US 3532271 A Abstract available in Claims available in Description (OCR text may contain errors) Oct. 6, 1970 F. F. POLNAUER SPRAY NOZZLES WITH SPIRAL FLOW FLUID 4 Sheets-Sheet 1 Fi led Feb. 23, 1967 FIG. Oct. 6, 1970 F. F. POLNAUER SPRAY NOZZLES WITH SPIRAL FLOW FLUID 4 Sheets-Sheet 2 Filed Feb. 25, 1967 R E U A N m 0 MP EF l NV FREDERICK Z ATTORNEYS Oct. 6, 1970 F. F. POLNAUER 3,532,271 SPRAY NOZZLES WITH SPIRAL FLOW FLUID Filed Feb. 23, 1967 4 Sheets-Sheet 5 INVENTOR FREDERICK F. POLNAUER BY M ATTORNEYS V SPRAY NOZZLES WITH SPIRAL FLOW FLUID Filed Feb. 23, 1967 4 Sheets-Sheet L I INIVEIQTOR 0 FREDERICK s POLNAUER ///i 4/ BY ATTORNEYS 3,532,271 SPRAY NOZZLES WITH SPIRAL FLOW FLUID Frederick F. Polnauer, 250 Riverside Drive, New York, N.Y. 10025 Filed Feb. 23, 1967, Ser. No. 618,172 Int. Cl. A26c 1/06; B]: 1/34 US. Cl. 239-1 19 Claims ABSTRACT OF THE DISCLOSURE This invention relates to spray nozzles and in particular to an improved nozzle for the distribution of fluids, such as liquids, gases and other sprayable materials, into a cone-shaped spray of very fine droplets that are discharged in a uniform pattern and wherein the pressure head of the fluid to be sprayed is efficiently and effectively converted into kinetic energy of rotation in a circulation chamber. BACKGROUND OF THE INVENTION Spray nozzles of the type using a logarithmic spiral flow for the fluid, are known in the art. In one such nozzle, which is described in British Pat. No. 760,972, it is claimed that optimal flow conditions caused by the formation of a logarithmic spiral flow, and thus maximum spray nozzle efficiency, can be obtained by controlling (or is substantially dependent upon) only two major nozzle dimensions. These dimensions are the inlet width and largest radius of the swirl chamber, that is, the chamber in which the logarithmic spiral fluid flow is obtained. The aforesaid British patent states that a ratio of inlet width to largest radius not larger than should be maintained in the swirl chamber. However, 1 have found that such a conclusion is not valid for most cases and instead, critical ratios among at least six parameters of the nozzle are of decisive importance in obtaining an effective nozzle. Further, I have found that the existing nozzles of this general type are lacking in many aspects of providing a good mechanical design wherein nozzle parameters may be readily changed by replacing nozzle components and at the same time maintaining certain predetermined relationships between the nozzle components such as alignment of the outlet orifice with the axis of the swirl chamber. SUMMARY OF THE INVENTION In accordance with my invention, an improved spray nozzle of the logarithmic spiral flow type is provided in which the outlet orifice opening is concentrically aligned with the axis of the swirl chamber at all times. Further, in accordance with a preferred embodiment of my invention, a spray nozzle is provided which has a replaceable outlet orifice plate and a replaceable swirl chamber so that the total configuration of both the outlet orifice plate as well as the swirl chamber may be varied and combined according to certain concepts which are an important part of this invention to attain maximum performance efficiency under a wide range of operational conditions. United States Patent 0 3,532,271 Patented Oct. 6, 1970 In addition, in accordance with the present invention, I have developed certain novel methods and certain design criteria for determining the parameters of logarithmic spiral flow spray nozzles which enable such nozzles to be designed with a considerable degree of predictability of spray performance, i.e., fluid flow rate and spray cone angle, rather than by the usual cut-andtry methods. Further, I have also developed certain novel methods and design criteria by maintaining ratios of nozzle parameters within predetermined ranges, which enable the patternation index of a logarithmic spiral flow, spray type nozzle to be described and predicted. A It is therefore an object of this invention to provide a spray nozzle wherein the fluid flow characteristics approach the theoretical flow-dynamics of a preferably logarithmic spiral resulting in an axial-symmetric flow and a uniformly distributed atomized fluid forming a conical spray and wherein the energy losses through friction and particle impact are minimized. A further object is to provide a spray nozzle in which the outlet orifice can be readily kept in alignment with the axis of the swirl chamber. Still another object is to provide novel methods for mathematically determining the parameters of logarithmic spiral flow type nozzles in order to achieve certain operational characteristics. An additional object is to provide novel methods and design criteria for use in constructing logarithmic spiral flow type spray nozzles in which the patternation index can be maintained below a given value. A further object is to provide novel methods and design criteria which permit the description and prediction of the patternation index of a logarithmic spiral flow spray type nozzle. Another object is to provide novel methods and design criteria used in the construction of logarithmic spiral fiow type nozzles in which the spray performance can be predicted. Other objects and advantages of the present invention will become more apparent upon reference to the following specification and annexed drawings, in which: FIG. 1 is a longitudinal cross-sectional view through the center of one embodiment of the spray nozzle constructed according to the principles of the invention; FIG. 2 is a top plan view of the swirl chamber body shown in FIG. 1, with a swirl chamber whose wall is shaped according to a logarithmic spiral; FIG. 3 is a longitudinal cross-sectional view taken along line 3--3 of FIG. 2; FIG. 4 is a cross-sectional view taken along line 44 of FIG. 2; FIG. 5 is a top plan view of the orifice plate; FIG. 6 is a cross-sectional view of the orifice plate taken through line 66 of FIG. 5'; FIG. 7 is a longtiudinal cross-sectional view through the center of a fuel injection spray nozzle made in accordance with the principles of the present invention; FIG. 8 is a cross-sectional view taken along line 88 of FIG. 7; FIG. 9 is a cross-sectional view taken along line 9-9 of FIG. 7; FIG. 10 is a top plan of the nozzle of FIG. 7; FIG. 11 is a bottom plan view of the nozzle of FIG. 7; FIG. 12 is a longtiudinal cross-sectional view through the center of a fuel injection nozzle for gas turbines made in accordance with the principles of this invention; FIG. 13 is a top plan view of the nozzle of FIG. 12; FIG. 14 is a longitudinal plan view of the nozzle of FIG. 12; FIG. 15 is a longitudinal sectional view through the center of another embodiment of the fuel injection nozzle 3 for gas turbines according to the principles of this invention; FIG. 16 is a longitudinal sectional view through the center of a further embodiment of a fuel injection nozzle for gas turbines, according to the principles of this invention; FIG. 17 is a sectional view of an improved embodiment of the orifice plate; and 'FIG. 18 is a sectional view of an improved embodiment of the swirl chamber. Referring to FIG. 1, a preferred embodiment of a spray nozzle constructed according to my invention comprises a housing 1 of stepped cylindrical