Printer Friendly

Energy solution for laboratory facilities.

THE ENIGMA

To design a ventilation system, the volume of airflow (L / s [cfm]) necessary to meet the requirement must first be determined. This determination remains unknown for laboratory fume hoods where the safety of the user is the primary requirement. The assumption that higher face velocity is better results in oversized HVAC systems in laboratory facilities that do not necessarily meet the needs of user safety.

In 1979, an article published in ASHRAE Journal recommended that a face velocity of 0.3 m/s (60 fpm), with the sash fully open, was sufficient to protect the user (Fuller and Etchells 1979). The recommendation was based on the observation that researchers work with the hood sash partly closed and are, therefore, protected by higher face velocities during normal operation. At the time, variable-frequency drives and variablevolume controls were not yet developed. The recommendation was made under the assumption that a fume hood is a constantvolume airflow device. The volumetric airflow of the fume hood was determined using the relationship Q = VA where Q is volumetric flow (L/s [cfm]), V is face velocity (m/s [fpm]), and A is the face area with the sash fully open ([m.sub.2] [[ft.sub.2]]).

Constantly challenged with energy conservation, engineers realized that laboratory hoods with horizontal sliding sash on a double track have half the open face area of a vertical sash hood of the same size. By using face velocity as a constant, hoods with horizontal sliding sash would require only half the volumetric air flow as hoods with vertical rising sash. A program to convert hoods from vertical to horizontal sash configuration was initiated. The sash conversion program met stiff opposition from some in the research community that preferred vertical rising sash for ergonomic reasons. A research chemist devised a "dry ice" test, using typical laboratory implements, to demonstrate that horizontal sash hoods do not contain as well as vertical sash hoods using the same face velocity with the sash fully open (ASHRAE 1995). Consequently, the sash conversion program failed as the result of an erroneous assumption. If containment is a function of volumetric flow Q, and given that volumetric flow Q is associated with face velocity V by Q = VA, it does not necessarily follow that containment is a function of face velocity V. The objective of safe and efficient laboratory hoods can be improved by finding an alternative that does not rely on face velocity to calculate the volumetric flow of air through a fume hood.

The current standard requires that a design open face area be designated but does not indicate what the open face area should be (ANSI/AIHA/ASSE 2012). Because the hood sashes can be positioned anywhere from fully open to closed, open face area A is a variable. Consider a medium sized hood that is 2.4 m (8 ft) long. The design open face area can be as small as 0.092 [m.sup.2] (1 [ft.sup.2]), or, as large as 1.9 [m.sup.2] (20 [ft.sup.2]). The standard also prescribes a range of possible face velocities V as little as 0.3 m/s (60 fpm), or as great as 0.6 m/s (120 fpm) (ANSI/AIHA/ASSE 2012). Using the relationship Q = VA, the resulting volumetric air flow can be as little as 28 L/s (60 cfm) or as large as 760 L/s (2400 cfm) for the same hood. Volumetric air flow Q determined in this manor is ambiguous and does not assure containment or maximize fume hood efficiency. In the absence of consensus that a specific value of face velocity V is known to contain over any open face area, the relationship Q = VA is not a viable means to determine the volumetric flow of air Q through a laboratory fume hood.

A fume hood is a rigid structure. Once built, the interior volume is constant and does not change. It is reasoned, then, that the volume of air that flows through the hood should also be constant while in use. The hood sash move to create larger and smaller openings making open face area A a variable. If Q is constant and A is variable, given the relationship Q = VA, the face velocity V must also be a variable. The volumetric flow of air through a partially enclosed space and the velocity of the air at a plane in that space are distinctly different aspects of air flow and must not be thought of as the same. Face velocity has been used to characterize fume hood performance because it is easy to measure and because of the lack of an alternative. Reasoning that hoods of the same size should have the same volumetric air flow and given that the laboratory design standard warns that face velocity alone is not an indicator of hood performance, then design air flow Q should be determined independently from face velocity V (ANSI/AIHA/ASSE 2012). Determining the volume of air necessary to assure fume hood containment has remained an enigma.

