JPS6323007A - Hydrodynamic wall surface - Google Patents

Abstract

(57) [Abstract] This bulletin contains application data before electronic filing, so abstract data is not recorded.

Description

DETAILED DESCRIPTION OF THE INVENTION Field of the Invention The present invention relates to the modification of boundary layer turbulence flowing over hydrodynamic walls.

[Prior art and its problems]

Many studies have been conducted in recent years on the effect of small wall shapes on turbulent boundary layers.

A row of small longitudinal rips across the turbulent boundary layer region of the wall on the wall? Special attention has heretofore been paid to the provision of so-called riblet walls extending in the flow direction of the flowing fluid, and experimental results have shown that net or wall drag reductions of approximately 70% can be achieved.

The use of rectangular fins in a paper entitled "Drag Characteristics of V-Grooves and Transverse Curvature Riblets" (presented by M. J. Walsh at the Symposium on Viscous Drag Reduction, Dallas, Texas, November 7-8, 1979). turbulent burst velocity (i.e., the low-velocity longitudinal vortices or "
3 by reducing the rupture velocity of the "streak"
An early study by Nie, Klein, and Johnson (1966) was mentioned, in which a reduction in drag of -4 mm could be obtained. Walsh's paper reports on the study of a series of alternative rib profiles, in which he claims that drag reductions of up to 7 mm can be achieved by using V-groove prelaminates. This has been described by experimenters as representing about the best reduction in skin friction drag that has been achieved to date.

Drag reduction is achieved by R, E, Fupulco (Reno, Nevada, 19
January 10-13, 1983, AIAA-83-0377,
As has been suggested by a number of sources, such as the AIAA 21st Space Science Conference (AIAA 21st Space Science Conference), random luck in the spanwise direction of the streak is related to the ability of the total limit or pret. Johnahan and Smith (Col.), Boulder, 198
March 12-14, 5, AIAA-85-0547,A
The IAA-Shear Flow Control Conference) found that Walsh was able to obtain his optimal results by using v-groove riblets and small-height cylindrical riblets to create areas of low-velocity streaks within a limited area at the top of the wall. However, their experiments also showed an increase in drag of 3 to 8 degrees. Very recently, s, p
, Wilkinson followed Johnahan and Smith in using a rectangular replet to fix or transmit a slow streak using a riblet. Attempts were made to control the rupture, but it has not yet been reported whether a net drag reduction was achieved thereby.

The results reported from these and other previous studies all indicate that the effectiveness of libretto is rather limited and limited, and therefore alternative solutions are sought.

Skin layer friction by providing superimposed replet surfaces on large eddy current breakers (LEBU's) as manbrators for the outer regions of the boundary layer? It has been proposed to effectively reduce the
No., M. J., Walsh et al. - NASA case LAR-13
286-1), such an arrangement considerably increases complexity and is prone to wear, especially accidental damage. Another suggestion (“
``On the drag reduction effect of sharkskin'', Vera ζ Yard, Hoppe, Rife, AIAA-85-0546), using local flow conditions on the ridgeline pattern structure similar to the scale shape on sharkskin to improve tangential mixing. It has been suggested that it is appropriate to use a so-called vortex-generating surface that has a shape that emits a vortex.

It has been argued that this can lead to a significant reduction in drag, but researchers have concluded that it is inherently complex and difficult to imitate a forceful scale surface with low drag characteristics. Probably.

In contrast to the above attempts to reduce skin layer friction by rather complex means, replet surfaces are less susceptible to damage and they can be applied in a relatively straight forward manner, e.g. by preforming and extrusion layers? It can be formed by cutting, pressing, or coating. What if the skin layer friction drag? It is clear that it would be desirable to use such a surface if it could be greatly reduced.

The mechanism by which such wall-climbing prets are successful is not yet fully understood, although different researchers have published conflicting materials regarding the change in the net reduction effect on skin layer friction depending on lip length. is probably important. Even if it is assumed that the replets produce boundary layer turbulence close to laminar conditions, in the equivalent turbulent and laminar regimes the skin layer friction drag in the laminar case is greater than that in the turbulent case. Does that mean it's going to be 80 times smaller? Keep in mind that their efficiency is low. If this improvement potential of this large proportion were achieved, the use of replet surfaces would be able to offer significant advantages over the combined drag reduction schemes described above.