shape with a male thread 5 formed along a portion of its largest outer diameter. The one open end at the top of the housing 1 has a bore 6 in which a swirl chamber body 2, having a bottom wall 12 and a chamber 9, and an orifice plate 3 are located. The lower end of housing 1 contains longitudinal inlet bore 7 concentric with the longitudinal axis AA of the body provided with a female thread 8 into which the liquid to be sprayed is admitted. The swirl chamber body 2 and orifice plate 3 are tightly fitted into the bore 6 to prevent leakage from the bore 6 of housing 1. As seen, both parts 2 and 3 are readily removed from the bore 6 for the purpose of replacing them. The orifice plate 3 has an orifice and the body 2 has a chamber 9 both of which are concentric with the longitudinal axis AA of the body. This concentric arrangement of orifice plate 3 and swirl chamber body 2 within the bore 6 of housing 1 is of great importance regarding the capability of mass-producing this type of nozzle with a high degree of precision and maintaining very small dimensional tolerances. Since the bore 6 serves as a gage, or master, for the peripheral diamenters of swirl chamber body 2 and orifice plate 3, this arrangement is important for making it possible that different replacement parts of 2 and 3 will always fit into the housing. Such replacements will be required whenever these parts have been worn off or damaged by usage or must be replaced to accommodate varying operational conditions. Swirl chamber body 2 and orifice plate 3 are firmly held in place within the housing bore 6 by a threaded cap 4 which presses down upon the upper surface of the orifice plate 3, and thereby also prevents leakage of the fluid from the body. A seal 11 may also be used between the body upper wall and the cap to prevent leakage. The inner wall of the cap 4 carries a female thread 5a to engage the male thread 5 of housing 1. After passing from a fluid supply conduit (not shown) through the inlet bore 7, the liquid flow enters an inlet 13 (see FIG. 2) which is tangential to the inner wall 14 of the swirl chamber 9. The fluid is circulated in chamber 9 along a generalized logarithmic spiral and is discharged through the outlet orifice 10 in the shape of a hollow cone 11 (FIG. 1). The axially symmetric thinwalled conical shell of discharged fluid at the outer edge of the outlet orifice 10 is torn apart into very fine droplets due to the effect of the centrifugal force of the circulating fluid. In FIG. 2, the swirl chamber body 2 is shown in a top plan view and the flow path of the fluid entering the tangetial inlet 13 is indicated by the arrow. According to FIG. 4, which represents a sectional view along line 4-4 of FIG. 2, the liquid passes upwardly through a cutout 12a in the bottom section 12 through inlet passage 13 and thereafter thinning inwardly above the bottom wall 12 of the swirl chamber into a horizontal position. FIG. 3 is a longitudinal sectional view along line 3-3 of FIG. 2. From FIGS. 2 and 3 may be seen four major dimensional parameters of this spray nozzle considered as important, according to my invention, which affect the etliciency and effectiveness of the performance of the spray nozzle. They are the height of the Swirl chamber H; largest radius of the swirl chamber R; width of the tangential inlet B close to the inlet opening into the swirl chamber; and the thickness of the rib S formed by the inner wall of the swirl chamber at the inlet. The inlet side wall 14 of the swirl chamber 9 is preferably shaped according to an outer turn of a true logarithmic spiral. It has been found, in general, that a logarithmic-spirally shaped swirl chamber is preferable in efiiciency to a circularly shaped chamber. However, there may be varying conditions under which it would sufiice to use variations to the logarithmic spiral, e.g. circular chambers. With regard to the number of inlets into the swirl chamber, it can be stated that one inlet is generally most preferable since it causes a minimum of clogging and inner fluid friction. FIG. 5 shows the orifice plate 3 in a top plan view and FIG. 6 shows a sectional view of FIG. 5. From FIG. 6 may be seen the two other major dimensional parameters, the diameter D of the outlet orifice 10 and the axial thickness L of the orifice plate near the outlet orifice. The cone angle 2 b of the hollow spray cone formed by the fine droplets is shown in FIG. 6. FIGS. 7-11 show another embodiment of my invention which is useful in applications such as fuel injection in oil burners or in any type of combustion chamber. The same reference numerals are used for similar parts, where applicable. This embodiment comprises a housing or nozzle body 16 of generally cylindrical shape with two sets of male threads 17 and 18 formed on two different portions of its outer diameter. One open end of body 16 has a concentric longitudinal bore 19 in which the swirl chamber body 2 and orifice plate 3 are located, as in FIG. 1. The lower end of housing 16 has a concentric longitudinal bore 20, of smaller diameter than bore 19, into which the fluid to be sprayed is admitted. The swirl chamber body 2 and orifice plate 3 are closely fitted into the bore 19 to prevent leakage from the bore 19 and the nozle body 16. Both parts 2 and 3 are easily removed from the bore 19 for replacement purpose. Swirl chamber body 2 and orifice plate 3 are firmly held in place within the bore 19 of the housing 16 by a threaded cap 21 which bears down under pressure upon the upper surface of orifice plate 3, and thereby also prevents leakage of the fluid. The inner wall of cap 21 has a female thread 17a engaging into the male thread 17 of nozzle body 16. As in FIG. 1, the swirl chamber and the outlet orifice are held concentric with the longitudinal axis of the nozzle by the inner wall of body 16 surrounding bore 19 which extends above the swirl chamber body and engages a portion of the orifice plate. The threads 18 are provided to engage corresponding threads of the combustion chamber, or any fitting to hold the nozzle body 16. The fitting or chamber reaches a stop against a shoulder 16a. FIGS. 12-14 show a further embodiment of my invention which has particular utility for fuel injection in combustion chambers of gas turbines. As seen best in FIG. 12, the spray nozzle of this embodiment includes a housing or nozzle body 22 of cylindrical shape with two sets of threads 23 and 24 formed on two difffierent portions of its outer diameter. One open end of nozzle body 22 has a longitudinal bore 25 into which the swirl chamber body 2 and orifice plate 3 are closely fitted. Again, the outlet orifice 10 and swirl chamber 9 are concentric with the longitudinal axis of the body and the bore. The opposite end of housing 2 contains a concentric longitudinal bore 26 of reduced diameter through which the fluid to be sprayed is admitted. Swirl chamber body 2 and orifice plate 3 are closely fitted into bore 25 to prevent leakage within bore 25 and nozzle body 22. Parts 2 and 3 can easily be removed from bore 25 if so required. Both parts 2 and 3 are firmly held in place within bore 25 by a threaded cap 27 which presses down on the upper surface of orifice plate 3 and thus also prevents leakage of the fluid. The inner wall of cap 27 carries a female thread 23a which engages the 1 