US3736797A

US3736797A - Venturi device

Info

Publication number: US3736797A
Application number: US00165608A
Authority: US
Inventors: W Brown
Original assignee: Individual
Current assignee: Individual
Priority date: 1969-05-21
Filing date: 1971-07-23
Publication date: 1973-06-05
Anticipated expiration: 1990-06-05

Abstract

A compact, high differential, low loss, curvilinear Venturi utilizing a convergent section which is modified to provide a means for creating and constantly maintaining turbulent action of the boundary layer in the general locale of the point of greatest constriction, regardless of Reynolds Number, flow patterns and types of media being measured, a geometric throat of no length aspect and a recovery section tangential to or in continuous surface relation thereto.

Description

United States Patent 1 Brown 1 June 5, 1973 [54] VENTURI DEVICE Primary ExaminerRichard C. Queisser Assistant ExaminerArthur E. Korkosz [76] Inventor: William R. Brown, 341 W. Mt. Vernon St, Lansdale, Pa 19446 Attorney-Henry N. laul, Jr., Wallace D. Newcomb and Frank J. Benasutti et al. [22] Filed: July 23, 1971 [57] ABSTRACT A compact, high differential, low loss, curvilinear Venturi utilizing a convergent section which is modified to provide a means for creating and con- [63] Continuation-impart of Ser. No. 826,624, May 21,

1969, pat 3,636,765 stantly maintaining turbulent action of the boundary layer in the general locale of the point of greatest con- 52 U.S.Cl .73 213 striction regardless of Reynolds Number, flow P 51 Int. Cl ..G0lf 1/00 terns and yp of media being measured, a geometric [58] Field of Search ..73/213; 138/44 throat of no length aspect and a recovery Section tam gential to or in continuous surface relation thereto. [56] References Cited UNITED STATES PATENTS 11 Claims, 19 Drawing Figures 3,636,765 l/l972 Brown ..73/2l3 ll l2 r/ A V/f 20 I3 g J L FLOW PATENTEDJUN 5 I975 SHEEI 1 OF 3 Fi.l

/// /l//Z// /j'/L/ //A /////////////f///fl INVENTOR.

William R. Brown fiwhfmi ATTORNEYS.

PATENTEUJUH 5 I973 3.736.797

SHEET 2 0F 3 Fig. 6 Fig. 8

FLOW-- FLOW- Q INVENTOR.

Wi Ilium R. Brown ATTORNEYS.

PATENTEUJUN sma 3.736.797

SHEET 3 [1F 3 "ESTIMATED" MEAN VALUE OF "c" 0 925 (SIZE 3.75x2.50)

0.930 |V|AL\L 0.920 A u n C 0.910

PIPE REYNOLDS N0. (|0' f I g I8 MEAN VALUE OF "c"= 0.780(SIZE 3.75x2.50)

PIPE REYNOLDS N0.x |0" Fig. I9

INVENTOR. William R. Brown ATTORNEYS.

VENTURI DEVICE CROSS REFERENCE TO RELATED APPLICATION This application is a Continuation-In-Part of my copending application Ser. No. 826,624, now U.S. Pat. No. 3,636,765, filed May 21, 1969, and discloses and claims, in part, the subject matter contained therein, which is incorporated herein by reference.

BACKGROUND OF THE INVENTION This invention relates to fluid and gas flow primary measuring Venturi devices which produce a variable differential pressure in relation to the Venturi size, Beta Ratio and flow rate, utilized to actuate secondary instrumentation of various types.

