US3076533A - Stabilisation of wind-excited structures - Google Patents

Description

Feb. 5, 1963 Filed NOV. 24, 1958 C. S CRUTON ETAL STABILISATION OF WIND-EXCITED STRUCTURES 5 Sheets-Sheet l A -4-- a 0 i I BOUNDARY or INSTAB/L/TY 7],, MAX F02 wsms/ury 01/:

To yam-Ex SHEDD/NG (d) y mx INVENTORS: CHRISTOPHER SCRUTON and I IS EUGENE SEPH SHE gni "{a a Attorneys for Applicants 1963 c. SCRUTON ETAL 3,076,533

STABILISATION OF WIND-EXCITED STRUCTURES Filed NOV. 24, 1958 5 Sheets-Sheet 2 l o-'- BOUNDARY or nvsms/ur 20 MAX FIG.3.

INVENTORS: CHRISTOPHER SCRU'ION and DENIS EUG JOSEP WALSHE Attorneys for Applicants Feb. 5, 1963 c. SCRUTON ETAL 3,076,533

'STABIL'ISATION OF WIND-EXCITED STRUCTURE-S Filed Nov. 24, 1958 s Sheets-Sheet 3 8-02 BOUNDARY or INSTABILITY '7 MAX. 0-029 A70 701/4 6"0a/ F I G 0 4 o 4 -L a I o .QMSs 60 o BOUNDAQY or wsma/L/r 7. NZD u QM 5s 40 P INVENTORS: CHRISTOPHER SCRUTON and DENIS EUGENE JOSEPH ALSHE Attorneys for- Applicants Feb. 5,1963

Filed Nov. 24, 1958 C. SCRUTON ETAL STABILISATION OF WIND-EXCITED STRUCTURES 5 Sheets-Sheet 5 STACK w/rH a-smer HEL/CAL STEAKES /0 PITCH /5x DIAMETER PLAIN STACK INVENTORS: CHRISTOPHER SCRUTON and DENIS EUGENE JOSEPH WALSHE Attorneys for Applicants 3,976,533 STABlLldATIfiN OF WENDEXUTED SERUQ'EURES Christopher Scruton, Hampton, and Denis Eugene loseph Walshe, Surbiton, England, assignors to National Re search Development Corporation, London, England, a British corporation Filed N v. 24, 195%, Ser. No. 776,075 Claims priority, application Great Britain Nov. 29, 1957 3 tllaims. Cl. lab-34) This invention relates to the stabilisation of wind-exciting structures such as chimney stacks, towers, suspended pipes and cables, and other elongated bodies of generally regular and simple geometrical shape, whether cantilever supported or secured at both ends. The invention is also applicable to structures immersed in fluids other than air, such as periscopes of submarines, and to component parts of more complex structures, although the more complex the structure, the less likely it becomes, for reasons of mechanical practicability and aesthetic acceptability, that the invention can with advantage be applied thereto.

It is an object of the present invention to provide a simple means for stabilising blufi' (i.e. non-streamlined), elongated bodies of substantially regular geometrical shape when exposed to fluid flow in a direction transverse to their lengths. The stabilising means according to the present invention may be applied to an existing body or incorporated during manufacture or erection.

The body to be stabilised according to the present invention may be generally vertical or horizontal, or may lie at any other angle to the horizon, and is stabilised by the formation on or the securing to the external surface thereof of one or more relatively low or narrow helical fins or ribs (hereinafter called strakes) extending around the body for a significant proportion of the axial length thereof. The said significant proportion will vary with the particular body concerned, and probably also with the site conditions, such as adjacent structures. In general, however, for a tall cylindrical stack a proportion of between one third and two thirds of the axial length, measured from the top, will sufiice for stabilisation.

In any given structure, the optimum height of a stroke and the pitch of the helix will probably best be determined empirically, usually as a result or" wind-tunnel tests, although it appears at present that, for cylindrical or polygonal prismatic structures such as pi es and chimney stacks, the optimum ratio of stralze height to cylinder diameter or equivalent characteristic transverse dimension D will normally lie between 0.02 and 0.2, whilst the optimum helix pitch, at least for three equiangnlarly spaced strakes, is of the order of fifteen times the said diameter or characteristic transverse dimension D.