Experts in the field of ventilation favor the concept of laminar flow as a means to achieve hood containment (ANSI/ AIHA/ASSE 2012; ASHRAE 1995; Bell et al. 2002). A hood design, considered state-of-the-art, prefers laminar flow that is minimally turbulent (Bell et al. 2002). Laminar airflow through a partially enclosed space would require a consistent cross-sectional area, such as a straight run of uniform duct. This is inconsistent with the structure of a traditional fume hood (Figure 1). Air enters the hood through openings in the movable sash. Once inside, the air expands into the hood interior that has a larger cross-sectional area than the open face of the hood. Then, the air must compress as it squeezes through narrow slots in the rear of the hood, exiting the hood interior. Laminar flow only occurs as the air flows through the slots at a high velocity. Because the exhaust slots cause resistance to the air that flows through the open face, the air inside the hood is chaotic and turbulent. For a traditional fume hood, increasing face velocity to improve containment only increases turbulence in the hood interior work space and at the open face.

A critical aspect of containment is that air at the open face of the fume hood flows away from the breathing zone of the user. Health and safety professionals presume to insure researcher safety by measuring and documenting air velocity at the open face of the hood (ANSI/AIHA/ASSE 2012; ACGIH 2013). The open face is an imaginary plane that slices through a turbulent mass of air in motion at the hood entrance. At any point on the open face, air may be entering or spilling from the opening as the majority of the air mass flows into the hood. Airflow through a traditional fume hood defies vector analysis because the direction of the air entering the hood from the room is not necessarily perpendicular to the plane of the open face. Face velocity measurement provides a resolution of true vectors that is perpendicular to the open face (ASHRAE 1995). Face velocity measurements provide information on magnitude (velocity) only, but the true direction of air entering a hood is unknown. The turbulent nature of air in free space makes the open face of the hood susceptible to outside influences such as movement in front of the hood and the way the air enters the room. The concept of laminar flow does not provide a practical model to characterize the motion of air flowing through a laboratory fume hood.

THE SEARCH

As a mentor, Frank Fuller ignited a smoke stick inside a fume hood to demonstrate the "rolling" effect that occurs in the upper portion of the hood. Practitioners have long been aware of this phenomenon (ANSI/AIHA/ASSE 2012; ASHRAE 1995; Bell et al. 2002; Meisenzahl 2014). Turbulence, such as vortices, is thought to be detrimental to fume hood performance. The current testing standard states that if smoke is enveloped in a vortex the test should be considered a failure (ASHRAE 1995). Laboratory hood design emphasizes the minimization of turbulence such as vortices (Bell et al. 2002). Yet, the rolling effect in the top of a hood occurs naturally and appears as a repeating pattern of motion as the air tends to rotate in the same direction. Rather than disregarding the vortex effect, consider it an avenue to a better understanding of how air flows through a partially enclosed space. Curiosity drove the search to enhance the vortex phenomenon in a way that improves fume hood containment and makes the direction of air motion inside the hood more predictable (ASHRAE 1995, Meisenzahl 2014).

The vortex is enhanced by changing the architecture of the hood interior. The interior geometry of the vortex hood is derived from the Fibonacci sequence where each number in the series is the sum of the two previous numbers (e.g., 1,2, 3, 5, 8, 13, 21,34). When a number in the series is divided by the previous number, the result is an infinitely repeating decimal known as the golden number ([phi] = 1.618 ...). If the base of a rectangle is the unit is one, and, the height is [phi], the result is known as the golden rectangle (Figure 2).

A characteristic of this geometry is that if a perfect square is removed from the golden rectangle the result is another, smaller golden rectangle. This repeats infinitely. When the diagonal corners of the squares are connected by a smooth curve, the result is known as the golden spiral (Figure 2). This is a shape that often repeats in nature.

The geometry derived from the Fibonacci sequence is translated into an interior architecture for a laboratory fume hood. Figure 3 represents two variations of the vortex hood design as viewed in profile from the center of the hood. The golden rectangle represents the hood interior profile. The golden number [phi] provides the proportion of the hood depth D to the interior height H of the hood, or H = 1.618(D). A single slot (5) is used to communicate air from the hood interior to the exhaust duct. Slot (5) is located a distance S equal to the working depth D above the work surface (6), or S = D. The height of the open face is located a distance O equal to or less than the working depth D above the work surface, or O [less than or equal to] D. The vortex baffle (1,3, and 4) is an arched vault extending over the length of the hood interior. The vortex baffle may be semicircular (3) in shape or it can be five flat panels (4) that approximates the semicircular shape.