corresponding male thread 23 of housing 22. The peripheral portion 28 of cap 27 is of circular configuration and its diameter can accommodate the inside diameter 29 of an air shroud member 30. The purpose of air shroud 30 is to coll the surface of the nozzle and to keep it free of harmful deposits. This is accomplished by introducing an air stream through several longitudinal channels 31 provided in the cylindrical outside portion of cap 27. The air flows up to the inner wall 33 of the top portion of the shroud and is there properly distributed over the face of the nozzle. The inside diameter 29 of shroud 30 fits tightly on the peripheral portion 28 of cap 27, and is thus firmly held in place. Several longitudinal cutouts 32 in the shroud equal in number and matching with the longitudinal channels 31 of cap 27 enable the air to enter the shroud. A lock ring 39a locks cap 27 and body 22 firmly together and thus prevents loosening of the connection provided by the threaded portion 23 of nozzle body 22. The male threaded portion 24 serves to fit the nozzle body 22 into a female threaded fitting contained in the combustor wall, or in a main manifold distributor if the combustion chamber has more than one nozzle. A cylindrical filter 34 of the cartridge type is placed within the bore 26 at the lower end of housing 22 to filter the fuel. The filter screen 34 is brazed to a collar 35 whose peripheral circular diameter 36 can be accommodated within the inside diameter 37 of the cylindrical bore 38 formed at the fuel inlet portion of the nozzle. A retaining snap ring 39 holds collar 35 of filter 34 firmly in place. FIG. 15 shows another embodmient of my invention which may also be used for fuel injection in the combustion chambers of gas turbines. In FIG. 15 a swirl chamber body 40 of cylindrical shape is shown with two sets of male threads 41 and 42 formed on two different portions of its outer diameter. The closed end of body 40 is integrally formed with the spiral swirl chamber 43. The outside diameter 44 of swirl chamber body 40 is closely fitted into a matching inner bore of a cap 45, which is the orifice body 45 and which has formed at one end thereof the outlet orifice 10. The inner wall of cap 45 carries a female thread 41a for mating with the male thread 41 of housing 40. These threads 41 and 41a mate on the body below the swirl chamber to preserve the axial alignment. The threaded cap 45 bears down on the upper surface of housing 40 and thus prevents leakage of the fluid. Again, the axis of the swirl chamber and of the outlet orifice are held in line with the longitudinal axis of the body. However, the gage action accomplished by the wall 6 of the main body bore in FIG. 1, is lost. The peripheral portion 46 of cap 45 is of circular condiameter 47 of an air shroud 48. Several longitudinal cutouts 48:: are provided on the shroud 48 and matching longitudinal channels 49 in the cap 45, so that the air can flow to the inner wall 50 of shroud 48 and is there properly distributed over the face of the nozzle. The inside diameter 47 of shroud 48 fits tightly on the peripheral portion 46 of cap 45. To prevent loosening of the nozzle assembly during operation, a fixed joint 51 is provided to secure the assembly firmly. The male threaded portion 42 serves to fit swirl chamber body 40 into a female threaded fitting of the combustor wall or of a manifold fuel distributor. The strainer assembly 52 is similar to the one shown in FIG. 12. FIG. 16 shows a further embodiment of my invention to be preferably used for fuel injection in gas turbines. This embodiment differs from the embodiment shown in FIGS. 14 and 15 in several respects. In FIG. 16 the nozzle body 53 is of cylindrical shape with two male threaded portions 54 and 55 along two different portions of its outer diameter. Adjacent the left open end of body 53 (as shown in the drawing) are located the swirl chamber body 2 and orifice plate 3. Body 2 and orifice plate 3 are closely fitted into a bore 56 of a threaded cap 57 which holds parts 2 and 3 firmly in place within bore 56 and aligned with the bore axis by pressing down on the upper surface of orifice plate 3, thus preventing leakage of the fluid. The inner wall of cap 57 carries a female thread 54a to engage the corresponding male thread 54 on body 53. Parts 2 and 3 can easily be removed from bore 56 if so required. Similar to the nozzle shown in FIG. 12, the embodiment of FIG. 16 also contains an air shroud 30, a lock ring 39a and a strainer assembly 52. FIG. 17 shows a sectional view of an improved embodiment of the orifice plate 3. Here, improved orifice plate 58 has an upwardly slightly slanting inner wall 59 as shown in the drawing. This has been found advantageous in smoothly guiding the horizontal streamlines into a vertical direction at the outlet orifice and thus minimizing fluid friction. FIG. 18 shows a sectional view of an improved swirl chamber useful with the nozzles of the present invention. Here, the swirl chamber body 60 contains a swirl chamber 61 with its bottom wall carrying a concentric conical ridge 62. This ridge has been found useful in smoothly guiding the streamlines upward toward the orifice hole and thus minimizing inner fluid friction. In all of the foregoing nozzle designs, it should be noted that the concentricity of the orifice or orifice cover plate is independent from the concentricity (or lack of it) of the threaded portion of the cap which holds the orifice cover plate down or which forms the orifice. This is highly advantageous. Further, in those embodiments where a cap is used to hold the cover plate down, in assembling the nozzle and threading the cap on the housing, no twisting or tilting stress is exerted on either the cover plate and/or the swirl chamber body which could otherwise lead to a distortion of both and causes eccentricity of the axis or deflection of it. From the viewpoint of thermal stresses the nozzle design of my invention is also advantageous for several reasons. For example, in a typical application the temperature of the oil in a swirl chamber is F, while outside the nozzle in the combustion chamber it may be 1000 F. and at the area of circulating air on the outer periphery of the nozzle about 700 F. Therefore, temperature gradients are set up which may lead to buckling or warping of the spiral swirl chamber body insert, and particularly the cover plate. By letting the cover plate protrude above the upper rim of the housing body, the orifice cover plate obtains more strength and resistance against buckling. Heretofore, the theoretical advantages which should be possible in spray nozzles using a circulation chamber wherein a fluid film is torn apart by centrifugal force at the outlet orifice of the circulation chamber, have not been achieved. From investigations in this area I have found that the failure to achieve optimal results with this type of spray nozzle has mainly been due to large frictional losses, lack of an axially-symmetrical flow and impact of the fluid particles within the swirl chamber and at its inlet and outlet passgaes. I have also found that these failures have been caused by a basic lack of understanding of the effect of the total nozzle geometry as represented by the six important geometric parameters, B, D, H, L, R and S, referred to