Existing curvilinear Venturis utilize a type of internal contour which, without modification, produce an uncontrolled type of boundary layer in the sense that as the flow rate decreases and viscosity and boundary layer thickness increase, the characteristic of this layer changes from a turbulent to a laminar type. When this transition occurs and as long as this characteristic prevails, an area of flow separation or velocity defect is present at or slightly upstream of the point of greatest constriction, (i.e. the throat). Therefore, low pressure measurements taken in this general locale are adversely affected and, in turn, such important factors as stability, accuracy, linearity and head loss are severely impaired. This separation (described as a velocity defect) is caused by the formation of eddies and the resultant backflow into this area because of the inability of the laminar type boundary layer to adhere to the wall of the Venturi up to and beyond the point of greatest constriction when confronted with the adverse or higher pressure gradients inherently present in the divergent or recovery section of the Venturi. In addition to the detrimental effects mentioned, pressure distribution in this general locale varies considerably in that it is not necessarily uniform or symmetrical and measurements taken not only are of less magnitude than can be achieved, but do not necessarily represent the average or mean velocity or pressure present in this area.

Thus, the actual differential measurements transmitted do not accurately represent the difference between the averages of both high and low pressure readings and the overall claimed accuracies of existing devices are highly questionable. The optimum location for low pressure measurement can not be utilized in existing curvilinear Venturis, as clearly indicated in ASME Publication No. 61-WA-80. The location of these taps vary with size and are always located somewhat up stream of the point of greatest constriction which places them in the curved upstream convergent section.

Low pressure generation can be achieved by locating taps in boundary regions of strong curvature, but the stability of the Coefficient is very sensitive to flow alignment, body form and varying drap force conditions. Today manufacturers generally locate their low pressure taps in the curved upstream sections of their curvilinear Venturis. Their Coefficient curves prove that the sensitivities mentioned are present. These (upstream) locations of the low pressure tap not only detract from the performance of existing curvilinear Venturis at relatively high flow rates when the boundary layer is turbulent, as clearly indicated by their relatively higher coefficient values and differentials which are lower than optimum for a given size and Beta Ratio Venturi and the fluctuations or unstableness of water columns as evidenced by the plus or minus deviations from the mean (or linear) coefficient values, but also they (locations) are in the immediate and exact area or locale where the separations of flow and boundary layer occur when the flow rate is reduced and the boundary layer changes from a turbulent to a laminar characteristic. Thus, they arein the exact area where separation occurs when these-conditions prevail and the undesirable results thereof are, therefore, experienced.

Major deviations from linearity in the coefficient curves of existing curvilinear Venturis range from approximately 180,000 Reynolds Numbers to 600,000 Rn, depending on size and Beta Ratio and, usually, accuracy statements are based on the individual calibra tion of each Venturi to insure an acceptable performance coinciding with customer specifications. Geometrically similar models of different sizes having identical Beta Ratios, do not necessarily produce identical coefficient values or closely similar points of deviation from linearity when tested under similar or identical conditions.

Further, test data showing effects of a single elbow immediately upstream of existing curvilinear Venturis indicate at least seven percent deviation from the nominal coefficient value, whereas, similar tests on my invention indicate only a l to 1% percent deviation. Very slight deviations in the mounting or installation of existing curvilinear Venturis, insofar as concentricity of the Venturi with the main conduit, produce sizeable changes in coefficient values and unpredictable overall performance. Whereas tests of my invention indicate that deviations in coefficient value are, as a practical matter, not measurable when the eccentricity is equivalent to 56 of an inch; while even Z1 of an inch eccentricity produces less than i of one percent deviation from the mean coefficient value. Therefore, it is not necessary to take any special precautions in the installation of my invention, whereas, a centering means is definitely required with existing curvilinear Venturis. Tests indicates that in the case of existing curvilinear Venturis, an eccentricity of only 1/10 of an inch produced a 1.3 percent in coefficient value, confirming data published by manufacturers.