Air enters the hood interior work space through the front opening (2). The tendency of a fluid to stick to smooth surfaces is known as the Coanda effect. The vortex hood uses this effect by directing the airflow across the work surface (6) and up along the rear lining of the hood towards the single slot (5). Initially, most of the air is exhausted through the slot. Air, which does not go through the slot, sticks to and follows the interior surface of the vortex baffle (1, 3, 4). The flowing air follows the arched shape of the baffle up, across, and down. The air mass rolls in upon itself and rotates beneath the vortex baffle.

For the profile shown in Figure 3, the air mass will rotate in a counterclockwise direction. Air that continues to flow through the open face (2) is induced by the rotating air beneath the baffle forming a stable unitary vortex. The vortex exhibits a fly-wheel effect such that all of the air inside the hood is directed towards the single slot (5). This causes a low pressure area, at the open face, in the middle of the opening (2) so that the exhaust air flows away from the breathing zone of the user standing in front of the hood.

To demonstrate the vortex effect, a chamber was constructed (Figure 4) with an interior shape prescribed by the vortex hood design (Figure 3). The chamber is constructed of clear acrylic sheet using hinges and fasteners. Smoke is produced from a theatrical fog machine using 700 W, AC120/ 60Hz, with a capacity of 1180 L/s (2500 cfm). Smoke is injected into the front bottom of the chamber through 1 1/4 in. (3.18 cm) PVC pipe. The pipe manifold, fixed at both ends to the bottom corners of the chamber, has 1/4 in. (0.63 cm) diameter holes at 1 in. (2.54 cm) intervals facing the chamber interior. The chamber is exhausted by a mixed flow in-line duct fan with 4 in. (10.16 cm) connections, 26 W, AC120/60Hz, which draws 33 L/s (70 cfm) of air out of the chamber. The chamber volume is 139 L (4.9 [ft.sub.3]) that resolves into about 14 air changes per minute. The open face area is 0.22 [m.sup.2] (2.42 [ft.sup.2]) with about 0.15 m/s (30 ft/min) face velocity.

The demonstration chamber verifies, by repeated observation, that the air inside the enclosure moves in the direction predicted by the golden spiral (Figure 2). The air flow inside the enclosure is turbulent and a stable vortex forms in the vaulted part of the chamber. The rotation of the air (vortex) inside the enclosure creates a flywheel effect that harnesses turbulence in a way that improves containment using a relatively small volume of exhaust air.

THE DISCOVERY

In 2014, the Journal of Occupational and Environmental Hygiene (JOEH) published an article describing an experiment that compares the traditional laboratory fume hood to the vortex hood (Meisenzahl 2014). The experiment was conducted with a fume hood having a dedicated fan with a variable speed drive. The design of the experiment was to thoroughly test a fume hood of traditional design, then modify the hood by removing the rear baffle (Figure 1) and replace it with a vortex baffle (Figure 3a). The modified hood was retested using the same conditions and test procedure used for the previous hood configuration. The results compared the performance of the traditional hood with vortex hood.

The experiment used the ASHRAE 110 tracer gas test standard using sulfurhexaflouride ([SF.sup.6]), and a Miran Series 205B ambient air analyzer that detected tracer gas in parts per billion (ppb) (Meisenzahl 2014, ASHRAE 1995). The data collection plan included more than 1000 instrument readings that were averaged into 108 data points (Meisenzahl 2014). Across all of the data, the vortex hood contains twice as well as the traditional hood, spilling only half as much tracer gas as the traditional hood configuration (Figure 5). The data collected during the experiment was highly variable. This was expected because of the turbulent conditions associated with air flowing through a partially enclosed space.

The initial analysis of the data (Meisenzahl 2014) was inconclusive regarding a causal relationship between volumetric air flow and containment within the hood.

The hood test standard specifies an ejector system that releases 4 L per minute (0.14 cfm) of tracer gas inside the hood. The standard prescribes measuring the concentration of tracer gas detected outside the hood (spill) rather than how much of the tracer gas is contained within the hood. An arbitrary value of spill, such as 0.1 parts per million (ppm), is used as a lower limit to determine if the test is a success or failure. The test standard does not require measuring the concentration of tracer gas inside the hood (ASHRAE 1995). The highest concentration of tracer gas inside the hood would be at the discharge of the ejector, where the test gas begins mixing with air. The full range of tracer gas content from highest to lowest concentration, in the air stream, remains unknown. Measuring only what spills out of the hood does not characterize the full range of hood containment.