[Means for solving problems]

According to the invention, a hydrodynamic wall is provided with a series of elongated protrusions extending in the direction of fluid flow relative to it for deforming the turbulent boundary layer thereon, the protrusions being adjacent in each repeating pattern. The protrusions are arranged in a spanker-oriented repeating pattern that protrudes to different heights.

Preferably, the protrusions form a substantially continuous spanwise row of replets. Similarly, it is desirable to form protrusions with a sharp pointed profile.

In the simplest case, the repeating pattern consists of alternating high and low riblets.

However, three or more height glue plet patterns can be formed with advantages over the known uniform height glue plet surfaces. Desirably, such a pattern is symmetrical.

If we consider the example of alternating high riblets and low riblets, we find that the lower riblets cause the turbulent eddy motion in the boundary layer to penetrate all parts of the wall surface, especially deep into their grooves, approximately 4 mm above their height in the wall soil. 3/3 is established as an effective surface and serves to suppress displacement of turbulent flow from the wall.

The alternating high riblets are sized relative to the low riblets and the boundary layer thickness such that they have a movement that extends this effect to its limit. That is, because of their large scale, their action complements that of the low riblets, and the turbulent vortex motion is maintained far from the wall surface.

If the pattern consists of three different height lift patterns, this effect plays one step further, but the additional improvement is now less than the improvement already provided by the two different height lifted patterns. Let's become. 3
Using a pattern with lips having more than two different heights will generally not provide sufficient further improvement to justify its complexity.

The truth seems to be that the already known types of glue plates act in a totally or primarily passive manner and do not have any significant influence on the development of the vortex itself. This also applies to the wall shape of the present invention, but a greater reduction in skin layer friction can be obtained.

This is because, in addition to providing its own contribution, the lower replet modifies the deep boundary layer region so that the higher replet can act much further away from the wall as well.

Depending on their geometry, replets can also be somewhat effective in suppressing the spanner-directed gradient associated with the formation of longitudinal vortices or "streaks" in the boundary layer and initiating secondary small-scale longitudinal vortices. has. however,
This does not seem to be a significant effect given the relatively small drag reduction effects obtained in published studies on replet surfaces, and it is likely that the effects of the improvements provided by the present invention will not affect replets. does not depend on

The relatively low efficiency of known glue bullet surfaces also tends to make their performance self-limiting. The turbulent eddy motion of the beams by the wall platets tends to reduce the shear stress in the wall, which reduces the effective height of the platets and thus has a shearing effect that inhibits their efficiency. In the present invention, the high replet is present to reinforce the effect of the low replet and thus to further push the turbulent vortex motion away from the wall surface. At the same time, this automatically compensates for any small replet efficiency losses. Another quadratic effect is due to the fact that there is a slight progressive decrease in wall shear stress over the critical flow evolution length (i.e. streamwise distance along the wall) in the transition region and the early part of the turbulent boundary layer. You can also get some benefits. A small replet with a height commensurate with the initial region of the turbulent boundary layer will therefore have progressively little effect along its length, but the taller replets now present will increase the effective region of the lip. Expand to the length of a large flow development.

In a preferred configuration according to the invention, at least the higher platelet has side surfaces sloped at an angle that varies depending on the height of the projection, each side surface having an intermediate region having a smaller slope than the adjacent regions above and below. and has a side surface forming the lower portion of the concave cusp of the associated protrusion. In the simplest case, the protrusion pattern consists of alternating high and low protrusions, the small and large protrusions being arranged in such a way that the turbulent vortex motion in the boundary layer is displaced away from the wall surface.

Small protrusions by themselves would not be able to establish an effective surface on the upper part of the wall by more than about three quarters of their height, but tall protrusions have the effect that they magnify this effect to its extreme. It can be dimensioned for small protrusions.

For best effect, the transition to the concave-sided point should be positioned at or near the effective height of the wall established by the projection. Preferably, the lower transition region has a generally planar surface so as to form a series of V-grooves with steep sides between the protrusions within its lowest extent. These act, as already mentioned, to displace turbulent motion away from the wall surface. The longitudinal vortices displaced on the active wall are then limited by concave cusps projecting into this area, which, due to their different external shape, can act more strongly on the vortices. can. If the displacement of the effective wall surface becomes large due to the combination of projection heights, and the effective surface is placed above the height of the smallest projection, no benefit can be gained by deforming those sides. do not have.

