Planning for fume hoods in the design of science buildings

msny architects are not sufficiently famil- iar with fume hood systems to keep the special requirements of such systems in mind at the time schematic ...
2 downloads 0 Views 3MB Size
XXXVII. Planning for Fume Hoods in the Design of Science Buildings* HAROLD HOROWITZ, Architectural Services Staff, Notional Science Foundation, Washington, D. C. 20550 I n the review of science building designs we continue to see repetition of design errors that aught to be recognized and avoided. It has become obvious that an important reason for this is that many sspects of the design are established before the fume hood system is first considered and commitments are mrtde that force unhappy compromises. Unfortunately, msny architects are not sufficiently familiar with fume hood systems to keep the special requirements of such systems in mind a t the time schematic studies are under way. The fume hood system msy not be thought about until the stage st which lahoratory furnishings are being laid out in the individual lahoratory rooms or the stage a t which the consulting mechanical engineer begins his work. By this time, many decisions have been made which may seriously limit the possibility of obtaining a satisfactory fume hood system. The desirability of not locating fume hoods near doors or along principal traffic lmes in lahoratory rooms, and the desirability af keeping horizontal duct runs to a minimum need not necessarily be in conilict with each other if the locations of service shafts are considered a t the very beginning of the schematic planning sctivitv. I n a lahoratory building with msny fume hoods, the distribution of senrice shaftsis as fundamental as column spacing, and needs to he considered the moment the the first work begins in connection with partition arrangements. If it is recognized a t the beginning that i t is not desirable to have fume hoods located close to

* Presented to the Pacific Coast Physical Plant Administrators of Colleges and Universities, Nov. 15, 1966. -Ermtum The 1966 Threshold Limit Value for mercury recommended by the American Conference of Governmental Industrial Hygienists is 0.1 mg/ma, and the 1966 Threshold Limit Value recommended for organic compounds of mercury is 0.01 mg/ma. "Threshold Limit Values for 1966" in THIS JOURNAL 44, A 45 (1967) and in the reprint volume "Safety in the Chemioal Litbaratory" (page 32) give an incorrect value for mercury and no value for organic mercury compounds because of an oversight by the editor of this column.

doors, i t follows that the service chases are best located away from the probable location of doors. Many good laboratory plans have been developed which achieve both objectives without compromise. I t is possible, but rerely s s an afterthought. Similarly, fume haod outlets on the roof are there to provide for the dispersion of toxic and odoriferous materials. They must perform this function without contaminating sir intakes or endangering personnel who must work on the roof. Our experience has shown that these matters cannot be solved sa,tisfactorily by the mechanical engineer alone, since the solution to this set of problems must affect the appearance of the building; if this is not recognised a t the outset of schematic stage planning, it is probable that the design solution will not lend itself to satisfactory fume dispersion. I t is not usual for a designer to concern himself a t the early design stages of most types of buildings with such matters as sir intake locations and exhaust locations. However, i t is necessary to be concerned with location of air intakes and exhaust outlets when designing science laboratories to avoid the necessity of subsequently having to compromise satisfactory mechanical performance with aesthetics.

Auxiliary Air Fume Hoods with Supplementary Air Introduced Within theHoods Should Not Be Used The matter of auxiliary air fume hoods was discussed briefly in an earlier article,' but only one of the two types of auxiliary air fume hoods was mentioned. We did not mention the design in which the auxiliary air is introduced within the hood enclosure itself. This type of hood is svsilable fmm several manufacturers and we have also seen installations which were improvised in the physical plant department shop. Our reason for not mentioning this type of hood was, in retrospect, s. foolish one. We feel that this type of hood is inherently unsafe and we were reluctant to call special attention to it since we were making disparaging remarks about auxiliary air hoods in general. Our failore to mention this type of hood led to several situations reported to us, where salesmen

"Fume Hoods for Science Labmalwies," by Harold Ilorowite, S. A. IIeider and C. N. Dugan, American Institute of Axhiteels Journal, Julv. 1965. Remints available on request fidm the authon

