Past disasters have shaped current codes and standards (Sidebar). Operating parameters for these systems have been formulated by studying fires and the physical conditions encountered in spaces engulfed in thermal smoke. Because no value can be assigned to a human life, systems must be designed, constructed, and maintained such that past failures are not repeated and additional human life and property is not lost.
Some of the most common smoke control system requirements and recommendations for high-rise buildings by local and model building codes, the NFPA, and ASHRAE include:
The final design of a smoke control system may be quite complex or relatively simple, depending on the owner's needs, engineer's experience, and requirements of the local authority issuing the approval for occupancy. However, regardless of the design, the system's effectiveness in many cases relies on the tightness of the building and stairwells, which cannot be easily predicted. To compensate for this, operational cushions, including varying volume dampers and variable-speed drives can be employed. Unfortunately, these controls often require closer coordination of the mechanical, automatic temperature controls (ATC), and balancing contractors. The added components also increase the chance for component failure.
Such a system must be tested and adjusted as a team effort and should be coordinated by the project's commissioning agent (or "authority")(CA), with all roles defined in the commissioning specification and agreed upon prior to the testing. The successful operation of an engineered smoke control system utilizes separately contracted parties, each having a stake in the operation of the system. These parties should be coordinated so that they have knowledge of how their portion of work might affect system functionality.
In today's construction, specialized contractors who have a stake in the operation of engineered smoke control systems include the drywall contractor, door installers, glazing installers, sheet metal contractors, electrical contractors, ATC contractors, fire alarm manufacturers/distributors, mechanical contractors, balancing contractors, and others. Pertinent sections of the architect's specification that affect the system's performance include the General Requirements under Division 1, Doors and Windows under Division 8, Finishes in Division 9, Specialties in Division 10, and Mechanical, Electrical, ATC, and Balancing in Divisions 15 and up.
Insufficiently detailed project specifications can reduce the engineering of these systems to a submittal review process; a process that might not be adequately detailed to carry out the design intent of the system. The designers (including the architect) should not only fine tune specifications with respect to engineered smoke control systems, but should scrutinize submittals (regardless of the thoroughness of the specification) and reject any alternatives that might reduce the effectiveness of the system.
The CA should facilitate smoke control system testing efforts by providing a test schedule and coordinating at least one kickoff meeting with the owner, the construction manager, the design engineer, mechanical contractor, ATC contractor, air balancer, building maintenance staff, and the fire alarm service contractor (if externally hired). A property insurance representative or the local authority may also want to attend. In addition, the CA should provide test data forms in advance for review and test equipment and instrumentation, including:
For zoned smoke control systems, testing should be carried out with procedures customized for the building being tested. Pressure differentials are commonly determined with instruments such as a calibrated Shortridge air data multimeter or a Modus digital Manometer. "Tell Tales" and smoke devices can assist in determining flow direction and whether or not a specific diffuser or register is supplying or exhausting air.
Typically, one or two members of the commissioning team in the zone of incidence will take measurements, and another will be stationed at the bas computer to review the operation of hvac devices and monitor alarms. The general contractor and any subcontractors involved in portions of the smoke control system should be available to assist with the test; building operations and maintenance staff should also be present for training purposes.
Specific systems focusing on engineered smoke control that require testing may include:
Figure 1 demonstrates an example of staffing that might be used to test a typical smoke control zone or other portions of a smoke control system, as listed above.
Problem 1: Inappropriate signal input for control of pressure during smoke events.
Example: Several designs specify modulating airflow (via a variable-frequency drive or using automatic control dampers) to maintain the required stairwell or hoistway pressure. These volume control devices often receive input from a differential pressure sensor. If, for example, the sixth floor is chosen as a differential pressure reference and the sixth floor location is either a negatively pressurized smoke zone or a positively pressurized adjacent zone, control devices can cause the net stairwell pressure to either be too low or too high, respectively.
Resolution: Avoid the use of stairwell or hoistway pressure control devices that require input from a differential pressure device sensing between a possible smoke zone or adjacent zone. In a 10-story building, a smoke alarm will affect at least three floors if a pressure sandwich scheme is utilized as outlined by ASHRAE and NFPA-92A. This means that by using this control, there is at least a 30% chance that the stairwell pressure will be adversely affected during a smoke incident.
Problem 2: Response time of typical bas is inadequate for maintenance of smoke control pressure requirements.
Example: Designs often attempt to use bas controls to maintain a constant pressure in a pressurized stairwell. Pressure fluctuations in the stairwell occur frequently and are nearly instantaneous with the opening or closing of stairwell doors. Utilizing standard bas controls to try to maintain stairwell pressure to a fixed setpoint will generally not work well.
Resolution: Experience indicates that a properly calibrated, constant-volume stairwell pressurization system is generally the most effective. Opening and closing doors will modify the stairwell pressure, but trying to incorporate a response system to eliminate the effect of instantaneous pressure changes is not often practical. If such a system is to be incorporated, it should utilize a mechanical response (i.e., via a spring-loaded volume control device), such as through the use of Phoenix(r) air control valves or similar systems.
