure of 50 Pa, the basic volume leakage of the test chamber shall not exceed 5 m3/h; BS 476-31 [11] specified that at a differential pressure of 50 Pa, the basic volume leakage of the test chamber shall not exceed 7 m3/h, and the new CNS 15038 [3] specified that the basic leakage of the test chamber shall be less than 7 m3/h. In addition, UL 1784 [9] and JIS A1516 [4] do not specify the requirements of the basic volume leakage of the test chamber. The actual volume leakage of the door test body is defined as the total volume leakage minus the basic volume leakage of the test chamber. From the above, it can be seen that the basic volume leakage of the test chamber does not theoretically affect the actual volume leakage of the door. This is a simple mathematical procedure in which by subtracting the "basic volume leakage of the test chamber" from the "sum of the volume leakage of the test chamber", the volume leakage of the door test body can be found [27], calculated by Equation (2) and converted to the actual volume leakage of the test body in a standard condition.

$$Q\_a{}^{\prime} = \frac{Q\_a}{\left(T + 273.15\right)} \times \left[\mathbf{k} \times \left(p\_a + p\_m\right) - 3.795 \times 10^{-3} \times M\_w \times p\_{H\_2O}\right] \tag{2}$$

*Qt* : The sum of the volume leakage of the test body and the basic volume leakage of the test chamber (m3/h);

*Q<sup>b</sup>* : Basic volume leakage of the test chamber (m3/h);

*Qa*: Actual volume leakage at temperature (T + 273.15) and pressure (Pa + Pm) of the test body (m3/h), *<sup>Q</sup><sup>a</sup>* <sup>=</sup> *<sup>Q</sup><sup>t</sup>* <sup>−</sup> *<sup>Q</sup><sup>b</sup>* ;

*Q<sup>a</sup>* 0 : Actual volume leakage of the test body under standard conditions (m3/h);

*T*: Air temperature (◦C);

*<sup>k</sup>*: Constant (293.15/101,325) = 2.89 <sup>×</sup> <sup>10</sup>−<sup>3</sup> ;

*pa*: Atmospheric pressure (Pa);

*pm*: Increased value of pressure (Pa);

*Mw*: Relative humidity (%);

*pH*2*O*: Saturation vapor pressure (Pa).

This test method excludes the step of measuring the basic volume leakage of the test chamber due to the field operating environment, which is bound to raise questions regarding the impartiality of this method. However, the basic volume leakage of the test chamber is in the range of 0~0.5 m3/h after repeated tests in the field by taping the contact surface of the plastic sheeting to the door or wall. Theoretically, this method should allow one to more easily achieve airtightness as compared to a chamber consisting of a complex heating system, steel plates, piping, welding, screws, and adhesive strips in a laboratory. Although it is possible to apply adhesives in the field to bring the basic volume leakage of the test chamber to a level closer to 0 m3/h, the basic volume leakage of the test chamber in the field should, for the sake of conservativeness, be 0.5 m3/h for subsequent calculations. Under different environmental conditions, the measured volume leakage values in the field should be corrected from Equation (2) to Equation (3), as the correction coefficients for different environmental conditions may vary.

$$Q\_d \, ^\prime = \frac{Q\_l - 0.5}{\left(T + 273.15\right)} \times \left[k \times \left(p\_a + p\_m\right) - 3.795 \times 10^{-3} \times M\_w \times p\_{H\_2O}\right] \tag{3}$$

CNS 15038 [3], ISO 5925-1 [5], DIN 18095-1 [7], and BS 476-31 [11] all specify temperatures in the range of 25 ± 15 ◦C, i.e., from 10 to 40 ◦C for ambient-temperatures testing. In CNS 15038 [3], the volume leakage under a pressure difference of 10, 25, and 50 Pa should be measured separately, with the benchmark of 25 Pa pressure difference, which is converted to the volume leakage under standard conditions that should not be greater than 25 m3/h. In addition, the volume leakage under a pressure difference of 10 Pa and 50 Pa should be free of abnormalities. The field test in this study excluded the detection of 50 Pa differential pressure for two reasons, the first of which is based on a comprehensive fire combustion study [28]. The differential pressure in general fires ranges from 10 to 15 Pa, while the field test in this study considered 25 Pa differential pressure as the benchmark, which is already a stringent standard and sufficient to meet the pressure derived from fires. The second is that the test chamber created with plastic tape and plastic sheeting may not be able to sustain the force of 50 Pa differential pressure for prolonged periods of time, which may cause the tape to loosen from the door frame. Although this can be improved by using a greater amount of, and better, tape, the test duration, operating cost, and the cleanliness of the door test body (which affects tape adhesion) are taken into consideration. Under conditions that do not affect the test results, the operation of measuring 50 Pa of differential pressure was excluded from the test results, and the field test was conducted only for the volume leakage measurement of 10 Pa and 25 Pa of differential pressure.