Other tests of existing curvilinear Venturis versus curvilinear Venturis modified in accordance with my invention indicate that in one case the differential was increased by 20 percent and the point of deviation from linearity of this Venturi was reduced from approximately 180,000 Pipe Reynolds Number to approxi; mately 30,000 Pipe Rn. In the case of another style curvilinear Venturi, the results were even more dramatic in that the differential was increased approximately forty percent and the point of deviation from linearity was reduced from approximately 600,000 Pipe Reynolds Number to approximately 25,000 Rn. Test parameters were approximately from zero to 700,000 Rn, and Venturis having similar or identical size and Beta Ratios were tested under identical conditions. A more or less conventional Venturi, modified to the extent of having conical-like forms installed in its flat cylindrical throat, also was tested and my invention produced a truly linear coefficient curve down to approximately 30,000 Rn. vs. 130,000 Rn. a coefficient value approxi- SUMMARY OF INVENTION My invention overcomes the deficiencies and inherent undesirable peculiarities of the existing art in curvi linear Venturis and produces very definite improvements not only in this specific area but, in addition, in the general area of Venturi primary measuring devices.

My invention covers the inclusion of localized areas or completely annular bands of raised or roughened serrations projecting inwardly from the wall (singular type or series) or depressions in the wall of the convergent section immediately upstream and/or slightly downstream of the point of greatest constriction and low pressure measurement, which produce and maintain the boundary layer in a turbulent state, regardless of Reynolds Numbers or viscosity of the media being measured. The means for creating and maintaining this condition of the boundary layer can be in the form of a straight knurls approximately 90 to the direction of flow or longitudinal axis of the Venturi, a diamond type knurl, or grooves or dimples in the wall of the upstream section and adjacent to the point of greatest constriction and, if desired, somewhat downstream of this point.

These raised or depressed means cause and maintain the boundary layer to be and to remain in a turbulent condition, regardless of flow rate, viscosity and other flow conditions, thus insuring accurate and predictable, optimum measurement of the low or throat pressure and avoidance of flow separation.

Accordingly, it is an object of this invention to provide a new and novel differential pressure producing and measuring device having varied and unique means for creating and maintaining a constant turbulent boundary layer characteristic, regardless of changes in Reynolds Number or flow rate, thereby eliminating or greatly reducing the very numerous detrimental effects caused, primarily, by the transition to and presence of laminar boundary layers which are inherent in the design of existing Venturis when the flow rate is reduced to relatively low flow rates.

It is another object to provide a short, compact device, which insures a very high degree of measuring accuracy over extremely wide ranges of flow regardless of conditions of flow encountered and Beta Ratio used, in addition to producing very high differentials and very low head losses without sacrificing any of the desirable results produced by the conventional long form Herschel type Venturis.

It is another object of the invention to produce a device having widely diversified and practical utility in the flowmetering field and the capability for actuating known types of secondary instrumentation.

These and other objects of the invention will become apparent from the following description with references to the drawings.

BRIEF DESCRIPTION OF THE DRAWINGS FIG. 1 is a longitudinal sectional view of the preferred embodiment of my invention in a section of pipe;

FIG. 2 is an enlarged fragmentary perspective view of a portion of the device shown in FIG. 1 showing one of numerous forms of roughened areas that can be used in accordance with my invention;

FIG. 3 is a view taken as indicated by the lines and arrows 3-3 in FIG. 2;

FIG. 4 is an enlarged fragmentary plan view of a portion of the device shown in FIG. 1 in accordance with another embodiment of my invention;

FIG. 5 is a view taken as indicated by the lines and arrows 5-5 in FIG. 4;

FIG. 6 is a view similar to FIG. 4 showing another embodiment of my invention;

FIG. 7 is a view taken as indicated by the lines and arrows 7-7 in FIG. 6;

FIG. 8 is a view similar to FIG. 4 showing another embodiment of my invention;

FIG. 9 is a view taken as indicated by the lines and arrows 9-9 in FIG. 8;

FIG. 10 is a view similar to FIG. 4 showing another embodiment of my invention;

FIG. 11 is a view taken as indicated by the lines and arrows 11-11 in FIG. 10;

FIG. 12 is a view similar to FIG. 4 showing another embodiment of my invention;

FIG. 13 is a view taken as indicated by the lines and arrows 13-13 in FIG. 12;

FIG. 14 is a view similar to FIG. 4 showing another embodiment of my invention;