The experiment comparing the traditional hood with the vortex design involved taking data at different times over several days. The only way to calibrate the experiment was to make sure that the highest concentration of tracer gas inside the hood was the same for each session data was collected. The analog instrument used in the experiment had a range large enough to measure very high as well as very low concentrations of tracer gas (Meisenzahl 2014). At the beginning of each session, the instrument probe was placed on top of the tracer gas ejector to record the highest concentration of tracer gas inside the hood. For the vortex design, the highest concentration inside the hood [C.sup.i] was measured at 222,600 parts per billion (ppb). Although this data was collected as a means of calibration, the availability of this information was important to the discovery of a direct causal relationship between volumetric airflow and hood containment.

All of the data collected to compare the traditional hood with the vortex hood was done using three progressive values of volumetric air flow. These values were 240, 380, and 570 L/ s (500, 800, and 1200 cfm). The data for the vortex hood was sorted by volumetric air flow. The values of spill concentration [C.sup.o] are tabulated in Table I. The way in which the spill concentration [C.sup.o] decreases as airflow increases suggests a consistent pattern associating volumetric airflow with containment.

The three data points [C.sup.o] connected with a smooth curve presents a logarithmic function (Figure 6). Tracer gas spill concentration is significant at lower airflow. The spill concentration decreases rapidly, approaching zero asymtotically, as volumetric airflow increases. Once past the middle of the curve, increase of airflow causes little increase in containment. There is an area on the curve (Figure 6), at about 425 L/ s (900 cfm), where containment is maximized and exhaust airflow is minimized.

Fuller and Etchells (1979) understood the flow of air through a fume hood as a dilution process characterized as a logarithmic function. A further contribution of his work is the hood index (HI), which is a comparative scale of values expressed as follows:

HI = -log ([C.sup.o]/[C.sup.i]) (1)

Because data was collected for both concentration of tracer gas outside the hood [C.sup.o] and concentration inside the hood [C.sup.i], values of the HI for the vortex hood are calculated and tabulated in Table I. When the values of HI are graphed with respect to volumetric airflow, the result is a straight line (Figure 7).

This was unexpected, because [C.sup.o] and [C.sup.i] are the averages of more than 600 instrument readings of highly variable data. It only takes two points to make a straight line. The probability of landing a third point on the same straight line, by random chance, is nearly impossible. This verifies a logarithmic function characterizing a direct causal relationship between volumetric airflow and fume hood containment.

HI expressed as a straight line is as follows:

HI = [V.sup.c](Q) + b (2)

where Q is volumetric air flow (L/s [cfm]), [V.sup.c] is the slope of the straight line that becomes the vortex constant, and b, the yintercept, is the value of HI when airflow is zero. With three data points, values for [V.sup.c] and b are easily solved using linear algebra. Setting the two expressions of HI equal to each other, the result is:

-log ([C.sup.o]/[C.sup.i]) = [V.sup.c](Q) + b (3)

When Equation 3 is rearranged to solve for volumetric airflow, the result is:

Q = [-log ([C.sup.o]/[C.sup.i]) -b]/[V.sup.c] (4)

This equation defines volumetric airflow Q through a vortex hood that is independent of face velocity V and open face area A providing an alternative to Q = VA.

Substituting the value estimated to be the most efficient airflow volume, Q = 425 L/s (900 cfm) in Equation 2, HI becomes 2.28. Using Equation 1 iteratively, the dilution ratio [C.sup.o]/[C.sup.i] becomes 1/190. Also, when Q = 425 L/s (900 cfm) is divided by the internal volume of hood used in the experiment, 2124 L (75 [ft.sub.3]), the result is 0.2 air changes per second, or, 12 air changes per minute (Meisenzahl 2014). Because the concept of vortex ventilation is based on shape and proportion, it is expected that any hood designed as shown in Figure 3, regardless of size, will also have the same values for the vortex constant [V.sup.c] and y-intercept b.

