Fume Hoods and Lab Control
(Avoiding Unsafe Laboratory Design Practices)

Scientific laboratories are equipped with fume hoods in order to improve the level of safety of people conducting experiments with products that would represent a threat to the health, and possibly the life, of someone using the products in some unprotected way. Working in a laboratory is a high-risk activity⁽¹⁾, and there are a number of things that can make that activity categorically unsafe.

Introduction

Anything that increases the risk of exposure of laboratory workers to dangerous products will make the laboratory an unsafe place to work. Factors to consider are:

1. the overall design of the laboratory,
2. the selection of laboratory equipment such as fume hoods,
3. the institution of safe working procedures, and
4. the design and operation of control systems for fume hood exhaust and laboratory pressurization.

This latter factor is particularly insidious because the operation of control systems is rarely completely understood and verified and, worse, its importance can be easily underestimated. For example, the very presence of fume hoods, along with a good supply of outside air can be mistakenly assumed to be all that is needed to meet safety requirements.

The fact is that if the control system does not meet certain specific design criteria, the operation of any fume hoods and laboratory pressurization will be clearly unsafe.

This claim can be easily substantiated by examining the scientific principles involved in the operation of fume hood and laboratory pressurization control systems. This paper therefore focuses on the key issues in control system operation, which are frequently glossed over by designers, thus leading to unsafe conditions in the laboratory. Designers must always bear in mind all the other safety considerations that are beyond the scope of this paper.

General Causes of Unsafe Control Operation

Knowing what causes unsafe control operation requires knowing first what is needed for the safe operation of a fume hood and laboratory pressurization control system. In general, a control system will be safe if:

• with regard to the air entering the fume hood, there is a constant average face velocity at all positions of the sash,

• the control system is able to change the fume hood exhaust flow virtually instantaneously in response to changes in sash position, and

• the laboratory pressurization control system reacts stably in response to changes in fume hood exhaust airflow, and is not affected by disturbances such as the opening and closing of laboratory doors and changes in duct static pressure.

Face Velocity

There are numerous agencies, industry societies, and standards organizations^(2,3,4,5) that call for a constant average fume hood face velocity (on the order of 100 feet per minute). In general, if the face velocity is too low, the hood will not be capable of containing fumes generated inside the fume hood. If it is too high, delicate equipment inside the hood may be upset and, in addition, turbulent airflow resulting from the high velocity can cause fumes to escape from the hood. The American Society of Heating, Refrigerating, and Air Conditioning Engineers⁽⁵⁾ has clear guidelines on the allowable variations in face velocity.

Response to Sash Position Changes

The face velocity can remain constant only if the ratio of exhaust flow to sash opening is maintained (average face velocity = exhaust flow/sash opening, e.g. 100 ft/min = 1000 ft/min &#divide; 10 ft²). The literature⁽⁶⁾ indicates that the fume hood control system must be able to change the fume hood exhaust flow within three seconds in response to changes in sash position; otherwise, fumes may escape from the hood. Dynamic fume hood containment tests conducted by Phoenix Controls Corporation⁽⁷⁾ suggest that the response time of the face velocity control system must be virtually instantaneous to prevent the escape of fumes from the hood when the sash is moved from the closed to the open position.

Laboratory Pressurization

Because of the nature of the substances being used, many laboratories must be maintained at a constant pressure (positive, negative, or neutral) with respect to their surroundings. Should the required pressure not be maintained, laboratory activities will be put in jeopardy. In the worst case, toxic substances could escape from the laboratory into neighbouring areas, thus contaminating them, with possibly tragic results.

Specific Causes of Unsafe Control Operation

Causes of Face Velocity Variations

The primary reasons that the face velocity is not maintained at an average constant value are as follows:

• It is not controlled at all. This would be the case with all traditional hoods - whether or not they have a bypass. The area of a bypass opening is considerably smaller than the fully open area of a sash. Assuming a ratio of one to three for the bypass area to the open sash area, one can perform a simple calculation to show that there will be a wide variation in face velocity from the fully open to the fully closed positions of the sash. This does not take into account that someone might reduce the bypass area even further by covering it with paper (such as notices, instructions, reminders, and so on).