Long structures or" bluff cylindrical section are prone to oscillate in winds when they have low natural frequencies and small values of structural damping. The oscillations occur usually in bending modes in a direction transverse to that of the wind and at the natural frequency of the structure. Practical examples of such oscillations are to be found in the swaying of tall smokestacks and in the galloping of transmission lines. Fundamentally, these instabilities are probably not essentially different in character from the single-degree-of-freedorn oscillations of suspension bridges, since both are due to how separation and the production of vortices, but smoke stacks are usually of simple cylindrical or polygonal shape and sulficient is known of their aerodynamic characteristics to explain the mechanism of the instability.

It is already known to improve the stability of bluff, elongated bodies such as chimney stacks by increasing their structural damping or by increasing their natural frequencies of oscillation. An example of the application of structural damping is the incorporation of shockabsorber type dampers in stays attached to the body, but it is usually difficult to find economic means of producing satisfactory increases in the natural oscillation frequencies of structures, and this is especially true of a cantilever structure such as a chimney stack.

A third method of improving stability is by modifying the shape which the structure presents to the wind or fluid flow. The present invention falls in this category, and aims at providing a means for stabilising bluff elongated bodies which does not require th addition of damping devices to the structure and avoids the difficult design problems encountered in obtaining large increases of natural frequencies. Certain modifications of the external shape for this purpose--such as the well-known splitter plate in the wake of a cylinder-are applicable only when the direction of flow of the ambient fluid relative to the body is con tent and prescribed, as when the body is moved in one direction through still air or water. A perforated shroud surrounding a cylinder with clearance, and saw-tooth spoilers on its external surface, have been proposed before, but the present invention is considered to have an easier and more economical application, and to be aesthetically more acceptable.

One source of excitation is the shedding of discrete vortices and this may be exemplified by the swaying of tall stacks. The alternate shedding of vortices, first from one side of the stack and then from the other side, re sults in a flow pattern known as a vortex street or man vortex trail, in which are produced a double row of evenly spaced, staggered vortices of equal strength but alternate rotations. The growth and shedding of these vortices produce changes in the circulation round the stack that are equal but of opposite sign, and a periodic force in a direction normal to the wind stream results. The frequency of this force when the stack is stationary is given by the Strouhal relation where D is a characteristic transverse typical dimension of the stack measured at right angles to its axis and the wind direction; V is the wind velocity; and S is a number depending on the shape of the body and varies be tween 0.15 for a square section cylinder to 0.20 for one of circular section. it from experimental results that for an oscillating elongated regular body there are certain ranges of wind speed for which the oscillations themselves control the vortex frequency. Vortex-shedding, however, does not always account fully for the instabilities of bluil bodies.

In addition to the swaying oscillations of tall stacks induced by wind, oscillatory ovalling of the circular section as an elastic ring has been observed near the top of stacks. The frequency of the ovalling and the wind speed at which the oscillations occur suggest that this second type of oscillation is also due to vortex shedding. Such oscillations have been remedied by the addition'of stiffening rings near the top of the stack.

Since the method of stabilisation according to the present invention depends on the breakup of the regular formations of vortices alternately from opposite sides of the elongated bluil bodythe so-called Karman streets-- which give rise to aeolian instability (see FIG. 1 be low), it is unlikely that, in any given instance, a single helical strake will suflice. Strake height and helix pitch both play their respective parts in the final result, but clearly there are practical upper and lower limits for both. Thus, for example, as the helix pitch tends to infinity,

sewa e the stroke approximates more and more closely to a lateral fin which, if it projects transversely of the wind direction, merely constitutes a corresponding increase in the characteristic transverse dimension D in the above-mentioned Strouhal relation, and contributes nothing to vortex break-up. Conversely, as the pitch tends to zero, the straize approaches a solid outer layer which again merely increases the characteristic transverse dimension D.

Again, as the ratio of strake height to characteristic transverse dimension D diminishes, the eiiectiveness of the strake on the air layer adjacent the body surface decreases, whilst, conversely, as this ratio increases the actual load on the body due to air resistance of the strake itself increases and may become prohibitivelylarge.

Between the obvious practical upper and lower limits or" these two parameters, therefore, there lies an infinite number of possibilities for any given case. Economic and aesthetic considerations also play a part in the final determination of the optimum number and dimensions of the straltes or ribs to be formed or applied on the body. Too many strokes become unduly costly without contributing pro rata to the final stability, whilst a low value of helix pitch may unduly increase the cost of manufacture and produce an unsightly result.

