This study carried out field experiments and computer simulations to analyze the piston effect by the movement of elevator cars and the influence of the piston effect on the existing smoke control performance in case of using escape or emergency elevators for the safe egress of high-rise building occupants.
Field experiments were conducted to assess an elevator hoistway flow coefficient using lower-level elevators and piston effect using elevators in multi-use high-rise buildings.
Computer simulations were conducted using 'FLUENT', the computational flow dynamics(CFD) simulation program, and 'CONTAMW', the network analysis program. FLUENT was used for assessing the elevator hoistway flow coefficient, and CONTAMW was used for modeling a building plan and then analyzing smoke control performance considering the piston effect.
Finally, the critical pressure in a real multi-use high-rise building was obtained from the newly calculated flow coefficient in the elevator hoistway. The influence of the critical pressure on the smoke control performance was also analyzed, and the results are as follows.
1. Experimental measurements and Analyses
An elevator hoistway flow coefficient was assessed through the experiments using elevators in 3-story and 2-story buildings. The experiments on the elevator's piston effect were carried out using elevators in 33-story, 44-story and 54-story high-rise buildings, respectively.
According to the results of flow coefficient experiments, average flow coefficient was calculated as 0.954. Considering the 4σ to guarantee 99.99 % reliance, it became 0.86.
This result was 3.6 % bigger than the value of 0.83 that Klote and Tamura suggested. The maximum critical pressure was decreased about 7 % with the same condition of elevator and elevator shaft. This analysis can result in the significant change in construction cost by applying a more realistic and less value of elevator piston effect when considering the smoke control performance in high-rise buildings.
From the measurement of the pressure change by the piston effect in a type of hoistway, the results are as follows.
(1) In case of 1-shaft 1-elevator, the pressure change by the piston effect was measured and the value was ranged from 21 Pa to 29 Pa. It is that the piston effect caused by elevator operations had the biggest value. Using the same conditions as the experiments, the calculated maximum critical pressure was 34 Pa. It is close to the result of experiments.
(2) The more elevators were installed at 1-shaft, the more pressure changes decreased. Especially, in case of 1-shaft 4-elevators, the pressure was decreased at the point where it is expected to increase. The range of the pressure change less than 4 Pa explained little influence by the piston effect.
2. Modeling and Analysis
The computer simulation to calculate flow coefficient was performed using FLUENT, commercial CFD program. The simulation for the evaluation of the smoke control performance with piston effect was also performed using CONTAMW developed by NIST.
As a result of the flow coefficient simulation, the coefficient was calculated as 0.86 considering the safety margin. The deviation of the flow coefficient according to the elevator velocities was less than 4 %, which showed little influence of elevator velocities on the flow coefficient. This value is more conservative than 0.88 calculated through the experiments, and the flow coefficient which apply to calculate the maximum critical pressure by the elevator piston effect, became 0.86 considering the result of experiments and the simulation.
From the result of the simulation for the evaluation of the smoke control performance with the piston effect, it was found that exterior walls' airtightness hardly had an effect on the smoke control performance such as differential pressure, smoke protection air velocity, and so on. The differential pressure between a living room and exterior walls, however, increased according to the increase of exterior walls' airtightness. Air volume passing through the exterior walls was steadily irrespective of exterior walls' airtightness.
Existing smoke control air volume that maintains the differential pressure, 50 Pa, didn't satisfy the requirements of NFSC, and smoke control air volume must be increased by 108 % in comparison with existing smoke control air volume to satisfy the requirements of NFSC.
To satisfy the existing smoke protection air velocity, 0.7 m/s of the requirements of NFSC, smoke control air volume should be increased by 47 % in comparison with existing smoke control air volume.
Upon opening the lobby door temporarily during maintaining the differential pressure of 50 Pa between the lobby and the living room and smoke protection air velocity of 0.7 m/s considering the piston effect caused by elevator operation, the piston effect disappears and results in the increase of opening forces of the lobby door to 249.6 N and 616.8 N, which are far above the NFSC requirement of 110 N.
3. Performance estimation of smoke control system with piston effect
The flow coefficient of elevator hoistway was 0.88 and 0.86 respectively according to the result of experiment and computer simulation. To be conservative the flow coefficient of elevator hoistway of 0.86 is recommended to be used in the smoke control design.
In the hoistway type having three or more elevators, the pressure change caused by piston effect is negligible due to the large area of hoistway.
According to the result of experiments, the maximum pressure change was 29 Pa in case of 1-shaft 1-elevator. The suggested flow coefficient is 0.86 as above. The maximum critical pressure caused by the elevator piston effect was calculated as 34 Pa using this flow coefficient.
The difference between the maximum critical pressure by calculation and that by experiments is 5 Pa. There is about 15 % gap. It means that smoke control design can be performed using the calculated maximum critical pressure.
The followings are results of the pressure change caused by the piston effect in elevator hoistway.
The exterior walls' airtightness hardly had an effect on the smoke control performance such as differential pressure, smoke protection air velocity, and so on. The differential pressure between a living room and exterior walls, however, increased according to the increase of exterior walls' airtightness. Air volume passing through the exterior walls was steadily irrespective of exterior walls' airtightness.
In case of decompressing in hoistway by piston effect, the existing smoke control air volume can not satisfy the requirement of NFSC and the smoke generated from fire room will contaminate an elevator lobby and hoistway.
Under these conditions the evacuee can not open the door. So, when the piston effect disappears suddenly, an over pressure venting system will be needed.
In order to get the optimized smoke control performance in a high-rise building, not only smoke control air volume required by NFSC but also additional smoke control air volume considering the piston effect are needed, and PBD(Performance Based Design) applying the features of building has to be required.