火災發生時，煙毒從建築物開口部的流竄、擴散，是造成人命傷亡的主因，因此開口部裝設之門窗等設備，對於煙流的阻隔性相當重要。
本研究參考國內外建築物開口部設備遮煙性能及其試驗法之相關文獻，建立一套遮煙試驗裝置。並利用該裝置，進行實際建築用門之遮煙性能測試，了解實際建築用門之遮煙性能，期提供修訂相關法規之參考。經研究後歸納出以下結論：
一、常、中溫下本裝置的氣密性能，可控制在100 Pa時1.2 m3/h以下之洩漏量。
二、中溫下本裝置的恆溫性能，可控制在200℃±10%以內之穩定溫度。
三、提出適用於本裝置之建築用門遮煙測定方法，即為Qd (試體洩漏量) = Qt (總洩漏量) – Qa (基礎洩漏量)。
四、常溫下針對同一門試體而言，正反兩面之洩漏量差異不大，而正壓量測法比負壓量測法之洩漏量大，且皆隨著壓力差增加而遞增。
五、一般單扇木質門在不同壓差下隨著溫度的增加，因熱膨脹效應，洩漏量逐漸降低；在不同溫度下隨著壓力差的增加，洩漏量會相對的遞增。
六、國內實際建築用門(一般單扇木質門)之洩漏量在常溫20Pa時最大達26.1 m3/h･m2，有改善之必要。

When fires happen,the spread of smoke at the openings of buildings is the primary reason for loss of life,thus windows and doors installed at the openings is very important on impeding smoke flow.
This study refers to domestic and foreign literature relating to building opening equipment smoke separation capabilities and testing methods to establish a smoke separation testing device. The device is then used to conduct smoke separation testing on actual building doors to understand the smoke separation capacities of actual building doors,in hopes of creating a reference for the modification of related regulations and statutes. Research resulted in the following conclusions：
1.Airtight capacities of this device under ambient and medium temperatures can control leakage under 100 Pa at 1.2 m3/h.
2.Under medium temperatures, the constant-temp capacities of this device can control stable temperatures within 200℃±10%.
3.Proposes smoke separation testing methods for building doors that are suitable for this device, or Qd (test door leakage) = Qt (total leakage) – Qa (base leakage).
4.Under ambient temperatures, for the same test door, there is not a great difference in leakage created by the two sides,and tested leakage is greater for positive pressure to measure than for negative pressure to measure, and both increase with the elevation of pressure differential.
5.Under different pressure with increase in temperatures, common single leaf wooden doors have lower leakage due to the thermal expansion effect;under different temperatures,with the increase of pressure differential,leakage increases correspondingly.
6.The doors actually used in domestic architecture (common single leaf wooden doors) produce a leakage at ambient temperature 20Pa of a maximum 26.1 m3/h･m2,which should be improved upon.

1.ISO 5925/1：Fire Test-Evaluation of Performance of Smoke Control Door assemblies-Part 1：Ambient Temperature Test,Geneva,Switzerland,1981.
2.ISO/TR 5925/2：Fire Test-Smoke Control Door and Shutter Assemblies-Part 2：Commentary on Test Method and Test Data Application,Geneva,Switzerland,1981.
3.陳宏原，「鋼製防火門高溫行為研究」，國立台灣科技大學營建工程碩士論文，民國85年6月。
4.D.M.Brani & W.Z.Black,"Two-Zone Model for a Single-Room Fire",Fire Safety, pp.189-216,1992.
5.CNS 11227：建築用防火門耐火試驗法，第8節室溫遮煙試驗，民國90年。
6.劉祥至，「防煙區劃之火災模式分析與實驗驗證之研究」，國立雲林科技大學機械碩士論文，民國90年6月。
7.J.Rakic,"The Performance of Unit Entry Doors When Exposedto Simulated Sprinkler Controlled Fires",pp.1-13,2000.
8.George Hadjisophocleous,"A Two-Zone Fire Growth and Smoke Movement Model for Multi-Compartment Buildings",Fire Safety,pp.257-285,2000.
9.鄭孟昌，「建築物開口部構件常溫遮煙性實驗方法之研究」，國立台灣科技大學建築碩士論文，民國95年6月。
10.Liu Fang,"Opening Flow Coefficient and Its Effect on Fire Smoke Flow",Journal of Chongqing Jianzhu University,pp.86-92 ,2000.
11.Daniel Gross,"Estimating Air Leakage Through Doors for Smoke Control",Fire Safety,pp.171-177,1991.
12.「防火設備の性能評價業務方法書」，平成17年10月。
13.「日本建設省告示第2564號の別記漏煙試驗方法」，平成17年10月。
14.JIS A 1516：建具の氣密性試驗方法，1998。
15.DIN 18095 Part 1：Smoke control doors;Concepts and requirements,German October 1988.
16.DIN 18095 Part 2：Smoke control doors;type testing for durability and leakage,German,March 1991.
17.Y.Cooper,"Measuring the Leakage of Door Assemblies During Standard Fire Exposures",pp.1-12,1981.
18.林大惠，「防火門遮煙試驗基準研究」，內政部建築研究所報告，民國92年12月。
19.CNS 14514：建築物防火區劃貫穿部耐火試驗法。
20.UL 1784：Air leakage Tests of Door Assemblies,1990.
21.BS 476 Part 31：Methods for measuring smoke penetration through doorsets and shutter assemblies,1983.
22.CNS 15038：建築用門遮煙性試驗法，民國95年11月。
23.田中哮義，「建築火災安全工學入門」，平成14年改訂版。