US2368143A - Ventilator - Google Patents

Description

F. R. KELLEY Jam.. 30, 1945.

VENTILATOR Filed March 2l 1941 am R Patented Jan. 30, 1945 PATENT OFFICE 368,143 vNtIrn'ron y Keller/Lambeth, 1:4! Ailiii fion Magica (2 1, 1941, serial No. 384,443 1 claiiif (c1. sas-7n) .Mvfinvootionrolatos yeotlatorslpartioularlv toexhaustventilators offthe sort illustrated and described inlletters Patent No. 1,639,187, granted tome-August A16192'7, and consists in certain refinements.aodolaborationfin-SttuotorelDvi-vir- `W i the obioois rot.. mum elciency is obtained' evolitotor ,cons ut .A s .ao.exe,mp1ary,.em ment0the--invention-is illustrated in the accompanyme drawing, inv/Ilich.:. .f fe: :Lr-1f ,Fig.,1,i`s al view or theventilator, partly in side elevation; and partly in 4 vortioa1.-seotion;- t if., Fie, 111s ,a View infrontolevatioo of.: a f sertanraa. angefthat is arrangedperpherally ofthe mouth or discharge opening Qftheventilator; 4 ma.; `Fig. III is achart that ,indicates comparative efciencies atlowwindvelocities. In general, the construction of my improved ventilatorl compares with, ,the `,ventilator of ,my above-noted patent, to which reference may be made for such features of structurel asr dovnot immediately concern my present linvention. Briefly, the Ventilatorfincludesga(,tubular base 3 that in service is rigidly assembled over, or upon,

the upper end o f the duct or passage (not-s l1,own that opens-through the roof ordeck of the building,for ship,` or other -enclosureto beventilated,h Upon such base a curved cowl 4 is mounftedfoijUv rotation on^a yvertical axis. ,`-1n this case, 4'them mounting consists in aV spindle 6, rigid withthe' cowl, and abearing block` I that, borneby*the*,r base 3vertica1ly sustains them pointell.` end Aof said spindle. Lateral support for the spindle provided by means of an anti-friction bearingr 12, j which bearing 12 is -rigidly secured v to thetbasle 3,, by means of a1 lspider '2*0. thusumounted, the# @CWI 1S adapted froolv 'to rotatofunoogthe ence'of the air currentsrorwind n the outeruat- ,A moophore in 11.11.011 the ventilator.. ism servico Situated o ,vane 9, ,beine Jinstron.Mental in man1.- fesi @1,101 known, Wav ,to maintain .the oowlwroeo with its dishargoopoios otmoothl position. downwind- With 'th cowl ,maintained in `suoli position. theel?, @wenn .Slamvrfthe oo` and around the. bodrio? tooloomoo@ Ate aspiraties effect t the mouth .oflthecowl- Suche aspirating effect, augmented bythefbuoyancy :or stack eflectof the airl inthe ducff-uponlwhichm they Ventilator is mounted, produces ,the desired-Y outward'flow of air from the enclosure being ventilated.. Y. I An outwardly-Sharing "lflange` encompassing the discharge mouth 1, increases this aspirating "u effect.. As-described in my. SaidT prior patenten i, ejector tubo 5....op'orates furthertoinorease the., air-*exhausting capacity of ,theA ventilator. The ejector tube is of frusto-conical form; andjt pro.- jects through, andis-rieidwith, the .posterior ,wall ofthe-cowtit is nolinedto the ver-ticahaxisof the cowl, with its smallersendirected;upward within the cowl and towards themouth 1.

It will be perceived thatthe hollow-.base 3 of the structure is of substantially frusto-conical form, whereby the venacontraota @effective and.-4 th? flow is maintained as nearly streamline as maybe-sx 5; if twin". 2*" illi.

.Such isthe, character ofythefventilator. ,for-i. which 4tho,.ironrovelntntsf.-of.my present inventionrhave:'beengpanticularly ,desigrlvla '2:1'5' f.; fa ffy: Bofill@V 4roooodins witloa:desoriptionA of Suo imnroxomentseitmav bozwoll to i.o.onshierzo.ortazi scientinc data pertaining to natural,.(as tiatin-,l. y guished from forced or mechanical) ventilation. Natural ventilationrrdepends-upon two factorsstack effect, and kinetic energy of wind. Stack eflecttislthe `buoyancy force .that tends toemoveh a column of air upward in a stack,-the-airin the stack being at highertemperaturethan the surrounding' atmosphere. j Stack `elect is substantially independentA of Vthe action, of wind in .the outer atmosphere,eand does not exist' when the temperaturesinside and outside of the stack areequal. l An explanation' of the eects of the kinetic energy of wind prevailingfin the outer atmosphere is not so simple. yWind action follows different laws at different-velocities. ,v