Une practical embodiment of the present invention, together with certain experimental data relevant thereto, will now be described, by way of illustration only, with reference to the accompanying drawings in which:

FIGURE 1 is a curve showing the effects of structural damping on the instability of a square-section elongated prism where the wind direction is normal to one face;

FIGURE 2 is a curve similar to FIGURE 1 for a circular-section cylinder;

FIGURE 3 is a side view of a cylindrical column to which the present invention has been applied, mounted for test in a wind tunnel;

FIGURES 4-7 are curves similar to FIGURE 1 of test results on the column of FIG. 3, and

FIGURE 8 shows comparative curves of stability for a plain cylinder fixed at one end only and the same cylinder fitted with strakes according to the present invention.

In FIGURES l, 2 and 4-8, the following symbols have the following meanings:

D=characteristic transverse dimension typical of the structure-e.g. diameter, or distance across flats of a polygonal section.

M :mass per unit length of a body,

N :frequency of oscillations in cycles per second.

V=wind speed in feet per second.

6s:logarithmic decrement due to structural damping.

n =maxirnum linear displacement per unit dimension (D) from the zero position during a period of oscillation.

=density of air (or other fluid).

FIGURE 1 of the accompanying drawings shows an experimentally-dctermined stability diagram for a long rigid prism of square section when constrained to oscillate in a motion at right angles to its span and to the wind direction. The instability region bounded by curve (a)(b)(c) (aeolian instability) was evidently due to vortex-excitation. 'The region above the curve (c)(d) (galloping instability) cannot be explained in the same way, but it is consistent with theories based on the changes of aerodynamic force which occur during a cycle of oscillation and are due to the variation of the eiiective wind inclination produced by the motion of the body.

FIGURE 2 shows the corresponding stability diagram for a circular section cylinder. It will be seen that only aeolian instability is apparent.

. In order to make the instability curves of general applicability, the ordinates and abscissae of the curves are shown as dimensionless or scalar quantities representing, respectively, structural damping and wind velocity.

FIGURE 3 of the accompanying drawings shows a d cylindrical body 1 of circular section mounted in a wind tunnel 2. Three helical stralces 3 are equiangularly spaced around its circumference, the helix pitch being about 15 diameters (D) of the body 1. Various tests were carried out on this arrangement with straites 3 of diilerent (radial) heights to determine the critical wind speeds for various values of structural damping, the ordinates and abscissae being expressed non-dimensionally, as in FIGURES l and 2. The results of the tests are shown in FIGURES 4-7, the heights of the strakes 3 being noted on each figure as the ratio (H) of radial height to the characteristic transverse dimension (diameter) D.

FIGURE 4 shows that quite a low strake height ratio (H =0.029) is effective in reducing aerodynamic excitation, and hence the amount of structural damping required to eliminate the oscillations. By increasing the value of H to 0.059 (FIGURE 5), further improvement is obtained, but the most notable eiIect is the general increase in the critical (non-dimensional) wind speed. Still further improvements in both aeolian instability and critical speed are shown in FIGURE 6, where H=0.088, whilst in FIGURE 7, with H =Q.ll8, the aeolian instability region is reduced to a very small area at stillhigh'er values of critical speed, and only a very small value of structural damping is required under these conditions to eliminate the oscillations. No further tests were made beyond the value of H in FIGURE 7, since the area of instability was regarded as of no practical significance.

In every test, the wind speed was raised to a high value to ascertain whether the straltes 3 introduce a region of galloping instability, but no evidence was found.

The tests described above were made on a rigid cylinder spring-mounted in a wind-tunnel to give approximately two-dimensional conditions. To demonstrate the efiicacy of the device under the three-dimensional conditions which occur more commonly in practice, further tests were carried out on a dynamic model of a tall stack. This consisted of a long metal tube fixed at its base but otherwise permitted to bend freely. FIGURE 8 shows the results of wind-tunnel tests made on the plain model and on the model fitted with a stroke configuration as for the tests illustrated in FIG. 7 with H =0.ll8. The strakes are just as effective for these conditions, the oscillations of the stack fitted with strakes being only just discernible at a maximum amplitude of 0.05D as compared with 0.75D for the plain stacks both being measured at the lowest value of damping (1.3).

From the foregoing it will be seen that helical strakes 3 are effective in reducing or suppressing wind-excited oscillations in elongated cylindrical bodies of circular section. The height of the strakes in any given case will depend on the amount of inherent damping in the structure, this latter being lowest for welded steel stacks and the like, but in general the value of H need not exceed 1/8 for even the most lightly damped structure.

While the essential concept of the invention relates to its influence on the air or other fluid flow, the addition of strakes will have some structural effects which, however, can only be beneficial since the natural frequencies and the structural damping will both be increased by the addition of the strakes.