above. Heretofore, nozzles of the swirl chamber type have been designed primarily by an empirical, or cut-and-try method with little regard to the inter-relationship of the aforementioned six geometric variables. Further, the prior art design methods did not permit any advance prediction as to efiicacy of the nozzle and its spray, for example, in terms of the cone angle and weight flow rate of the fluid. Further, the prior art nozzles are not able to establish certain nozzle design criteria by which a patternation 7 index of below a certain value can be achieved, or to predict the patternation index in general. The effect and use of the patternation index is described below. When designing nozzles of the spiral swirl chamber type, two criteria are usually specified by the purchaser for whom the design is being made. These are the cone angle (2 b) and the actual weight flow rate (W of the fluid. Starting with these two criteria, the following is a method that I have evolved which can be used for determining certain of the nozzles geometric parameters. As is known, the discharge coefiicient (K) and spray cone angle (Zr/1) of a nozzle are functions of the geometric parameters of the nozzle and the nozzle pressure drop. For a given nozzle having the following properties: the weight flow rate for an ideal fluid (W is given by: idea1 out V 'Y If we know the cofiicient of discharge (K) for a particular nozzle pressure drop (AP), we can write the actual weight flow rate (W as: act ideal It was found that the coefiicient of discharge at a given reference pressure drop (K is an empirically derived function (f of the nozzle area ratio A /A By defining the area ratio as ir. or then ( rer=f1( Further, it was found that the discharge coefficient K for any AP can be related to the discharge coefiicient at the reference pressure drop by a correction factor (C as follows: K p ref The correction factor C is a function (f of the actual nozzle pressure drop, p=fz( In a similar manner, a set of empirically derived functions relate the spray cone angle (Zip) for a particular actual pressure drop to the area ratio out Thus, the spray cone angle for the given reference pressure drop (2p is a function (f of the area ratio, as follows: To relate the spray cone angle 2x11 for any given pressure drop to the spray cone angle at the reference pressure drop (2% a correction factor f ref (degrees) f3 is used as follows: 2w mr The correction factor Cw is a function (f4) of the actual nozzle pressure drop, and is given by: 2=f4( To summarize, there are four basic empirically derived relationships to consider in defining the nozzle: ( ref f1( which states that the coefficient of discharge at a reference nozzle pressure drop of AP is a function (f of the area ratio. ( p=f2( which states that the correction factor used to adjust the discharge coefiicient K for pressure drops other than the reference pressure drop is a function (f2) Of the actual nozzle pressure drop. which states that the spray cone angle at a reference nozzle pressure drop is a function (f of the area ratio. (4) C2=f4( which states that the correction factor used to adjust the spray cone angle to pressure drops other than the reference pressure drop is a function (f of the actual nozzle pressure drop. A nozzle can readily be designed in accordance with Equations 14, once the various functions f through f are determined. For example, consider that a nozzle must be constructed that will give a given weight flow rate W and spray cone angle 231 at a given nozzle pressure drop AP. First step Using Equation 4, calculate 02 for the given nozzle pressure drop AP. Second step We know that 1 2 re =4 l r Cw from the definition of Since we have from the First Step and 2 1/ is given, We can solve Equation 3 for the area ratio out i.e., m 1 3 f out 02% Third step Using Equation 2, calculate C for the given nozzle pressure drop AP. Fourth step 9 We know from before that: aet= ideal and 1dea1= out V 'l it follows that: [Gm/ZyyAP where D is in ft., W in lbs./sec., 'y in lbs./ft. and AP in lbs./ft. Sixth step 'By using the ratios of the other nozzle dimensions, as defined by the requirements for good patternation which are discussed in detail below, we are able to determine the important nozzle dimensions other than D. In this procedure one must bear in mind that nozzle inlet area A as defined by B H must be kept at such a level so that the ratio A /A remains at a value which is predetermined by the Second step. It should be understood that by using the method described above, that empirical relationships of the nozzle passages (parameters B, D and H) are established for a given flow rate and cone angle. A typical procedure will now be described for deriving f f f and 1, which are used in Equations 1, 2, 3, and 4 above. Of course, any other suitable procedure may be used in accordance with conventionally accepted techniques. In this procedure a family of logarithmic spray type nozzles of similar design, but with different A (inlet area) and A (outlet areas), are operated over a wide range of nozzle pressure drops, for example from to 700 p.s.i. The spray cone angle (230) and the actual nozzle fiow rate (W are measured at specific values of pressure drops. From each of these measurements the value of the coeflicient of discharge for a given pressure drop can be computed as: measured flow rate W 'y=fuel weight density A =nozzle outlet orifice area=1rD /4 and g gravitational constant. In one specific run of the typical procedure being described, twenty-seven different nozzle combinations were used in determining K and 2 1/ at various AP points, with three different D, H and L parameters, while B, S and R were held constant. From the total of twenty-seven, nine nozzle combinations were used to determine K and 21/1 with D and H varying and the parameters L, B, S, and R being held constant. The twenty-seven nozzle combinations used were believed to be suflicient to provide the necessary data within the ordinarily accepted limits of experimental and design error. Of course, fewer or more combinations can be used as the accuracy requirement dictates. The data obtained in the run with the various nozzle combinations is plotted to give two families of curves. The first family represents K (ordinate) vs. A /A (abscissa) for the various nozzle combinations at given values of AP, each curve of the family being at one value of AP for nozzles of the number of available combinations. The second family represents the measured 2\// (ordinate) vs. A /A for the various nozzle combinations at given values of AP, each curve of the family being at one value of AP for the nozzles of the number of available combinations. To make the data of the curves more easily usable for general analytical design of the nozzles, the curves were translated into numerical equations to obtain 11, f f and L in the following manner: Step 1.The discharge coefficient K is obtained from the first family of curves as a function of the A /A ratio. (a) Since the discharge coeflicient K varies with nozzle pressure drop AP at a constant A /A value, a reference pressure drop is selected at which the K vs. A /A relationship is to be found. This gives K of Equation 1. In the experimental run being described, a 300 p.s.i. pressure drop was selected as K to obtain f since