Harold Horowitz is Superviaary Architect,, Architectural Services Staff, Instil~ltiunal Relations, National Science Foundation. He established the Foundation's Architectural Services Stan and developed its current progrhms among which are the evaluation of architectural aspects of proposals for grants to support the construction or renovation of college and university science facilities; review of progress and satisfactory completion of grants; consultation on science facility design to srehitects; promotion of the develo~mentof new and improved design criteria, and planning data for science facilities. Mr. Horowite received the B.A. in Architecture from the Institute of Design, Illinois Institute of Technology, and Msster in Architecture from MIT. He is a member of the American Institute of Architects, National Associat,ion of Phvsical Plant Administrators of Colfeges and Universities, and is a consultant to the Facilities Panel, Commission on Undergraduate Education in the Biological Sciences. He has published many papers and reports and served an various committees dealing with building design in relation to safetp. used our article's reference to the discomfort associated with the auxiliary air hood with sir supplied externally, as a means of promoting the sale of the hood with t,he auxiliary sir supplied internally. After having heard such reports, we realize it would have been better t,o explain that this type of hood is the most objectionable of all for a number of specific reasons. First of all, supplying auxiliary air within the hood reduces the volume of air drawn through the haod face to the point where there is insufficient air entering the hood to provide adequate capture velocity. Seeondly, the haod with internally supplied air ~.equiresextremely careful balancing

Volume 44, Number 5, Moy 1967

/

A439

Safety

...

and maintenance. The supply air can never be permitted to approach too closely the amount of exhaust air because fumes will then be forced into the laboratory. Maintenance is a matter for real concern with this type of hood since t,he exhallst fan may deteriorate and lose efficiency at, a much greater rate than the s ~ p p l yfan, since the exhaust fan operates ill a c o r w sivc environment. I n addition, hoods with internally supplied auxiliary air most have a aystem of interlocking airflow-sensing and control devices which alltomatieally shut off the mxilisry supply air fan in the event something happens to slow or stop the exhaust fan. I n our view, aaziliary air hworls i n which the a w i l i a r y w p p l y i s introduced within the hood enclosure should nevcr be employed under any eircuncstances. The disadvantages are too great,!

The Importance of Determining Wind Direction Distribution

One of the most important kinds of information when planning a science building with fame hoods is the pattern of wind dimct,ion distribution a t the site, somebirnes referred to as the "wind rose pattern" because of its general shape. Nost people have general notions of the prevailing wind direction, however, such general notions, usually expressed in such terms a s U O u r prevailing winds are from the west,, except in the winter when we have strong winds from the southeast"-are generally useless. Whst is needed is an indication of the prevailing percentage distribution of occurrences of wind from the various points of the compass. Even more important than knowing the direetion the wind is coming from most. oiben, is knowledge of the direction the wind comes from least often. It is the prevailing leeward zone of a science building that is the crit,ical area wiLh respect to the location of air intakes. If one examines a number of wind rose patterns for different sites one will see that it is rare that the prevailing leeward direction is 180" opposed to the prevailing windward direction, the popular conception. The quality of wind direction distribution information available a t the mesent time is a great weakness in rational design of fume disnelsal svstems and location of

unit or research group that has been collecting wind distribution data for a period of years, one may have to fall back on the data available from the nearest Weather Bureau Station. Wind direction records from the Weather Bureau are far better than no information. but cannot be resoont h s t an slert university, with an active research program in the sciences requiring construction and renovation of a substmtial amount of facility space over the years, would make an effort to collect such data. using recording equipment Located in the campus areas where science facilities are concentrated. Such information could

A440

/

Journol of Chemical Edumtion

become very valuable hase data in science facility planning.

Keeping the Rain Out of Fume Hood Fans The function of the discharge terminal of a fume hood duet system is to project fumes sway from the building in such a way that air intakes will not be contaminated and persons in the vicinity of the building or working on the roof will not be sub,iected to the hazard of breathing the discharged fumes. One of the most eommnn observations is that concern about rain entering the discharge end of the duct has led to a. design which will not project fumes away. We have seen discharge ouG lets of exhaust fans aimed downward to spread fumes along the surface of the roof or to discharge horizontally which is nearly as bad. The fans may be positioned to discharge vertically but the opening of the duet terminal is covered by some type of weather cap which reduces velocity and changes the direction of discharge so that the fumes are not projected away from the building. Such concern aver rain penetration ia needless. Very simple means of protecting the fan can be found that do not compromise the effectivenessof the system. To begin with, it should be remembered that an appropriate terminal velocity of a t least 2000-3000 fpm is usually sufficient in itself to prevent rain penetration of the stack. The column of air leaving the terminal simply brushes the rain aside. If the fan is not operated continuously, there is the possibility of some rain penetration but this can easily be handled by a small drain hole a t the bottom of the fan h o u s ing, or the use of one of a. variety of details that have the effect of creating a drip within the duct to divert water and permit it to be discharged before entering the fan housing.