FIG. 15 is a view taken as indicated by the lines and arrows 15-15 in FIG. 14;

FIG. 16 is a view similar to FIG. 4 showing another embodiment of my invention;

FIG. 17 is a view taken as indicated by the lines and arrows 17-17 in FIG. 16;

FIG. 18 shows a coefficient curve, where coefficient value is plotted against Pipe Reynolds Numbers, produced by a typical curvilinear Venturi not modified in accordance with my invention and as now exists in the state of the art; and

FIG. 19 shows a coefficient curve, where coefficient values are plotted against Pipe Reynolds Numbers, produced by the identical curvilinear Venturi modified in accordance with my invention to produce and to constantly maintain a turbulent boundary layer characteristic regardless of Reynolds Number, flow patterns and types of media measured.

DESCRIPTION OF THE PREFERRED EMBODIMENTS Although specific forms of the invention have been selected for illustration in the drawings and the following description is drawn in specific terms for the purpose of describing these forms of the invention, this description is not intended to limit the scope of the invention which is defined in the appended claims.

Referring to FIG. 1, a device in accordance with the preferred embodiment of my invention in a curvilinear Venturi tube (designated generally 20), is shown in its environmental relationship with the pipe conduits l3 and 14 with which it is associated. The tube includes a peripheral flange 10 which is mounted in between two flanges 11 and 12 provided on the ends of

pipe section

13 and 14 respectively. The flanges l1 and 12 are held clamped together by bolts such as 15.

A high pressure piezometer opening 23 and continuant hole 22, is provided in the peripheral flange 10. The opening 23 is in the upstream surface of flange and communicates with an open type, annular pressure averaging annulus 24 encircling the convergent end of the Venturi.

An annular low pressure averaging chamber 25 in the wall of the Venturi encircles the area of greatest constriction and a series of low pressure piezometer openings 26 spaced circumferentially about the inner surface of this area of greatest constriction communicate with this chamber 25. In the arrangement illustrated, there are four such equally spaced low pressure piezometer openings. This number can vary.

In the embodiment illustrated in FIG. 1, the means for producing the results desired is shown as a series of straight ridges or serrations 27 in this inner surface of the wall of the Venturi. However, this means cannot only be any one of a number of inwardly projecting forms as shown, but also can be any one of a number of depression-like means machined or cast into the surface of the inner wall of the Venturi, such as those shown in FIGS. 10. through 13.

The Venturi includes an internal annular surface having the sectional configuration generally of a conventional curvilinear body generating from a point of stagnation or zero velocity and highest pressure at the upstream end through an inclined or arcuate surface which curves inwardly and accelerates the media being measured. This acceleration of flow and corresponding reduction of pressure, continues to the throat or point of greatest constriction and lowest pressure, commonly referred to as the geometric throat when no length aspect of throat is involved. The internal surface is inclined or curved outwardly from the shoulder gradually decelerating the flow and increasing the pressure to a point which terminates in a diameter closely approximating that of the internal or inside diameter of the pipe. Flow is in the direction indicated by the arrow. The curvature of the upstream or converging section can be formed by a single radius or by a number of radii and this condition also applies to all sections or portions and to the entire length of the inner wall surfaces of the Venturi device.

In the examples that follow, the dimensions given are for sizes 4 through 48 inch curvilinear Venturis. The roughened area or series of projections FIG. 2, shown in the form of ridges and valleys, is generally only located immediately upstream of the low pressure piezometer opening 26. The primary purpose of this means is to prevent the characteristic of the boundary layer from changing from turbulent to laminar and to constantly maintain this turbulent characteristic regardless of flow rate, flow patternand type of media flowing through this device. These modifications can be of a singular configuration in the locale described or can be a series of ridges and valleys of straight (approximately 90 to the longitudinal axis of the Venturi) or angled serrations located in the locale described or they can be continued in circumferential length to form a completely annular band of ridges and valleys or a single ridge or valley. There is no critical limiting factor insofar as the width of the band as measured on the horizontal axis is concerned. They can extend upstream to the entrance of the convergent section and downstream to the exit of the recovery or divergent section. However, for most practical purposes, the length of the band or series of multiple serrations or depressions will not exceed one-half inch immediately upstream of the throat piezometer as indicated by dimension 1" and, unless and band is used, the width or circumferential length designated as fcw will not exceed the diameter of the piezometer opening by approximately more than one-eighth of an inch. Further, the depth of the valleys and height of the projecting ridge or ridges designated as d and h generally will not exceed 0.120 inches.