Since stability diagrams for unstabilised polygonalsection prismatic structures lie between those of FIG- URES 1 and 2 for square and circular sections, respectively, it is reasonable to expect helical strakes 3 to be effective on such shapes, especially where the number of sides is not too small-cg. octagons, decagons, or dodecagons. Furthermore, since the strakes 3 are operative by virtue of their eiiect of breaking up vortex formations, it is reasonable to assume that they would be efiective against ovalling and butfeting. Butte-ting oscillations may be set up in a structure due to the eddying airflow downstream of a nearby structure, as in the case of a chimney situated to leeward of an adjacent chimney.

The invention is applicable to a variety of bluff, clon a as").

gated structures, as has been indicated above. In some cases, however, detail modifications may be made to suit local conditions. For example, in the cases of chimneys, towers, and the like, the important mode of oscillation is the fundamental, in which the movement of the upper parts of the structure is considerably more than that of the lower, and hence the upper portions are much more effective in extracting the necessary energy from the fluid stream to produce and maintain the oscillations. In these cases, therefore, it is only necessary to fit the strakes over the upper portion. In the experiments on the flexible model stack described previously, the strakes were omitted from the lower third of the total axial length of the stack without noticeable loss of effectiveness. Furthermore, the strakes themselves need not be continuous, short gaps being permissible at intervals along their lengths.

The helix pitch of a strake may have a minimum practical value for a given diameter or characteristic transverse dimension D below which no material advantage is gained. The maximum practioal value may in part depend on the number of strakes used, and vice versa. Tapering of a structure, such as the normal gradual enlargement towards the base, has no observable effect on the aeolian instability thereof, so that helix pitch may be maintained constant throughout the axial lengthof such a structure without detriment to the effectiveness of a strake.

Tests have been made on models in which the effects of change of pitch were observed. In the first tests, the strakes 3 on the cylinder shown at 1 in FIGURE 3 had their helix pitch halved-Le. the pitch was reduced to 7 /2 diameters. The same effective reductionvof oscilla tion by fluid excitation was observed. In the second tests, a truncated cone had strakes with a constant ratio between helix pitch and cone diameter, so that the pitch varied along the length of the cone. Again, effective suppression of oscillations was observed.

We claim:

1. In combination with a bluff elongated cantileversupported body of generally regular geometrical shape, means comprising at least one helical strake secured to the external surface of said body and extending in the direction of the longitudinal axis thereof over a substantial portion of the exposed length of said body for preventing the formation of Karman vortex streets downstream of said body and thus stabilizing said body against excitation by a fluid moving transverse to the longitudinal axis thereof.

2. In combination with a bluit elongated cantileversupported body of generally regular geometrical shape, means for stabilizing the body against excitation by a fluid moving transverse to the longitudinal axis thereof comprising at least one helical strake secured to the external surface of said body and extending in the direction of said axis over a substantial portion of the exposed length of said body, the radial depth of said strake relative to the characteristic transverse dimension of said body being in the ratio of about 0.02 to 0.2 so as to prevent the formation of Karman vortex streets downstream of such body.

3. In combination with a blufi elongated cantileversupported body of generally regular geometrical shape, means for stabilizing the body against excitation by a fluid mowing transverse to the longitudinal axis thereof comprising a plurality of strakes secured at equiangular intervals around the external surface of said body, each extending in a helical path at a pitch of not more than about 15 times the characteristic transverse dimension of said body for a substantial proportion of the exposed length of said body so as to prevent the formation of Karman vortex streets downstream of said body.

References Cited in the le of this patent UNITED STATES PATENTS 739,269 Tilden Sept. 15, 1903 935,129 Speed et al. Sept. 28, 1909 2,216,758 Schmidt Oct. 8, 1940 2,859,836 Wiener Nov. 11, 1958

Claims (1)

1. IN COMBINATION WITH A BLUFF ELONGATED CANTILEVERSUPPORTED BODY OF GENERALLY REGULAR GEOMETRICAL SHAPE, MEANS COMPRISING AT LEAST ONE HELICAL STRAKE SECURED TO THE EXTERNAL SURFACE OF SAID BODY AND EXTENDING IN THE DIRECTION OF THE LONGITUDINAL AXIS THEREOF OVER A SUBSTANTIAL PORTION OF THE EXPOSED LENGTH OF SAID BODY FOR PREVENTING THE FORMATION OF KARMAN VORTEX STREETS DOWNSTREAM OF SAID BODY AND THUS STABILIZING SAID BODY AGAINST EXCITATION BY A FLUID MOVING TRANSVERSE TO THE LONGITUDINAL AXIS THEREOF.

Images