Vertical Separation Between Fume Hood Outlets And Air Intakes Does Not Provide Protection For Air Intakes One of the most popular misconceptions is that a n air intake a t ground level will be adequately protected by a separation of several stories between it and fume hood discharge terminals on the roof. This conception is contrary to our nnderstsnding of airflow behavior around buildings and is also not supported by observations msde a t existing science buildings. One of the best examples we know is a university bioehemistry research laboratory that is five stories high. The air intakes are located a t ground level and fume hood outlets and a n incinerator outlet are located on the rooftop. This building is a long slab with its major axisoriented in the northeastjouthwest direction. The air intake is located near the center of the southeast elevation which turns out to be the prevailing leeward zone. The experience supports theoretical considerations completely, since strong solvent and chemical odors are distributed throughout the huilding hy the ventilating system with daily regularity. Occupants of this building have told us that they are aware of the beginning of the operation of the incinerator within min-

(Continued on page A44d)

Safety

...

utes, because the discharge from the flue is picked up a t the air intake and distributed throughout the building. The five story separation here has not provided protection.

Discharge the Fumes Above the Boundary of the Leeward Cavity Studies of air flow around huildings both in the field and in wind tunnels show that a buildina forms an obstruction to the air-

movement takes place between the cavity and the surrounding regions. The shape of the cavity is influenced by the shape of the huilding and other nearby obstructions to airflow, the velocity of airflow, and other meteorological conditions. We have already pointed out that air intakes should not be located in positions that are likely to fall within the prevailing leeward side of a. science building. However, this advice is really dependent upon the point a t which the fumes are discharged from exhaust stacks. If the fumes are discharged a t a height above the houndary of the cavity zone, they will he carried away from the building and not contained within the cavity. However, if the height is not sufficient for the fumes to he disehargedabove the boundary they may he contained and held within the cavity. If the air intakes

A442

/

Journal of Chemical Education

w e also within the cavity region, they will draw in contaminated air. There is no simple way to tell how high stacks should he to penetrate the cavity boundary. Photographs made of smoke moving past actual huildings, and simulation studies in wind tmnels show that the separation of the cavity from the air stream moving past may he sharply defined. However, the shape of the cavity and the height of the boundary for a given budding shape will vary with wind speed and he influenced by such meteorological conditions as wind gusts. There are a. number of papers in the literature describing studies in which wind tunnels tests have been used for making a. good approximation of the optimum height to insure dispersing fumes away from the vicinity of the point of discharge. Some authorities have formulated a rule of thumh which says that the height of the discharge point should be half again the height of the building. By building height they are r e ferring to the elevation of the leading edge or separation point a t which the cavlty begins to form. We are reluctant to support the rule of thumb that exhaust discharge points should be half again the height of the huilding beceuse we believe it is too oonservative for most situations. I t seems to us that it is increasingly conservative as huildings become taller and is also substantially infiuenced by the aerodynamics of the building shape and the relationship of the shape to the direction of the prevatling wind. Some smoke flow studies of low huildings, however, make the rule of thumb iLppear reasonably valid.

A common mistake in the design of science huildings is the attempt to hide fans, stacks, and other equipment on roofs of science huildings by means of parapets, wells, and other types of rooftop construotion in the hope of giving the building a neater, more orderly appearance. Such obstructions to aidow simply aggravate the situation by making it necessary to raise the height of the stacks still further. If stack heights are not raised, fume dispersal from the rooftop area will he seriously compromised and can lead to potentially hazardous conditions to all persons who must work on the roof while fume hoods are in operation, since the parapets and wells tend to retain fumes to even a greater extent than if they had not heen constructed. There are a number of huildings a t universities where, in our opinion, a prudent physicill plant director would insist that his work crews wear gas masks if performing maintenance operations while the fume hood fans are in operation. Solutions to this problem can be found readily once it is recognized that stacks are a necessary part of a science building ventilation system and the aesthetic design must incorporate some provision for accommodating them. A number of huildings have already been designed in which a series of stacks are visible, undisguised, hut so treated and organized that their appearance is entirely acceptable as a component of the form. This approach makes much more sense than trying to cover them up with false roofs or parapets whieh defeat their own purpose by making it necessary to raise the stacks still higher.