Problem 3: Sweeps and gaskets on stairwell doors are omitted or poorly installed.
Example: Excessive air leakage from the stairwell will occur, causing an inability to achieve the code-required minimum +0.15 in. wc pressure in the stairwell.
Resolution: Specify the acquisition of gasketed, fire-rated door assemblies, and the proper installation of adjustable door sweeps.
Problem 4: The use of open/close-only smoke dampering may result in excessive zone-of-incidence egress door forces.
Example: In an already heavily exhausted lab building, the complete closing of supply air smoke dampers serving an affected floor could cause excessive floor depressurization. Floor depressurization and egress stairwell pressurization can combine to make the door opening force exceed life safety limits.
Resolution: For sprinklered buildings, consider installing either analog control supply air smoke dampers or adjustable, mechanical stops such that supply air is not completely shut off from the floor of incidence. The analog setting or the mechanical stop should be adjusted during commissioning as zone differential pressures are being measured to prevent excessive floor depressurization.
It may be argued that allowing some supply air to enter could fan the flames. However, unless the floor is of zero leakage construction, "makeup" air will be supplied to the floor, but uncontrollably through smaller cracks, and most notably through elevator and stairwell doorways. Introducing a controlled volume of supply air will prevent excessive floor depressurization, thus reducing the force holding egress doors shut to an appropriate level.
Problem 5: Improper sequence of AHU functions with respect to smoke control under certain environmental conditions.
Example: During winter testing, an AHU commanded to enter smoke-purge mode (i.e., 100% outside air) tripped on freeze protection and shut down.
Resolution: The impact of the sequence of operation for smoke alarm events should be carefully considered during design and appropriate performance under maximally challenging conditions verified. Hard-wired AHU safeties that prevent unit destruction under normal operation (such as overload protection and high static pressure cutout) should not be disabled. Freeze protection for coils could be disabled under smoke alarm events.
One potential programming solution is to disable the freeze trips under any building smoke alarm condition and command both heating water and chilled water coil isolation valves to open to 100% during smoke purge mode whenever outside air temperature is below 40?F.
Problem 6: Improper programming of smoke detector interface with bas.
Example: In commissioning a new clinical inpatient building, it was found that patient room smoke detectors were programmed to provide an alarm signal to one of two (the north or the south) nurse call stations. However, smoke zoning for the patient floor was divided between an east-west boundary. Initiating a smoke alarm in any patient room along the north side of the building caused the west zone to enter smoke control mode, while the south patient rooms caused the east zone to enter smoke control mode.
Resolution: The project specification had detailed that the smoke detector alarms appearing at the nurse call stations would receive a signal from auxiliary contacts on each smoke detector. This signal was not to be intertwined with the fire alarm command center functions or bas functions affecting hvac systems. Therefore, the ATC contractor needed to reprogram each patient room alarm such that the nurse call station alarm would be separated from the normal bas functions of the detectors.
Problem 7: Improper programming of specific detector points integral to hvac smoke control functions.
Example: A hospital inpatient floor passes performance testing of the sequence of operations for smoke control mode, but fails using real smoke. East zone smoke control works until smoke passes through the return duct to the shaft on the west side. The west shaft return air smoke detector is mistakenly programmed as a west zone smoke detector. Zone smoke control was lost after the west side went into smoke control mode.
Resolution: The return air duct smoke detector was reprogrammed such that it maintained the necessary AHU purge function but did not trigger zone smoke control.
Problem 8: The 30-lb maximum stairwell door forces are readily exceeded with common stairwell doors, even with closers adjusted to provide minimum closing force.
Example: Although code specifies an upper stairwell pressure limit of +0.35 in. wc (to prevent excessive door force), testing has found that the 30-lb limit is often exceeded when stairwell pressure approaches +0.25 to +0.28 in. wc.
Resolution: Other members of the team (ranging from architects to O&M personnel) must be involved in crafting a solution. For example, it may be appropriate to consider alternate stairwell door designs, such as doors with vertical, pivoted hinges offset from the edge. These doors that rotate about an offset axis can significantly reduce the required opening force. Where more common stairwell doors are used, installers and O&M personnel must adjust closers to the minimum force needed to close the door. It is also feasible to consider use of special closers that reduce the door opening force.
Problem 9: Systems do not respond appropriately due to overrides placed on AHU dampers, fan speeds, variable-air volume (vav) setpoints, and other portions of systems involved in smoke control.
Example: Often, central AHUs and/or portions of the distribution system are overridden by the building O&M staff because they do not operate satisfactorily under automatic control. Systems are adjusted for the purposes of occupant thermal comfort, energy savings, or other environmental considerations, such as noise. It is often found during testing (and will likely be discovered during actual emergencies) that these overrides eliminate the smoke control capabilities of the system.
Resolution: It is important to ensure the central AHUs perform properly and in accordance with design intent in automatic mode such that functions do not have to be overridden. From then on, a policy should be in place that requires documentation of any manual overrides with consideration of potential impacts so that operators can take appropriate action during emergencies to ensure proper performance.