It is an advantage that to the extent that any of the small protrusions extend beyond the effective wall position, their sides are given a similarly varied slope such that the transition is at the height of the effective wall position. This causes a change in slope at a certain height above the actual wall surface from which the protrusion originates, but in special cases the protrusion pattern creates an effective wall surface that varies spanwise with the protrusion pattern, and The changing position of the slope on each protrusion is made possible to follow the change in their height.

The transition to the concave shape can be in the form of a generally sharp edge, but it can also be a rounded or chamfered transition. It is preferable that the concave surface has a relatively sharp edge with a continuous curve at the tip, but it can also be made up of a group of straight (or) curved sections, and for convenience, the tip itself can be flat or rounded. The present invention will be explained below with reference to the drawings.

〔Example〕

Referring to FIG. 1, a wall is depicted having a high replet)R and a low replet)r, and this pattern repeats in the spanwise direction, with the replet running streamwise on the wall in the turbulent boundary layer region.

It is desirable to start from the flow transition area on the wall.

Preferably, the streamwise extending pattern is continuous throughout the turbulent boundary layer region of the wall. It is not important that the replets align exactly with the fluid flow; distances of up to 10-15 are permissible. The replets are all shown as triangular, but other shapes can be used, including a concave shape between successive cusps, as shown in Figure 2 with the replets R' and r'. It is desirable to have a sharp point with a trough (the apex preferably forming an angle no greater than about 20°). Triangular glue plets separated by narrow flat valleys can also be used. Furthermore, although Figure 1 shows large and small glue plets with the same base width to give a relatively acute 2-sided triangle, it is possible to increase the base width to approximately twice the height of each replet. is desirable. That is, this relates to the cusp shape shown in FIG.

The dimensions and dimensional relationships of the replets can be expressed in the so-called dimensionless form of the so-called "wall modulus", where the experimental distance value is multiplied by the "wall unit".The scalar quantity ν is The fluid density ν is the fluid kinematic viscosity. In this example, the small rivet r has a dimensionless height h of 15 units, and the large rivet R has a height h of 25 units.
It has e. The width W+ of each replet is Z5 units giving a pitch si+ of 15 units for the pattern repeat.

Due to the presence of the small replet r, an effective surface for the turbulent boundary layer flow is established above the wall itself. The influence of replet r is large; the effective height of replet R is b + e),
That is, the large prets are sufficient to set this effective surface at about three quarters of their height on the wall so that its height above the effective surface is therefore about 13 units. Add their own influence to this upper region so that the effective surface is further displaced.

Many variations in the dimensions and geometry of the replets are possible within the scope of the present invention. The ratio of h:s is not less than 1:2 but 2
: Desirably not more than 1.

Within this range, the aspect ratio (height to width) can be particularly high for large plets. Regarding dimensions, the following dimensionless wall unit dimension range is desirable for low-velocity flow. For a small replet r, 2(h(20, L or 5(h(15) is preferable.

For large pret RK, s<h, e”(45, L-shaped 15(h, e (35
is desirable.

As for the pattern repetition pitch, +6<at<50-L<20<sI+40 is desirable.

Changes within the above ranges can be made individually for each parameter, but the large replets must protrude above the small replets, and a minimum protrusion of three wall units is necessary for their effects to be seen. Needless to say, this is said to be the case.

If a longer replet pattern is used and implemented, the pattern repeat pitch will be much shorter, but will maintain approximately the same base width for the individual triplets. In the case of replets with three different heights, i.e., when an intermediate height fret ri is interposed between the high replet and the low replet as shown in FIG. 3, the pattern repeat pitch is doubled. become. This is also true for the example of FIG. 4, where the height of the beam prets is reduced to a minimum in the centrally located region between the taller replets. The contours of the different heights may also be different.

Figures 5 through 7 illustrate wall replet patterns within the scope of the present invention having modified plet profiles. The effective wall surface is established at three quarters of the height of the small replets in a pattern of alternating replets of two different heights. The pattern pitch is 15 wall units, therefore + + it is the pitch of small and large glue plets (S=St).

The height of the small replet is 15 wall units and that of the high replet R is 25 wall units. That is, at the top of the effective wall they have a height of 1 δ units (hoe) for large replets, and a height of 3 units (h hoe) for small replets.
e). From the point of transition t, the sides of the ripplets extend upwardly with the already mentioned sharp-pointed concave profile, but in the lower part of the effective wall they form a V-groove with deep sides. There is.