THE IMPACT

Laboratory administrators, responsible for allocating facility space to various research and development programs, benefit from infrastructure that is safe, efficient, and generic throughout the facility. The chemical fume hood was originally, and should remain, a constant-volume airflow device while in use. A simple HIGH/LOW position switch is recommended for the constant volume vortex hood. The full range of face velocities, indicated in the laboratory hood standard, can only be met using a constant-volume airflow that allows the researcher to fine-tune containment (safety) by positioning the sash, with smaller openings having higher velocities (ANSI/ AIHA/ASSE 2012). The Magnehelic gauge is recommended for hood use because it is an accurate and reliable means to indicate static pressure that has a direct relationship to volumetric airflow. When the fume hood is set up and balanced, the Magnehelic gauge may be marked to indicate the static pressure that verifies the airflow requirement is met. To initiate hood use, the user only needs to select the high position of the switch and observe that the Magnehelic gauge is at the prescribed setting. In this configuration, fume hood operation is safe, reliable, consistent, and easily maintained.

The vortex hood, using 12 air changes per minute, has a dilution ratio of 1/190. This is useful for research chemists and industrial hygienists to design experiments. The threshold limit values (TLVs) of many chemicals are expressed as concentrations (ppm). The dilution ratio defines the reduction of concentrations in the vortex hood that can be used to plan and limit the amount of material used in an experiment. Vortex ventilation benefits safety professionals because they will be able to audit for specific values of volumetric air flow (L/s [cfm]) exhausting the laboratory hood. Technicians have become adept at measuring face velocity. one only needs to measure the open face area and hood depth to calculate volumetric air flow and determine if 12 air changes per minute is achieved. The volumetric airflow (L/s [cfm]) can be posted on the hood for auditing.

Fume hood design is simplified because the specifying engineer only needs to know two numbers. One is 1.6, which is the proportion of the hood working depth to the interior height of the vortex baffle. The other is 12 air changes per minute where the engineer, knowing the interior volume of the enclosure, can specifically calculate the exhaust load. A design face opening is no longer necessary. It does not matter if the hood has horizontal or vertical sash or if it has no sash at all, because the required air flow is a function of the interior volume of the hood which remains constant. The design of the vortex baffle is straightforward. It is a fixed rigid structure with no moving parts and requires no utility connections. The need for building automation system (BAS) controls is minimal for the vortex hood. A simple high/low switch is sufficient to conserve energy. The high setting is to achieve 12 air changes per minute when the hood is in use. The low setting is best determined by the design engineer with knowledge of the air-conditioning load (cfm [L/s]). With a reduced total exhaust load, automated controls used for cross-tracking between supply and exhaust air may not be necessary. It is proposed that the rotating vortex inside the hood is strong enough to stabilize the hood open face. Challenges that vex the traditional hood design, such as movement in front of the hood or the way supply air enters the room, are not a concern with the vortex hood.

The greatest impact of vortex ventilation would be realized by the owners of laboratory facilities. The operating cost of such buildings is driven by the amount of once-through exhaust air that must be heated and air conditioned. The following is an example that compares the operating cost of a traditional fume hood to the vortex hood using realistic assumptions. Consider nominal laboratory fume hood that is 1.83 m (6 ft) long with an interior length of 1.62 m (5.33 ft) and a sash full open with height of 0.8 m (2.66 ft.) for an open-face area of 1.3 [m.sup.2] (14.18 [ft.sup.2]). Using face velocity at 0.4 m/s (80 ft/ min), times 1.3 [m.sup.2] (14.18 [ft.sup.2]), is 0.53 [m.sub.3]/s or 530 L/s (1134 cfm). The same hood with a modified interior vortex shape and a working depth of 0.68 m (2.25 ft), multiplied by the golden number 1.6, has an interior height of 1.1 m (3.6 ft.). The interior volume of the hood is 1.62 m (5.33 ft) long times 0.68 m (2.25 ft.) deep times 1.1 m (3.6 ft.) high, equaling 1.2 [m.sub.3] (43 [ft.sub.3]). 1.2 [m.sub.3] equals 1212 L. Using 0.2 L/s (12 air changes per minute) times 1212 L (43 [ft.sub.3]) equals 240 L/s (516 cfm). The vortex hood requires about half of the volumetric air flow as a traditional hood of the same size. Using an estimated cost of $6 per cfm per year, the vortex hood can reduce the operating cost of a nominal six foot hood by $3700 per year.