In the flow of fluids in pipes, ducts, etc.,"it` has been found that there ,areftwo critical Velocities. Below the lowercritical velocity the fluid moves in parallel lines and pressuredrop is due towiscosity only. t Above-theiupper-critical-veloc-f" ity, the -ud moves rin aturbulen-t manner; fand; pressurefdropgrowsas the squarefofthe velocity.`- Between the upper andY lowerJ criticalfvelocities the: motionofthe fluid fis unstable Land fluctuaties between one or the other states of-motion. Thesei sameI circumstances apply to the movement of air in aventilator. f

The suction oraspirating effectf the ventilator depends upon the impact pressure of the wind upon the ventilator cowl and upon the viscous., drag of the air, and it would appear that the force tending to .disohargeraireirom the ventilator is equal to thelsum` of stack effect and suction.` And` on, such.V assumption, itisypossible .to

derive, a pseudo-rationalormula.for the amountllf. of air discharged through a ventilator mounted P1=H feetX (density of air in lbs. per cubic foot at to) X titu ifi-460 If P1 is expressed in inches of water and to is taken at 70 F., then degrees F.

H (tf-ta) P1- 36,250 where 36,250 is the ratio of the Weight of 144 cubic inches of water to the weight of 1 cubic foot of air, multiplied by 530, which is equal to tu plus 460.

The force produced by the impact of the wind on the building and transmitted through the inlet opening to the interior is wind velocity)2 density 2g where g is the acceleration due to gravity or 32.2 feet per second per second; the wind velocity is expressed in feet per second; the density of the air is expressed in pounds per cubic foot, and the force P2 in pounds per square foot of the area on which the wind is effective. In order that P2 may be expressed in inches of water and the velocity of the wind expressed in the equation in miles per hour, the velocity of wind in miles per hour is multiplied by 5280 (=feet in mile) 3600 (=seconds in hour) and the density of air (.0756 pound per cubic foot) is multiplied by l2 (=inches in foot) 62.4 (=wt. o f water per cu. ft.)

or 1/s.2. Thus, the equation becomes:

. 2802 I U P l in miles per hour 3600 X .0756 V2 P :klX V"X density 2 thus becomes d 't P2=r.5 V2

From the calculations above it will be seen that when the velocity of the Wind is expressed in Cil miles per hour, and Pz in inches of Water, V2 in the equation is multiplied by 1/zoalt Thus,

l V2 P2l=OXV2X206c-4120 hence the total dynamic force would be I Vi P2+P2 1370 In accordance with well-known formulae the velocity of air in feet per second equals \/2g.h, where'h is the head of air, or the differential pressure between the upper and lower ends of the air column divided by the density of the air;

`explained as follows:

that is 2g differential pressure in pounds per sq. ft. V:

density of air 1n pounds per cu. ft.

The velocity of the air under the three pressures P1, Pz and Pz may thus be expressed:

2g(P1+P2+P2/)5.2 V (m ft' per S@CJ-density in pounds per cu. ft.-

A vena contracta is formed when the air enters the base of the ventilator, hence the quantity lof air discharged is less than the product of velocity and area. There is, furthermore, some additional outlet resistance in the ventilator. The overall coeicient of discharge depends on the location of the ventilator, whether located on a ilat roof or a ridge; also on the form of the ventilator base, whether flaring or cylindrical, and on the freedom of exit of air. A high value for this coeicient is 0.69, which is obtainable only in a ventilator with a practically unrestricted outlet. In this case the discharge Q in cubic feet of air per hour would be Where A=the area of the ventilator throat in square inches; or

'I'hrough careful tests, however, it has been found that these assumptions are incorrect, and that the formula must be empirically modied. The fact is that the two forces of suction and a `stack effect do not add directly, for reasons with which this invention is not immediately concerned. Indeed, it appears that in combining these two forces there is introduced a resistance (a negative force) that makes the resultant force less than the sum of the two, and oddly enough at very low wind velocities there is actually a slight reduction in the discharge, below the point of discharge produced by temperature diierence alone. This action, being the result of natural forces, has a tendency to take place in all ventilators. The less efiioient the ventilator the greater is this reduction, and the greater is the wind velocity required to overcome it. The physical reason for this may be At low wind velocities there is a partial blanking-off of flow, due to the convergent streaming of Wind currents around the outlet opening of the ventilator. This holds true for ventilators of both the rotary and stationary types. This blanking-oi has its greatest effect in very low wind velocities. In the velocity range between one and four miles per hour, depending on the ventilator type, the critical velocity, known as the Osborne Reynolds velocity, is reached. This is the velocity at creased-while-'theVblanking-oif effect is much reduced, or may even disappear altogether.

Whereas the flange 8 and the ejector tube 5 of my patented structure have done much to nullify this blanking-oif effect at low wind velocities, the particular improvements of my present invention carry the nullication to the optimum. Specifically, I form on the upper limb of the eccentric, outwardly flaring flange 8 a lobe or flange .extension 8a. Such lobe provides immediately above the top of the cowl, in position directly over the discharge mouth 1, a localized accentuation of the eiect of the ange in the currents of air that stream externally from the back to the front of the cowl. Additionally, I flare outwardly the outer and larger end of the ejector tube 5, the effect being obtained by means of an outwardly flaring peripheral flange a. By virtue of these refinements practical perfection is realized.