Problem 10: Specification of elevator hoistway pressurization vs. establishing a design airflow rate.
Example: Controlling mechanical volume devices by sensing hoistway pressure will likely not be practical. The net open area around elevator doors ranges from 0.55 to 0.70 sq ft (per the 1999 ASHRAE Applications Handbook, Chapter 51). For a hoistway in a 14-story building with three elevator doors at each floor, there may be 30 sq ft of net open area around the elevator doors, neglecting system cracks and open ducts to machine rooms.
Resolution: Where hoistway pressurization is proposed in the design, the designers should be required to provide a design air volume with fixed volume control devices. For example, if the hoistway in the above example is specified to maintain +0.15 to +0.35 in. wc (similar to a pressurized stairwell), ASHRAE Applications, Chapter 51 Equation 9 would calculate that roughly 30,000 to 46,000 cfm of air would need to be supplied to the hoistway.
Several other common deficiencies have been discovered in design, construction, and operation and resolved through commissioning. Some of the most common issues include: corridor smoke boundary doors being caught on their jambs, mechanical difficulties with doors that prevent them from shutting upon being released from their magnets, improper zoning of smoke detectors, dampers binding, loose damper actuator linkages, and inappropriate (or lack of) system response.
The purpose of engineered smoke control systems is very simple: to prevent the loss of life and property. Commissioning a successful system requires appropriate design, which can be guided by such publications as ASHRAE Chapter 51 in the 1999 Applications Handbook, ASHRAE and the Society of Fire Protection Engineers' Design of Smoke Management Systems (John H. Klote and James A. Milke, copyright 1992), and NFPA-92A. It also requires developing detailed project specifications, paying extra attention to review of submittals affecting proper whole-system operation, correct installation and programming, and appropriately communicated contracts and specifications.
Every member of the construction team has a stake in proper system operation, and all should understand and accept their roles in making the system work. These roles should be defined prior to the bidding phase of the project.
Testing the systems should be a team effort coordinated by the CA, and should include building operations and maintenance staff for training purposes. Utilizing an effectively implemented commissioning program will help to assess potential problems before they occur and will verify system performance through detailed testing.
Lessons learned from commissioning of these systems will help architects and engineers avoid common pitfalls, assist them in developing more effective designs in the future, and provide owners and occupants with a safer building. ES
EDITOR'S NOTE: The images in figures 2, 3, and 4 do not transfer well to this website and as a result are not included here. Please reference the print version of Engineered Systems to view the images.
National and local codes generally demand that certain smoke control-related parameters must be met if a building exceeds a specified height, typically 70 ft. For instance, the building must contain at least one smoke-proof enclosure isolated on each floor protected by a vented vestibule. The enclosure (a fire-rated stairwell running the height of the building), should also be positively pressurized. The vented vestibule on each floor must be provided with clean makeup air near the floor level at a rate no less than 60 air changes per hour (ach), and exhausted at no less than 90 ach, making the vestibule's pressure negative. The remaining building stairwells must be positively pressurized to +0.15 to +0.35 in. wc with respect to each floor. The lower limit of 0.15 in. wc is designed to prevent the migration of smoke into the stairwell, while the upper limit of 0.35 in. wc ensures that stairwell doors are not too difficult to open due to excessive differential pressure.
There are several additional life safety measures that are required to be implemented in buildings that are not fully sprinklered; these requirements are often "softened" to recommendations in buildings that are sprinklered in accordance with NFPA-13. NFPA-92A recommends that even in a fully sprinklered building, designs that utilize smoke zoning maintain boundary differential pressures no less than 0.05 in. wc to provide proper smoke containment.
Commissioning Overview
The CA should review the smoke control system's design prior to installation to ensure compliance with building codes, NFPA standards, and the design intent. In addition to reviewing hvac systems and ATC sequences, the CA should also review specifications such as the type of stairwell doors to be used; installation of adjustable door sweeps and gaskets on stairwell doors; requirements for patchwork, finishes, fire stopping; the ATC specification for devices and response times, location of smoke detectors with respect to smoke zone boundaries, and smoke control overrides.1
Problems Found During Commissioning
Design review and testing assists the building team in proactively combating potential problems. The following are ten examples of deficiencies found during commissioning of various buildings and the corrective action taken.Design Standards Evolve
In 1993, the National Fire Protection Association (NFPA, Quincy, MA), one of the most accepted and authoritative sources on issues of fire control and prevention, initially published NFPA-92A, "Recommended Practice for Smoke-Control Systems." This document describes methods of designing and testing engineered smoke control systems. The smoke control ideology presented is compiled from publications by ASHRAE and by individual studies by NFPA members and consultation with leaders in the fire protection and hvac fields. Such publications have guided the model building and local building codes to establish requirements for building construction and renovation. The current edition (2000 Edition) of NFPA-92A was approved as an American National Standard on August 18, 2000.