If the presence of large ripples has the effect of increasing the effective wall surface, the small ripples are located entirely below the effective wall surface. FIG. 6 shows a modification of the column of FIG. 5 in which this occurs, but it shows that it begins to lean toward small replets. Since the height of the small replets is almost this small, they have a perfectly planar shape, while the sides of the large replets rise and have a concave shape across their effective height from the effective wall surface. granted.

However, it has been found that the variation in height between successive glue plets creates a similarly undulating effective wall surface, especially when the pattern is composed of two or more glue plets. This can be accommodated by a shape such as that shown in FIG. It is assumed that in this case the effective wall surface of the small replet R is equal to that of the example of FIG. 5, whereas in the case of the large replet R it corresponds to that of FIG. The inflection points of the small replets, i.e. their intersection with the effective wall surface, lie at the actual wall surface, i.e.
In the case of large plets, the inflection point is at the roller. Small replets with a concave external shape, while large replets have a concave external shape as shown in Figures 5 to 7. In order to effectively perform their functions, the concave surfaces are relatively sharp. The cusp of the edge must be formed. It will be appreciated that the above-mentioned effects can be achieved at least substantially in the case of deformations of the sloping sides of the figures. In particular, for practical reasons it may be desirable to blunt the inflection points or cusps on the plet surface. That is, in a similar manner, these surfaces can be provided with straight and/or curved sections both above and below the effective wall surface without unduly changing the general character of the external shape required for the above-mentioned function. Additionally, the cusps of the large plets can be flattened or rounded to reduce their inherent drag without adversely affecting their operation.

In all cases, it is possible to reduce the wetting area in the longitudinal direction of the flow (with a boundary layer thickness δ greater than It is also possible to have a short spanwise extending gap (small).

Our two simultaneous applications (
It is also possible to use either or both of the shapes that are the subject of ``Convoy 1.11-r Hydrodynamic Wall''). Regarding the increase in the height of the glue plets in the flow direction, which is a feature of case ①, if this is applied to the glue plets of different heights of the present invention, the degree of increase will be continuous or continuous as shown in Figure 8. It can be done in stages as shown in FIG. The self-limiting influence of the length of the increase in height and the reverse evolution and the progressive decrease in the wall shear stress progressively inoperate the smallest replets first, so that a stepwise increase in height can be provided in a staggered manner. In other words, Fig. 9 shows two glue pregs with different heights) rl @
In the case of a spanwise pattern of rt, it shows how a small riblet r1 is continued by a larger riblet r, then a third series of riblets higher than rl, and if more If a step is required, the latter is continued by a fourth, even larger series of glue plets, so that at each station in the machine direction the pattern always consists of glue plets of two different heights. You can do it like this. Figure 9 also. At the leading edge of the ripple pattern, how the ripplet can be faired into the wall to avoid increased drag due to flow meeting the steep front surface.
This is what is shown.

If the height of the rivet increases in progressive steps, the leading edge of each step can be similarly furled. (However, this point is not shown.) The dimensions of the previously mentioned prets are reasonable given the results performed for Matsuha numbers (M) of about 0.3 or less at relatively low free flow velocities. Seem. However, changes in one dimension, particularly the height of the riblet, have been found to be desirable for higher speeds and Matsuha numbers. The machine direction length of the projections would appear to have an important influence on the optimum dimensions, especially in the case of pine numbers greater than about 0.4M.

A wind cylinder test with a uniform row of single-height V-groove plets with a streamwise length of approximately 1 meter is required to give optimal results at high speeds This illustrates the changes in The following table showing the ripple dimensions at the leading edge of the ripples shows these at Matsuha numbers in the range 0.4-0.9 M in relation to the previous low-speed results in the range 0.1 M11), 12). test +
31, +41.

This shows data obtained by performing (5).

Riblet height/span ＋　　　＋ h,sRemu wall unit m13 0.8x10 12115 3x10 (31175X10 (4) 20 10x10' [5) 25 15x10' Ra is the length Reynolds number = L beam is 7° and di It is the length of the cut.

The above results show that in order to obtain the optimal effect on silkworms, an increase in the height of about 50 to 2 m at the leading edge of the special pret is required.
1=IX10 indicates that a 10 times increase in Ret is required compared to the low special value.