FIGS. 2 and 3 show the serrations in perspective and elevation views respectively as a series of parallel ridges and valleys approximately to the horizontal axis of the Venturi, immediately adjacent to and upstream of the low pressure piezometer opening.

FIGS. 4 and 5 of the drawings show the means in plan and elevation respectively as diamond shaped serrations or knurls. The general location and dimensional criteria as defined for the valleys and projections described with respect to FIGS. 2 and 3 also apply to this form of the invention.

FIGS. 6 and 7 show the means in plan and elevation respectively as a rectangular form. Here again, the general location and dimensional criteria applying to the form shown in FIG. 2 also apply tothis embodiment.

FIGS. 8 and 9 show one of the depressed type of means in theform of a series of V-shaped grooves machined or cast into the wall of the Venturi. The included angle dimension of these grooves a can vary greatly and the bottom of the grooves can be either sharp or incorporate a slight flat as illustrated. The depth of grooves It can vary from 0.003 inches to 0.150 inches, or deeper if Venturis over 48 inches are being manufactured. The circumferential width cw" of the grooves should be at least 56 inch longer than the diameter of the piezometer opening and the length 1" as related to the horizontal axis, should be approximately k inch. The general location should be as illustrated.

FIGS. 10 and 11 show another variation of the depressed type means in the form of straight grooves in the wall of the Venturi. The cw, I and h dimensions and qualifications stated in the description of the form in FIGS. 8 and 9 can also be applied to this type form. The width of the grooves, w, can vary from approximately 0.005 inches to approximately 0.050 inches. If very large Venturis are being manufactured, this dimension, as can the others, can be varied greatly, using basically a geometric pattern of proportionate increase.

FIGS. 12 and 13 of thedrawings show still another embodiment of the means for creating and maintaining a turbulent boundary layer and of reducing the overall drag forces, in the form of a series of depressed'spots or dimples in the wall of the Venturi. These dimples can be varied in size and configuration to suit manufacturing tooling and processes and size of the Venturi. The depressions illustrated are round and for an 8 inch size Venturi, for example, should be approximately five thirty-seconds of an inch in diameter and approximately 0.040 inches deep. The cw, and 1" dimensions of the areas are approximately 36 inch plus the diameter of the piezometer opening and approximately one-half inch respectively. These dimensions can be varied generally in a geometric pattern in relation to the size of the Venturi. None of these dimensions are critical.

wall much longer (farther downstream) FIGS. l4 and show another embodiment of the means in the form of a series of ridges raised above the normal curved plane of the upstream section of the Venturi. The h, cw and 1" dimensional criteria and the qualifications thereof as stated in the description of FIG. 2, also apply to the form shown in these figures.

FIG. 16 shows one form of a singular type protrusion upstream of the piezometer hole. The protrusion consists of a single ridge or serration extending inwardly from the inner wall of the Venturi and having a rounded upper surface. The width of the ridge is indicated by the small letter w in FIG. 16. The ridge extends in the circumferential direction a finite distance as shown in FIG. 16, but it would extend completely around the inner circumferential surface of the Venturi as indicated by the dash extension lines.

FIG. I17 shows the embodiment of FIG. 16 in section.

FIG. 18 shows a coefficient curve, where coefficient values are plotted against Pipe Reynolds Numbers, produced by a typical curvilinear Venturi without modification and in accordance with the specifications of existing art.