• Face velocity control is done by other than sash position sensing. Traditional techniques that use velocity sensors have proven unreliable because of: (1) the inability to measure the average face velocity, (2) a weak signal measurement, (3) a low signal-to-noise ratio, (4) low measurement accuracy, and (5) slow response.

• Exhaust flow capacity is not accurately controlled. Sash position can be measured to a high degree of precision, so the precision of face velocity control depends on the degree of precision with which the exhaust flow can be controlled (average face velocity = exhaust flow/sash opening, e.g. 100 ft/min = 1000 ft³/min &#divide; 10 ft².

• Auxiliary air to the hood is not controlled. In the case of hoods with auxiliary air supplied directly to the hood, there is a high probability that the airflow will vary with variations in duct static and hood flow. This adds an element of unpredictability to the system which is an unnecessary risk to take. In addition, if the auxiliary airflow has a downward velocity component of more than only 10 ft/min, it will have a guillotine effect on the air stream entering the hood, thus precluding a safe face velocity. In the worst case, on a loss of hood exhaust airflow, a continued flow of auxiliary air will positively pressurize the hood, causing a flow from the hood to the lab. It is worth noting that Public Works Canada prohibits the use of hoods with auxiliary air⁽⁸⁾.

Other factors which prevent the face velocity from being maintained at an average constant value are as follows:

• Supply air diffusers are too close to the hood.

• The hood is placed in a heavily trafficked area.

• Laboratory staff have careless work habits.

Such factors are related to overall laboratory design and operation and are outside the scope of this paper.

Causes of Slow System Response

The primary reasons that the control system may have a speed of response that exceeds the recommended three seconds are as follows:

• The face velocity is controlled by other than sash position sensing. All other techniques have simply proven to be too slow. Controlling from a sidewall velocity measurement is a catch-22 situation. The normally high speed of response of the velocity sensor has to be slowed down in order to provide an integrated output signal that does not "hop all over the place" due to momentary measurement changes. On the other hand, increasing the speed of response will lead to unstable control system operation.

• The control system has inherent time delays. An example of this would be a digital computer control system⁽⁹⁾ whose signal sampling time is too slow. This could result from the misapplication of other-than-laboratory controls to a laboratory application.

• The device that controls the exhaust airflow capacity is slow-reacting. An example of this would be capacity control devices that are regulated by slow-moving electric/electronic actuators.

Causes of Unstable Laboratory Pressurization

The primary reasons that the laboratory pressurization control may be unstable are as follows:

• It is controlled by other than the volumetric offset principle. The traditional form of pressurization control, which uses a differential pressure sensor measuring laboratory pressure with respect to a reference, does not work unless the two pressure measuring tips can be rendered immune to momentary fluctuations. This is virtually impossible except in the case of an airtight, airlock-equipped high containment lab with a rock-steady reference pressure point.

• There is a mismatch between the supply and exhaust airflow capacity control devices. If the supply and exhaust devices react at different speeds this will cause a variation in supply and exhaust flows and, consequently, a variation in laboratory pressurization. If the mismatch is severe, a laboratory that is supposed to be negative could become positive, with the resulting unwanted contamination of adjoining areas.

Obtaining Safe Control System Operation

Maintain Constant Average Face Velocity

Constant average face velocity is an achievable design objective with today's technology. Design guidelines are:

• The face velocity must be controlled. In other words, do not use hoods that have no form of face velocity control at all.

• Control the face velocity using the principle of sash position sensing.

• For hood exhaust airflow control, use only a variable speed motor control designed for rigorous laboratory service, or in the case of manifolded hoods, use a proven airflow control device (such as a controllable airflow Venturi air valve which has a precision that is rated as a percentage of actual airflow and not full scale value).

• Do not use hoods with an auxiliary air supply. If it exists already, convert it to a safe design with the proper face velocity control.

• Follow recommended general laboratory design and operating practices to avoid things like supply air diffusers that are too close to the hood, hoods that are placed in a heavily trafficked areas, and the development of careless work habits.