Vortex ventilation is well-suited for use in older laboratory buildings with high operating cost. It is relatively easy to retrofit a traditional laboratory hood to a vortex hood by replacing the rear baffle with an arched vault. The retrofit requires rebalancing of the HVAC system. but otherwise there is no impact on facility infrastructure or laboratory space. Vortex hood conversion makes an ideal capital improvement project because a one-year payback period can be expected on the project cost. This is especially true for older facilities that do not garner capital investment. Using less utilities (electricity, steam), the operating cost per floor space area will decrease incrementally. For new construction, the vortex baffle does not add to the first cost of the equipment. The substantial reduction of exhaust load will reduce the size and capacity of HVAC equipment and therefore improve the return on net asset. The simplicity of the vortex baffle is a safe and sure energy solution for laboratory facilities.

ACKNOWLEDGMENTS

In honored memory of Frank Fuller (deceased) and colleagues from DuPont: Barbara Dawson, Gee Joseph, Aaron Chen, John Rydzewski, Frank Olszewski, William Moye, Mark Mueller, Joseph Lala, and John Fitzpatrick; colleagues at Delaware Engineering & Design Corp., Steve Krinsky and John Leslie; and at Pennoni Engineering, Robert Hayden, Bill Davison, and Bob Opreska. A patent is pending for the design shown in Figure 3.

REFERENCES

ACGHI. 2013. Industrial ventilation, a manual of recommended practices, 28th ed. Cincinnati, OH: American Conference of Government Industrial Hygienists.

ANSI/AIHA/ASSE. 2012. Z9.5-2012 American National Standard for Laboratory Ventilation. Washington, D.C.: ANSI.

ASHRAE. 1995. ASHRAE Standard 110. Method of testing performance of laboratory fume hoods. Atlanta: ASHRAE.

Bell, C.G., H.E. Feustel, and D.J. Dickerhoff. 2002. Low flow fume hood. United States Patent Number: US 6,428,408 B1.

Fuller, F., and A.W. Etchells. 1979. The rating of laboratory hood performance. ASHRAE Journal 21(10):49-53.

Meisenzahl, L.R. 2014. Vortex ventilation in the laboratory environment. Journal of Occupational and Environmental Hygiene 11(10):672-9 2014

Lawrence R. Meisenzahl

Member ASHRAE

Lawrence R. Meisenzahl is retired from the DuPont Co. engineering department and is currently a Senior Consultant with Vortex Hoods, LLC, New Castle Delaware,

Caption: Figure 1 Traditional laboratory fume hood.

Caption: Figure 2 The golden rectangle/spiral.

Caption: Figure 3 Vortex fume hood design.

Caption: Figure 4 Vortex effect demonstration chamber.

Caption: Figure 5 Containment of traditional/vortex enclosure.

Caption: Figure 6 Concentration outside hood/volumetric airflow.
Table 1. The Vortex Hood

Airflow,     Concentration   Hood Index (HI),
L/s (cfm)    Outside Hood          log
             ([C.sub.o]),      ([C.sub.o]/
                  ppb           [C.sub.i])

240 (500)        8917             1.397
380 (800)        1460             2.183
570 (1200)        316             2.848

Figure 7 HI (-log ([C.sup.o]/[C.sup.i]))/volumetric airflow.

Air Flow-Liters/Sec   Hood Index-Log(Co/Ci)

240                   1.397
380                   2.183
570                   2.848

Note: Table made from line graph.
COPYRIGHT 2017 American Society of Heating, Refrigerating, and Air-Conditioning Engineers, Inc.
No portion of this article can be reproduced without the express written permission from the copyright holder.
Copyright 2017 Gale, Cengage Learning. All rights reserved.

Article Details
Printer friendly Cite/link Email Feedback
Author:Meisenzahl, Lawrence R.
Publication:ASHRAE Transactions
Date:Jan 1, 2017
Words:4849
Previous Article:Energy analysis of leed silver certified dining hall on an academic campus--a revisit three years after its initial certification.
Next Article:Evaluation of alternative refrigerants for mini-split air conditioners.
Topics:

Terms of use | Privacy policy | Copyright © 2022 Farlex, Inc. | Feedback | For webmasters |