this value is about the center of the range of pressure drops tested. Of course, any other suitable constant pressure drop line can be selected. (b) The experimental data of K vs. A /A plotted at the 300 p.s.i. pressure drop is then curve fitted by any suitable manual or automatic computational method, for example the least square method. The latter comprises applying the method of least squares to fit polynominals of degrees 1, 2, 3 and 4, of the form ref=f1( in out) through the data sets of pairs of values. This resulted in K at p.s.i., Or K300. Step 2.To derive the correction factor C as a function 1 of the actual nozzle pressure drop AP in Equation 2, families of curves K vs. A /A with pressure drops varying from 15 to 700 p.s.i. are again plotted. The values of A /A ratios are from the same nozzle combinations used in Step 1 and are identical with the respective values of A /A used in Step 1. The data for each nozzle pressure drop AP is taken from the experiments with the various nozzle combinations. If K is the actual discharge coefficient for any AP, the correction factor is As explained above, both K and K are derived from the experimental data. As a second part of Step 2, all C values are calculated for each given value Ai /A corresponding to the respective A /A values of the curve K, =f(x). It was found that in the nozles of the present invention, the C values at any specific pressure drop do not vary significantly for different area ratios. Therefore, the average C data can be plotted against varying AP and a fitted curve produced by the least square method as used in Step 1. The fitted curve equation gives C =f (AP) Step 3.--To obtain the cone angle 2,0 as a function 73 of the A /A ratio in Equation 3, a reference pressure drop is selected since the cone angle varies with nozzle pressure drop AP at a constant A /A value. In the experimental run being described, the 300 p.s.i. reference pressure drop was again used. The experimentally ob tained data 2 0 vs. A /A plotted in the second family of curves is extracted for the selected reference pressure drop and a curve plotted. This curve is then curve fitted 'by the least square method to give 2 l/ =f (x) Step 4.To obtain the correction factor as a function 2, of the actual nozzle pressure drop in Equation 4, so as to be able to adjust the spray cone angle 25b to pressure drops other than the reference pressure drop, the following is done: (a) Families of curves 2\// vs. A /A are plotted with pressure drops varying from 15 to 700 p.s.i. The values of A /A are from the same nozzle combinations as used in Step 1 and identical with the respective values of A /A of Step 1. The data for each cone angle is experimentally measured. If 2 1/ is the actual cone angle for any AP, the correction factor is 2 Wit. 1 l (b) Next, all values are calculated for each given values A /A corresponding to the respective A /A values of the curve 2,.,,=f (x). It was found that the C211 values for the nozzles of the present invention at any specific pressure, do not vary significantly for different area ratios. Therefore, the averages data can be plotted against varying AP and a fitted curve developed by the least square method as used in Step 1. The fitted curve results in the equation As should be apparent, a complete method has been described for either obtaining several of the nozzle parameters, if 2 h and W are given, or for predicting 2 h and W when the A /A ratio and pressure drop as known. One of the most difficult requirements to be attained in swirl chamber spray nozzles using a circulation chamber is the atomization of a fluid into fine droplets and discharging the atomized fluid into a uniformly distributed conical spray. The latter property of the spray is of basic importance in many applications of spray nozzles, and particularly for fuel injection in combustors of oil burners, gas turbines, and other internal combustion engines. The degree of uniform weight distribution of the spray achieved by a spray nozzle can be measured by the socalled patternation index (delta) which represents a quantitative measure for the level of distribution. For a better understanding of the patternation index concept which is involved in my invention, it is useful to present a short definition of one of several ways by which a patternation index may be obtained experimentally. In one method of obtaining the patternation index, a total number of X observations are taken on the percentage of fluid sprayed into, for example, six equal 60 sectors arranged in a circular or hexagonal catch basin. The sum of the absolute values of the differences between 16 /3% and the percentage falling into each of the six sectors of the catch basin during each of the X test runs is used as the patternation index. In general, the lower the patternation index, the better the distribution of fluid by the nozzle in the conical pattern. In many critical applications, for example, a spray nozzle used for fuel injection in the combustion chamber, a nozzle having a large patternation index would cause uneven, damaging fuel concentrations resulting in hot spots on the combustion wall or adjacent components, e.g., turbine blading. In the calculation of the parameters of swirl chamber nozzles it is very important to the designer to quantitatively evaluate as part of his design procedure the effect of the various geometric parameters which he must select in part arbitrarily. In accordance with this invention the patternation index is used as an important and basic criterion for the evaluation of the design as related to the total nozzle geometry and also to the quality of the manufacturing technique used to produce this type of swirl chamber spray nozzle. The use of the patternation index serves for optimization of the design and for manufacture evaluation. The prior art of designing swirl chamber nozzles with a circulation chamber fails to obtain such aforementioned design criteria, and particularly one containing the patternation index. It is an important part of my invention to use the patternation index as well as ranges of ratios of the six major nozzle dimensions as aforementioned, in a unique and novel way in describing and predicting the efficiency of nozzle design in terms of performance in patternation. It has been found that the optimal performance of a swirl chamber type with one inlet and spray nozzle is significantly affected by the six geometric parameters or dimensions, namely, the orifice diameter D; axial thickness of the orifice plate L at the outlet orifice; width of the tangential inlet B close to the inlet opening into the swirl chamber; height H and largest radius R of the swirl chamber; and the thickness S of the rib formed by the inner wall of the swirl chamber at the inlet. Each of these six geometric parameters are important for accomplishing two novel and major features of this invention, namely the description and prediction of nozzle performance in terms of the patternation index delta for a given set of these six geometric parameters. It has been discovered that there is no single parameter whose value is critical in itself. Instead, for any given value of a single parameter within a given range of geometric variables, stated subsequently, we can obtain optimal nozzle performance in terms of good