The chart in Fig. III aiiords a ready appreciation of the situation. 'Ihe abscissa represents velocity in miles per hour, and the ordinate the volume of air discharged per unit of time. The curve Z was plotted with data obtained in tests, under typical conditions in the field, on a conventional stationary ventilator, and the curve Y with the data obtained from a conventional rotary ventilator. It will be noted that both of these curves show e, marked drop in discharge with Wind velocities of between 0 and from 3 to 5 miles per hour, thus indicating that in the usual types of 'ventilators the negative effect produced by the external wind actually causes a drop in discharge below that which would be obtained with no Wind at all. The X curve shows the results with the improved ventilator of i my present invention. Significantly, there is no sag in the curve, and for all wind velocities of from 1 mile per hour upward a much greater discharge is obtained. At velocities above 12 miles per hour there may be ventilators in the art that will yield greater discharge than mine, but whether or not this is the case is of no controlling importance, because there never has been any problem in designing ventilators that are adequate in high wind velocities.

Since the blanking-oif effect, if it occurs at al1 in my ventilator, is negligible, I have been able to arrive at a reliable equation for discharge capacity. The equation is:

where the symbols represent the variables already dened. It is evident that in very few cases will the air inlets be always on the Windward side of the structure being ventilated. Usually there are openings on the opposite side also and in this case the pressure P2 Will be only a fraction of the impact pressure corresponding to the wind velocity. The inlet openings may also be shielded from the wind -action by surrounding obstacles. Again, these openings are of finite size and some resistance therefore exists to the flow of air through them, slightly reducing the net or effective air-moving force. For the average conditions encountered in actual practice, with inlet openings of reasonable size,

sideable margin for'the 4'factorof safety;j y

Obviously, 'the application of this formulaj`-` sl'ioulcl"{nt'beextemied took farl beyond the Irange of condi-tions -forwhich it wasfdevelope'd. e This range?include` heights yup :fto iiu5lefeet, fteiifiper" ature differences Iup 5to* 25" F.V candfwind velocities up to 15 miles per hour. On account of the regular form of the discharge curves resulting from the formula, it is believed, however, that there will occur no appreciable error in its use for extensions of as much as 50 per cent of these limits. This is of importance in some cases because of advantages to be gained through an elevated position of the ventilator, to make vuse of the greater wind velocities which prevail at higher levels.

In order that my improvements in ventilator structure may be fully understood and may be readily applied to ventilators of different sizes, I shall express the principaldimensions in percentage, with the diameter of the mouth ofthe cowl as the unit. The diameter of the flange 8 is substantially 122.9% the diameter of the mouth 1; the interval at which the center of the ange is spaced above the center of the mouth 'l is substantially 8.1%, and the extent of the lobe 8u, above the crest of flange 8 is substantially 11.6% of the diameter of mouth 1. The outward flare of the ange and lobe will be understood by reference to the angle a between surface of such parts and a vertical line-through the center of the mouth of the cowl, the value of the angle a being substantially 52. The overall length of the ejector tube 5 is substantially 53.5% of the diameter of the cowl mouth, while the length of the ejector tube without the flange is substantially 44.2%. The diameter of the inner end of the ejector tube is substantially 28%; the diameter of the tube at the region of junction with the flange 5a is substantially 44.2%, and the diameter at the outer end of the ange is substantially 58.2%, of the diameter of the mouth of the cowl.

The lobe 8a is symmetrical with respect to the vertical center line of cowl mouth 1, and, extending through substantially 28% of the periphery of the flange 8, the lobe is at its edge curved to merge with the curved edge of flange 8.

As above indicated the ventilator of the invention has been designed with eiiiciency at low wind velocities given foremost consideration. While the laws of aerodynamics have been given precedence over artistic merit, it will be perceived that the latter has not been sacrificed. This is a matter of practical importance, inasmuch as the ventilator in service will be exposed to view on the roofs of buildings, or above the decks of ships, or in other locations where architectural`merit is required. Within the ambit of the appended claim certain modifications are permissible.

I claim as my invention:

In a ventilator comprising a base, a cowl mounted on said base for rotation on a vertical axis, the mouth of said cowl extending in an approximately vertical plane, an upwardly flaring flange extending peripherally of and arranged eccentrically of said mouth, the breadth of such eccentric ange being maximum at the top of said mouth, and an ejector tube opening through the wall portion of the cowl opposite to said mouth; the improvements herein described that aiord increased ventilator efliciency at wind velocities of six miles per hour and less.

said improvements comprising an outwardly flaring peripheral flange on the outer end of said ejector tube, and a flange portion or lobe integral with said eccentric flange, said flange portion or lobe being centered above the mouth of said cowl and flaring upwardly and outwardly from said eccentric flange, said flange portion or lobe being of maximum breadth on a vertical plane that bisects said mouth of the cowl and of gradually decreasing breadth on opposite sides of said plane and merging with the periphery of said y eccentric flange at points spaced substantial intervals from said point of maximum breadth.

FRANK R. KELLEY.

Images