FIG. 19 shows a coefficient curve, where coefficient values are plotted against Pipe Reynolds Numbers, of the identical curvilinear Venturi referred to in FIG. 16, which was modified in accordance with my invention to produce and maintain a constant turbulent boundary layer regardless of Reynolds Number, flow patterns or type of media being measured.

When the boundary layer is turbulent, it remains attached to the wall of the Venturi up to and beyond, that is; (downstream) the shoulder and point of greatest constriction or what is designated as the geometric throat of a curvilinear type Venturi, and reduces or practically eliminates turbulence and cavitation of the main flow in the recovery section.

The turbulence, in the boundary layer, produced by any one of the means described in this specification and illustrated in FIG. Nos. 1 through 15, is a mixing type action and this mixing generates a momentum exchange by means of which the low speed boundary layer fluid literally borrows momentum from the high speed fluid outside the boundary layer. Thus, because it is assisted by the kinetic energy transfused into it by the outside flow or main flow, the turbulent boundary layer can flow farther against the adverse or higher pressure gradients in the divergent or downstream section than can the laminar type boundary, which as described, separates from the wall at a point upstream of the shoulder or point of greatest constriction and the point of optimum low pressure measurement.

The forward (downstream) tractive effect of the mass interchange when these phenomena occur, causes the turbulent boundary layer to remain attached to the and practically eliminates any cavitation and possibility of backflow and eddies from forming in the downstream section and thus eliminates the possibility of separation of velocity defect and unstableness in the area of low pressure measurement.

Further, the total magnitude of the drag forces, pressure plus viscosity, is considerably reduced.

Thus, it can be readily understood that it is very beneficial to maintain a turbulent characteristic in the boundary, regardless of changes in flow rates or Reynolds Numbers and various types of flow conditions and flow media.

The primary beneficial effects are: (a) increased velocities and lower throat pressures resulting in considerably higher differentials and lower coefficient values for a given size and Beta Venturi at a given flow rate; (b) a major extension of the truly linear or exponential (flat) characteristic of the coefficient curve down to extremely low Reynolds Numbers, which in other words, widens the Linear or accurate measuring range of the Venturi device; (0) reduction in the scatter of readings, from the mean coefficient value or, otherwise stated, greatly improved accuracies over the extended linear range; (d) significant improvement in stability of signals produced and transmitted by the Venturi; (2) less sensitivity to misalignment of Venturi with pipe and to flow conditions upstream of the Venturi; (f) elimination of flow separation or velocity defect at or even close to the point of greatest constriction and low pressure measurement. (In fact, this phenomena is, for all practical purposes, eliminated within the entire length of the Venturi); (g) reduced head losses or pressure drops, when defined as a net loss and when defined as a percentage of the differential produced; (h) as a result of (b), fluids cannot only be accurately measured over a much widerrange, but fluids of relatively higher viscosities (than water) can be accurately measured by this improved device; and, (i) the optimum location for the low pressure measuring tap, at the point of greatest constriction, can be utilized for all sizes and Beta Ratio Venturis, thereby eliminating the necessity for varying manufacturing procedures, simplifying calculations necessary to engineer a Venturi and eliminat ing the necessity for empirically determining the point of lowest pressure. Further, as a result of the invention means, the point of greatest constriction also provides a considerably higher fidelity of measurement in that it represents the mean or average velocity (or pressure) in this area, regardless of flow rate and flow conditions.

My invention has a laying length equal to or shorter than most known compact types of compact Venturis and can be produced in any of the known insert and modified insert and flanged types of configurations. Further, the materials of construction are virtually limitless. It can satisfactorily measure saturated steam, gases and clear or solids-bearing fluids, whether conductive or non-conductive. It can also be utilized for high pressure and temperature applications and for measurement of corrosive media. Its size and Beta Ratio can be varied to produce differentials and flow rates as desired and, unlike most compact Venturis, its accuracy and stability even when Beta Ratios of 0.8 are used, are not materially affected regardless of flow conditions upstream or downstream created by commonly used appurtenances such as, elbows, tees, valves etc., installed in the pipe line.