patternation by varying the other parameters within their stated ranges of limits. It was found that each of the following ten geometric ratios: constituting the geometric independent variables are the more important ones in determining the patternation index, but the value of any one of these does not fix the patternation performance. It also was found that nozzles having a patternation index delta less than 30, are generally acceptable for many spraying purposes. However, in applications as critical as fuel injection in combustors of gas turbines, the patternation index should be much less than 30. In general, it was found that the reliability or consistency of the paternation performance is very high for nozzles having a low patternation index. Since all the six independent variables or geometric parameters D, L, B, R, H, and S contained in the above stated ten ratios are important and since there is much interaction occurring among them, we cannot describe good nozzles by giving ranges on each variable separately. We can however, by defining regions for the above cited ten geometric ratios achieve two major features of this nozzle, the description and prediction of nozzle performance in terms of the patternation index delta. It should be noted that description and prediction of the performance in terms of the index delta are two separate features of this invention-and the conditions for each will be separately described. The ranges of the ten ratios specified in Groups I and II above are obtained in the manner hereafter described, or in any other suitable experimental and/or mathematical manner. The method to be described, however, has been found to be accurate to a relatively high degree within normally accepted limits of experimental error. As the first step, a number of experiments involving different nozzle combinations with varying combinations of parameters D, L, R, H, B and S are run and the patternation index delta is measured for each experiment, The delta measurement is carried out in any suitable way, for example, as described above. The data from each experiment is used for later computational purposes and it is conveniently utilized on an electronic computer, so it is, for example, placed on punched cards or tape. The second step is to find which of the six nozzle parameters are the variables affecting patternation and which combinations (thereof) (such as linear ratios, e.g., D/R; squares, exponential powers, etc.) must be investigated to determine their effect on the patternation behavior of the nozzle. As the first part of this step, the linear relationships of all of the combinations of six variables are investigated, for example by using a stepwise multiple regression program in a computer, to eliminate all insignificant variables. After the insignificant variables are eliminated a computer run is made to determines the most significant combinations (linear ratios, etc.) of parameters. Further regression analyses are then made to further eliminate non-essential combinations of parameters from this run. After all non-significant combinations of parameters are eliminated an equation is derived which is the basic patternation index delta equation. In the sample process being described, seven regression runs were made and the delta equation turned out to be: () (5) (PATTERNATION INDEX delta) L B S L B R -{Jr 5*!12 -+ga 54-94 5 4-95 m bidwhere g to g are factors representing influence coefficients and K is a constant. As the third step, the values of the most significant linear and square combinations of parameters are investigated. In the sample process being described, the following fourteen combinations of variables were tabulated for the various combinations of nozzles investigated. In the sample 243 nozzle combinations were used and 465 delta measurements were made with them. The data from the tabulation of each of the further investigated combinations of (fourteen) variables is then plotted for various given ranges of delta, e.g. delta=2025, 25-30, 35-40. To state it another way, a curve is made for delta (ordinate) vs. each one of the fourteen variables within a given range of delta. Each of the curves is evaluated to determine which of the variables influenced the patternation index delta more significantly. This is done by establishing a trend line and discarding those variables which indicate a small eifect on delta. To state it another way, those of the fourteen variables indicating an undefinable effect on delta were discarded, i.e., a screening process is used. Those variables (of the fourteen) which indicated a more significant effect, are maintained. As the fourth step, all of the computer regression analysis runs are screened for the data points which do not fit Equation 5. This identified those points where the regression analysis cannot predict patternation. These points lying outside the predictable range of Equation 5 established the critical ratios of Groups I and II. The experimental data for the sample run described above on the various nozzle combinations was made with water as the fluid at a gage pressure of 100 p.s.i. For fluids other than water, and pressures different than 100 p.s.i., suitable correction factors for density, viscosity and fluid pressure must be applied. However, the general technique described for determining the ranges of the ten ratios of Groups I and II can be used with any fluid at any pressure. For water, at 100 p.s.i., the following limits of the ten ratios have to be observed for the following objectives: (1) Description of performance.To describe the performance of all nozzles with a patternation index delta of less than, the ratios of Group I have to stay within the following limits: These four ratios as defined by the stated limits are of primary importance and determine all five nozzle dimensions with the exception of S. The ratios of Group II offer refinements in the definition of the four above ratio parameters. Group II comprises the ranges: The ratios of Group II are criteria for confirming or rejecting the ratios of Group I with regard to their suitability for describing nozzle performance. The ratios of of Group II are capable of screening possible selections of dimensions which are not compatible within themselves. (2) Prediction of perf0rmance.In order to be able to predict performance, this capability is determined by the ranges of three ratios: If we stay within the limits of these three ratio parameters, We can predict the performance of a nozzle in terms of the patternation index. It must be noted that nozzles designed within the limit of R/ D, L/ D and S/D will have a predictable performance, but not necessarily a good performance. In summary, the use of the ten ratios discussed above enable a nozzle designer toobtain a nozzle with a patternation index below a certain value, thirty in the example described with the ranges stated. The use of the three ratios R/D, L/D and S/D also permit the prediction of nozzle performance in terms of the patternation index delta. Further, Equation 5 enables a nozzle designer to ob tain by mathematical analysis the patternation index delta for my nozzle. This is extremely useful since it tells the designer whether or not the nozzle he designs has the required patternation index. Nozzles constructed in