The factors that determine the degree of optimality are numerous and complex. The shape of the underlying airframe with curvatures or discontinuities in the flow direction and/or the snow direction, and the possibility of non-uniform U-values over the range of the replet rows, introduce further influences and optimize the A light evaluation of riplet dimensions may be required. However, for general purposes,
If compression of Ret is achieved by increasing the leading edge height of the riblets or by spacing the riblets, then ReL( h; Re
It may be suggested that a suitable scaling factor of the leading edge height of the wall unit glue plate with respect to t, )K should be in the range 1:20 to 1:6.゛It is also desirable to calculate the increase in riplet size along the length of the row regardless of ReL, as discussed in detail in another patent filing of ours filed at the same time, Vehicle Row ■. .

〔effect〕

The present invention is typically applied to the base of an airframe where a reduction in skin layer friction is desired.

It therefore provides a means of reducing drag on the exterior surfaces of vehicles, including aircraft, land vehicles and seaplanes. It can also be used on rotating turbomachines as well as fixed vane seven-hole machines and on the inside surfaces of pipes and conduits. The present invention includes the possibility of providing a surface member with projections which can be applied to the fuselage in order to form surface protrusions in one layer with the associated fuselage and provide the necessary surface profile on its upper part, such as a sheet or tape.

[Brief explanation of drawings]

FIG. 1 is a schematic cross-sectional view of a wall surface formed perpendicular to the flow direction on the wall surface according to the present invention, and FIGS. 2 to 9 can be applied to the shape of FIG. 1 individually or in combination. Sectional views of a series of modifications. R/, rI...Riblet, hb...Height, W+...Width, +Re...Pitch. (4 others) Engraving of drawings (no change in content) Procedural amendments (August 1, 1988) Director General of the Patent Office Kunio Ogawa 2, Title of invention: Hydrodynamic wall surface 3, Person making amendments and Relationship Applicant Address Name Rolls φRoyce φPC 4, Agent Address Room 206, Shin-Otemachi Building, 2-2-1 Otemachi, Chiyoda-ku, Tokyo

Claims (1)

Claims: 1. A hydrodynamic wall surface with a series of long protrusions extending thereon in the direction of fluid flow for modifying the turbulent boundary layer at the top thereof, the protrusions being adjacent to each other in each pattern repetition. Said wall surface characterized in that the projections are arranged in a repeating spanwise pattern projecting at different heights. 2. The wall surface according to claim 1, wherein the heights differ by at least three wall units. 3. A wall surface according to claim 1 or 2, characterized in that the pattern consists of alternating individual protrusions of two different heights. 4. A wall surface according to claim 1 or 2, characterized in that the pattern consists of a symmetrical arrangement of protrusions of five different heights. 5. A wall according to any one of claims 1 to 4, characterized in that the projections form a substantially continuous spanwise row. 6. Claim 1, characterized in that the protrusion extends in the flow direction over at least the length of the turbulent boundary layer region.
Wall surfaces described in any one of Items 1 through 5. 7. The wall surface according to any one of claims 1 to 6, wherein at least a part of the pattern has a sharp pointed shape. 8. The wall surface according to any one of claims 1 to 7, wherein the sharp pointed protrusion has a V-groove. 9. Claim 7, characterized in that the sharp pointed projection has an outer shape with a concavely curved valley between the points.
The wall surface described in section. 10. Wall surface according to any one of claims 1 to 9, characterized in that valleys with flat bottoms are provided between at least some of the projections. 11. The effective height of each large height protrusion relative to the effective wall position established by the small protrusion immediately adjacent thereto is at least three wall units. A wall surface as described in any one of Item 10. 12. The wall surface according to any one of claims 1 to 11, wherein the height of the protrusion increases in the flow direction. 13. Claim 10, characterized in that the increase in height is stepwise, and a protrusion of a large height in a pattern preceding one step becomes a protrusion of a low height in a pattern following that step. The wall surface described in section. 14. The protrusion of at least a large height has side surfaces that slope at an angle that varies depending on the height of the protrusion, each side having an intermediate region of smaller slope than its upper and lower adjacent regions, and a concave shape of the associated protrusion; 14. A wall surface according to any one of claims 1 to 13, characterized in that it forms the lower part of a pointed part. 15. A wall surface according to claim 14, characterized in that all projections exceeding the effective height of the wall have side walls of varying slope such that the intermediate region is positioned approximately at the effective height of the wall. . 16. Claims 1 to 15 characterized in that there are flow direction gaps between the lengths of the protrusions, each gap width being less than half the thickness of the boundary layer in the flow direction. A wall surface described in any one of the above.