It will be understood that various changes in the configuration of the raised and depressed-type means which have been described herein and illustrated, may be made within the scope of the invention as expressed in the claims.

It will likewise be understood that the exact geometry of the internal contour of the tube shown can be varied somewhat to achieve variations of the results desired.

Further, it will be understood that the exact or theoretical location of the throat piezometer can be varied to suit local conditions.

Further, it will be understood that, regardless of the vertical height of the obstructions, they are in the main boundary layer of the fluid. They are slightly higher than the laminar sublayer as described by Vinard and Olsen as being approxiaately 25 percent as thick as the main boundary layer which is dimensionally defined in existing art. The general dimensional parameters of the depressed type means should adhere to the basic criteria described herein. However, as clearly indicated these criteria are not critical and can be varied considerably, without changing the basic concept or overall beneficial results produced as detailed herein.

Further, while many embodiments. ofthis invention show the means for creating and maintaining these beneficial effects in the form of a series of projections and depressions, it will be understood that these forms can be of a singular nature and both of these styles can be in the form of a completely annular band.

Further, it will be understood that these modifications or forms can, if desired, extend throughout the entire length of the Venturi.

It will be likewise understood that for measurement of solids-bearing fluids, the averaging chambers at the high and/or low pressure take-offs can be eliminated and single taps at both locations can be used.

It will be further understood that the Abstract of the Disclosure set forth above is intended to provide a non-legal technical statement of the contents of the disclosure in compliance with the Rules of Practice of the United States Patent Office, and is not intended to limit the scope of the invention described and claimed herein.

What is claimed is:

l. A device for sensing pressure in a line having fluid flowing therein, comprising: a curvilinear Venturi in said line having a low pressure tap communicating with said fluid; and means providing a constant turbulent boundary layer in the fluid flowing through said Venturi in the locale of said pressure tap, said means comprising a protrusion extending into said Venturi in the convergent section thereof immediately upstream of the low pressure tap, said protrusion comprising a serrated surface.

2. The invention of claim 1 wherein said protrusion comprises a substantially flat rectangular wafer-like member extending into said Venturi from the innersurface thereof.

3. A device for sensing pressure in a line having fluid flowing therein, comprising: a curvilinear Venturi in said line having a low pressure tap communicating with said fluid; and means providing a constant turbulent boundary layer in the fluid flowing through said Venturi in the locale of said pressure tap, said means comprising an interruption in the smooth surface of said Venturi in the convergent section thereof immediately upstream of the low pressure tap.

4. The invention of claim 3 wherein said interruption comprises a plurality of V-grooves in said surface.

5. The invention of claim 4 wherein said V-grooves lie substantially perpendicular to the axial direction of flow through said Venturi.

6. The invention of claim 3 wherein said interruption comprises a plurality of rectangular slots in the wall of said Venturi.

7. The invention of claim 3 wherein said interruption comprises a plurality of dimples in the wall of said Venturi.

8. The invention of claim 3 wherein the said interruption comprises a plurality of ridges extending perpendicularly to the direction of axial flow in said Venturi.

9. The invention of claim 3 wherein said interruption comprises a ridge extending into said Venturi.

10. The invention of claim 9 wherein said ridge extends completely circumferentially about the inner surface of said Venturi.

11. The invention of claim 3 wherein said interruption comprises a depression.

Claims

1. A device for sensing pressure in a line having fluid flowing therein, comprising: a curvilinear Venturi in said line having a low pressure tap communicating with said fluid; and means providing a constant turbulent boundary layer in the fluid flowing through said Venturi in the locale of said pressure tap, said means comprising a protrusion extending into said Venturi in the convergent section thereof immediately upstream of the low pressure tap, said protrusion comprising a serrated surface.

2. The invention of claim 1 wherein said protrusion comprises a substantially flat rectangular wafer-like member extending into said Venturi from the inner-surface thereof.

11. The invention of claim 3 wherein said interruption comprises a depression.