accordance with my invention have many significant advantages. Some of these are: 1) The ability to maintain a substantially constant angle of spray throughout a wide range of volume output under varying pressure heads of the fluid. (2) The ability to maintain a fairly constant coefllcient of discharge over a wide range of pressure heads and a wide range of volume output. (3) Very fine droplets at relatively low fluid pressure. (4) A minimum of pressure variation between the largest and smallest volume output, or flow rate, in order to keep the maximum pressure low. (5) Ability to have relatively large cross sections for passage of the fluid to minimize the change of clogging. (6) Uniform weight distribution of the droplets in the conical spray, i.e., good patternation over a wide range of volume output. (7) Production of fine droplets maintained over a Wide flow range. (8) High quality of atomization and high level of patternation are maintained over a relatively long life span of the nozzle. (9) When used for fuel injection into combustion chambers, stable combustion, high combustion efiiciency and faultless ignition under most unfavorable conditions over a long life span of the nozzle. (10) Etficient and effective operation when atomizing highly viscous fluids or fuels. (11) Long life span of the nozzle. (12) Effective and efficient operation of designs for smallest and largest volume output. (13) Maximally attainable ranges in flow rate. (14) Simple designs which are rugged in construction, can well resist operational stresses, e.g. due to thermal stresses. (15) Simple designs with very few component parts Which are suited for high-precision production, highest surface quality, best assembling method and easy maintenance and replacement of the components. (16) Simple designs which allow mathematical prediction of main performance parameters such as patternation, coefficient of discharge and spray cone angle. This avoids design and construction by costly and timewasting cut and try methods. (17) High combustion efficiency over a wide range of fuel/ air ratios. While preferred embodiments of the invention have been described above, it will be understood that these are illustrative only, and the invention is limited solely by the appended claims. What is claimed is: 1. A spray nozzle comprising body means formed with an inlet passage for receiving the fluid to be sprayed and a bore, means forming a swirl chamber having a portion which is in the shape of an are, said swirl chamber having an inlet opening for communication with said bore, and orifice means having an outlet in communication with said swirl chamber, said nozzle having an actual flow rate (W and a spray cone angle (2 in degrees which are related to the ratio (x) of the nozzle inlet area (B.H.) to nozzle outlet area 1rD /4 by the relationships: ( Kref f1( P M 't ref f3 =f where: B is the width of the tangential inlet close to the inlet opening of the swirl chamber, D is the diameter of orifice means outlet, H is the height of the swirl chamber, AP is the actual pressure drop of the nozzle, K is the nozzle discharge coefficient at a reference pressure drop, C is a correction factor to relate the nozzle discharge coefiicient at the reference pressure drop (K to the discharge coefficient at a particular pressure drop, 2%,, is the spray cone angle at the reference pressure drop, and is a correction factor relating the nozzle spray cone angle at the reference pressure drop to any pressure drop. 2. The method of constructing a spray type nozzle of the type having a swirl chamber at least a portion of which is in the shape of an arc of a spiral with a predictable patt rnation index value, comprising the steps of forming a body with an inlet passage for receiving the fluid to be sprayed, forming a swirl chamber having an inlet opening communicating with the inlet passage and forming an orifice means with an outlet in communication with the swirl chamber to have the ratios of the following parameters within the stated ranges where k through k are positive real number constants with k; being less than k k k and k k being less than k k and k k being less than k and it and k being less than k and D is the diameter of orifice means outlet, L is the thickness of the orifice means at its outlet, R is the largest radius of the swirl chamber, and S is the thickness of the rib formed by the inner wall of the swirl chamber at the swirl chamber inlet. 3. The method of claim 2 comprising the further step of constructing the nozzle with an inlet between said inlet passage and an inlet portion of the swirl chamber which inlet is generally tangential to an arcuate portion of the swirl chamber, the following ratios of parameters constructed within the stated ranges to make the patternation index less than a certain value, where k through k are positive real number constants where k is less than k k, and k k is less than k and k and k is less than k, and where B is the width of the tangential inlet portion close to inlet opening of the swirl chamber, and H is the height of the swirl chamber. 4. The method of claim 3 further comprising the step of constructing the nozzle with the following ratios of parameters within the stated ranges: H Z k13 and I014 where k through k are positive real number constants and wherein km is less than km, kn, km and k [(14 is less than [(10, k and k13 k is less than k and k and k is less than k 5. The method of constructing a logarithmic spray type nozzle with a predictable patternation index value, said nozzle having an inlet passage for receiving the fluid to be sprayed, a swirl chamber having an inlet opening communicating with the inlet passage and an orifice means with an outlet in communication with the swirl chamber comprising the steps of determining the nozzle parameters of inlet passage area (3-H) and orifice outlet area (1rD /4) for a given nozzle flow rate and cone angle, and constructing the nozzle with the ratios of the parameters B/ D and H /D maintained within predetermined limits where :1 and a are positive real number constants with a being less than a and B is the width of the tangential inlet close to the inlet opening of the swirl chamber, D is the diameter of orifice means outlet, and H is the height of the swirl chamber. 6. The method according to claim 5, further comprising the step of constructing the nozzle where the ratios of the parameters R/D and L/D are also maintained within predetermined limits where a and 11 are positive real number constants with a being less than a a and a 0 being less than a and a and :1 being less than (1 and L is the thickness of the orifice means at its outlet, and R is the largest radius of the swirl chamber. 7. The method according to claim 6, further comprising the step of constructing the nozzle with the ratios of the following parameters within the stated ranges: where a through a are positive real number constants where w, is less than all of the other constants a is less than a through a and a a is less than al through a a is less than a through a a is less than a a a and a a is less than a a and a a is less than a and a and a is less than a within the limits R L S 1 5 a 5 a and 4 5 5 wherein D is the diameter of orifice means outlet, L is the thickness of the orifice means at its outlet, R is the largest radius of the swirl chamber, and S is the thickness of the rib formed by the inner wall of the swirl chamber at the swirl chamber inlet, and wherein k through k are each positive real number constants with k being less than k and k and k; and k each being less than k k and k whereby a nozzle is formed in which the patternation index can be predicted. 