Ensure a Rapid Speed of Response

A safe speed of response is also an achievable design objective with today's technology. Design guidelines are:

• Use the sash-position sensing principle only. The input signal from a sash position sensor is virtually instantaneous.

• For hood exhaust airflow capacity control, use only a variable speed motor control designed for rigorous laboratory service, or in the case of manifolded hoods, use a proven airflow control device (such as a controllable airflow Venturi air valve controlled by a combination of pneumatic actuator and electronic analog controller).

Ensure Stable Laboratory Pressurization

Stable laboratory pressurization is another achievable design objective. Design guidelines are:

• Control the laboratory pressure by accurately controlling the supply and exhaust airflows. Avoid the use of direct pressurization control based on a signal from a differential pressure sensor, except in the case of a "completely" sealed, high-containment laboratory with a stable reference pressure point, in which situation the designer must still proceed with extreme caution.

• Avoid a mismatch of capacity control devices on supply and exhaust airflows. This is readily done by using proven airflow control devices (such as controllable airflow Venturis on the supply and exhaust in the case of manifolded hoods, or Venturis on the supply with variable speed controllers on single exhaust hoods).

Owner/Designer Liability

With today's technology, there are design alternatives that will result in the safe operation of fume hoods and pressurized laboratories. Given the safe alternatives, the owner and designer (and anyone else involved in the design) put themselves in a position of crushing liability if they knowingly choose an unsafe alternative. In the case of an accident that results in the death of a worker, this liability could be criminal.

In effect, overlooking just one of the operational problems mentioned in this paper could lead to an unsafe operation. Accordingly, the owner/designer is advised to proceed with extreme caution (even to the point of consulting with the firm's legal counsel and insurance representative to establish limits of liability) when designing a laboratory that uses substances that are in the slightest way hazardous to health. It is to be noted that liability insurance is unlikely to cover a designer's conscious act of neglect.

References

(1) Jones, Robert H., Editor. 1985. "Injuries and Death Needlessly Haunt R&D Laboratories." Research and Development, August. pp 70-74.

(2) Laboratory Fume Hoods. LF-10 1980. Washington, D.C.: Scientific Apparatus Makers Association.

(3) Prudent Practices for Handling Hazardous Chemicals in Laboratories. 1981. Washington, D.C.: National Academy Press of the National Science Foundation.

(4) 29 CFR Part 1910. Occupational Exposures to Hazardous Chemicals in Laboratories, Final Rule. Occupational Safety and Health Administration.

(5) 1991 ASHRAE Handbook, HVAC Systems and Applications. 1991. Atlanta, Georgia: American Society of Heating, Refrigerating, & Air-conditioning Engineers.

(6) Ahmed, O., and S. A. Bradley. 1990. "An Approach to Determining the Required Response Time for a VAV Fume Hood Control System." ASHRAE TRANSACTIONS 1990 V.96 Pt.2. Atlanta, Georgia: American Society of Heating, Refrigerating, & Air-conditioning Engineers.

(7) Dynamic Fume Hood Containment Tests. Video. 1992. Newton, Massachusetts: Phoenix Controls Corporation.

(8) Standards and Guidelines - MD 151128. 1988. Laboratory Fume Hoods, Section 3, Subsection 6, p 6 of 6. Ottawa, Canada: Public Works Canada.

(9) Technical Support Package, NASA Tech Briefs - ARC-12710. Ames Research Center. How Safe is Control Software. Moffet Field, California: National Aeronautics and Space Administration.

Author

Donald A. Coggan, PE, is recognized internationally as an expert in the field of control systems design and training. In addition to consulting directly to clients in the United States and Canada, he has addressed groups throughout North America as well and in Europe and Asia. He is the originator of a design evaluation technique called "Specifying for Maximum Value" based on principles set out by the Society of American Value Engineers (SAVE). Mr. Coggan has authored numerous technical publications including a training system and accompanying software for instrumentation technician evaluation for the Instrument Society of America (ISA). He has also co-edited Fundamentals of Industrial Control, the flagship volume of the ISA's Practical Guide Series.

Return from this fume hoods page
to the value engineering page.