9. A spray nozzle according to claim 8 wherein the ratio of parameters of the swirl chamber means and the orific means of the nozzle have the following range of values referenced back to water having an inlet gage pressure of 100 p.s.i. which permits the prediction of the nozzle in terms of a patternation index value. 10. A spray nozzle comprising body means formed with an inlet passage for receiving the fluid to be sprayed and a bore, means forming a swirl chamber having a portion which is in the shape of an arc of a curve, said swirl chamber having an inlet opening for communication with said bore, and orifice means having an outlet in communication with said swirl chamber, said swirl chamber means and said orifice means having the following ratios of the parameters and within the limits where D is the diameter of orifice means outlet, L is the thickness of the orifice means at its outlet, R is the largest radius of the swirl chamber, B is the width of the tangential portion of the inlet means close to the inlet opening of the swirl chamber H is the height of the swirl chamber and where h through h; are positive real number constants with k being less than h h and h h being less than h and I1 and 11 being less than h whereby a nozzle is formed in which the patternation index can be described. 11. A spray nozzle as in claim 10 wherein the nozzle is constructed with the ratios of the following nozzle parameters controlled so that the ratios of the para-meters of the nozzle are held within the following limits: and H R f hg and hg where h through h; are predetermined positive real number constants, and 117 is less than h h h and I1 it is less than I2 h and h k is less than it and h and 11 is less than I1 12. A spray nozzle according to claim 11 wherein the nozzle has the following ratios of parameters: R RH B H R referenced back to water having a gage pressure of psi. which enables the description of the nozzle in terms of a patternation index value. 13. A spray nozzle according to claim 10 wherein the nozzle has the following ratios of parameters: R L B H 12, 2.62, L82 and 4.77 referenced back to water having a gage pressure of 100 p.s.1. 14. A spray nozzle as in claim 13 wherein the selected ratios of the parameter values produce a patternation index for the nozzle of 30 or less. 15. A spray nozzle according to claim 1 wherein each of the constants hq, h and hg is less than any one of the constants h h and h.,; the constant h is less than h and the constant k is less than h 16. A spray nozzle as in claim 15 wherein the selected parameter ratios produce a patternation index of 30 or less. 17. The method of constructing a spray nozzle comprising forming a body with an inlet passage for receiving the fluid to be sprayed and a bore, forming a swirl chamber having an inlet and a portion which is in the shape of an arc and with an inlet opening for comrnunication with the bore, and the swirl chamber, the inlet opening having a portion which is generally tangential to an arcuate portion of the swirl chamber at its inlet opening, forming an orifice means having an outlet in communication with the swirl chamber, and con- 19 structing the nozzle with the parameters B, D, H, L, R and S of the nozzle where: B is the width of the tangential inlet close to the inlet opening of the swirl chamber, D is the diameter of orifice means outlet, H is the height of the swirl chamber, L is the thickness of the orifice means at its outlet, R is the largest radius of the swirl chamber, and S is the thickness of the rib formed by the inner wall of the swirl chamber at the swirl chamber inlet, so that the parameters have the ratios L B H S b1 b2; 3; n 5; 6 1 wherein 12 through. b, are positive real numbers and wherein b is less than b through b and b and b, is less than b through b b is less than b through b b,; is less than b b and b 12 is less than 12 and b and b is less than b the patternation index delta of the nozzle being calculable by the formula: where g through g are real number constants and K is also a constant. 18. The method of claim 17 further comprising the step of constructing the nozzle so that the parameters have the following additional ratios wherein b through b are positive real numbers and b is greater than 11 but less than b b is greater than b but less than b b is greater than b but less than b 17,; is greater than 11 but less than b 19. The method of determining parameters for a spray nozzle of the type having body means formed with an inlet passage for receiving the fluid to be sprayed and a bore, swirl chamber means having a portion which is in the shape of an arc and an inlet opening, said swirl chamber means also having an inlet means for communication between said bore and said inlet opening of said swirl chamber, said inlet means having a portion which is generally tangential to a portion of an arc of the swirl chamber at said inlet opening and orifice means having an outlet in communication with said swirl chamber, said nozzle when operating also having an actual flow rate (W and a spray cone angle (2 1/) in degrees which are related to the ratio (x) of the swirl chamber inlet area (B.H) to nozzle area 1rD /4 where: 20 B is the width of the tangential inlet close to the inlet opening of the swirl chamber, D is the diameter of orifice means outlet, H is the height of the swirl chamber, comprising the steps of operating a plurality of nozzles having different B and H parameters at different inlet fluid pressures which produce different actual pressure drops through the corresponding difierent nozzles, measuring the actual flow rate (W of each of the nozzles operated at the diflerent pressures to determine the respective nozzle discharge coefficient at the different pressures, measuring the spray cone angles (2,0) of the different nozzles at the different pressures, and producing by machine from the measurements made the functions f f f and 71 of the following equations: ( ref=f1 =f2 ref=lf3 2 1 14 AP is the actual pressure drop of the nozzle K is the nozzle discharge coeflicient at a reference pressure drop C is a correction factor to relate the nozzle discharge coefficient at the reference pressure drop (K to the discharge coeflicient at a particular pressure drop, M is the spray cone angle at the reference pressure drop, and C is a correction factor relating the nozzle spray cone angle at the reference pressure drop to any pressure drop. References Cited UNITED STATES PATENTS 2,218,110 10/1940 Hosmer et a]. 239-468 2,551,276 5/1951 'McMahan 239-403 3,182,916 5/1965 Schulz 239-468 2,378,348 6/1945 Wilmes et a1 239-491 2,719,755 10/ 1955 Stanley 239-492 X 2,751,253 6/1956 Purchas et al. 239-492 X 2,904,263 9/1959 Tate et a1 239-494 3,013,731 12/1961 Carlisle 239-104 X 3,198,214 8/1965 Lorenz 239-468 X FOREIGN PATENTS 760,972 11/1956 Great Britain. LLOYD L. KING, Primary Examiner US. Cl. XJR. 239--468, 492, 494 Patent Citations
Referenced by
Classifications
Rotate |