Influence of drag-reducing polymers on turbulence: effects of Reynolds number, concentration and mixing

 Measurements of turbulence properties of solutions of polymers have been made over a large range of drag-reduction, in a fully-developed channel flow. At flows close to maximum drag-reduction the Reynolds stresses were approximately zero over the whole cross section of the channel. Added mean polymer stresses were observed in the viscous wall region for small drag-reduction and over the whole cross-section for large drag-reduction. Even though the Reynolds stresses are zero, the velocity profile is not parabolic because of the presence of these mean stresses. We interpret the results by arguing that the interaction of turbulence with the polymers introduces mean and fluctuating polymer stresses which can create turbulence. The observation that the turbulence modification depends on the manner by which the polymers are introduced into the flow supports the notion that the polymers need to form aggregates in order to be effective. Received: 29 September 1998 / Accepted: 10 February 1999     

Design of Beams Subjected to Random Loads

Nonuniform Timoshenko beams subjected to a given stationary random excitation are considered. The general equations relating the spectral density function of the response to the cross spectral density of the load are derived. The optimal shape of the beam is defined as the shape which, for given constant volume of the beam, minimizes the maximum root-mean-square value of the bending stresses in the beam. The shape of the beam is described by a limited number of orthogonal design functions, and their optimal combination is found by sequential linear programming with move limits. From numerical results it is seen that slight modifications of the beam shape give a considerable reduction of maximum r.m.s. stress for most loading cases.     

Stress separation in the photoelastic study of centrifugal stresses

This technical note refers to the problem of stress separation in the photoelastic analysis of plane models under centrifugal stresses. Two methods are described in order to determine the sum of principal stresses. These methods, which are based on the compatibility equation, reduce the determination of the sum of principal stresses to the solution of a Laplace's or Poisson's equations. As an example of application, the separation of stresses in a rotating disk with two eccentric holes is shown and comparison with the stresses obtained by using the shear-difference method is made.     

Three-dimensional photoelastic analysis of a fiber-reinforced composite model

A three-dimensional photoelastic analysis using the “stress-freezing” technique was conducted to determine the stress distributions in the matrix of a unidirectionally fiber-reinforced composite model subjected to matrix shrinkage and normal transverse loading. The model, consisting of a square array of polycarbonate rods in an epoxy matrix, simulated a boron-filament-reinforced plastic composite with a fiber-volume fraction of 0.50 at the critical temperature of the matrix epoxy. The effects of matrix shrinkage were separated from those of external loading by analyzing two identical models, one loaded and the other unloaded. The Lamé-Maxwell equations of equilibrium were used to separate stresses along axes of symmetry on interior transverse slices. Axial stress components were obtained by subslicing. Results are presented in dimensionless form by dividing the stresses by the average stress through the section. A comparison with theoretical results for a boron-epoxy composite shows excellent agreement, although Poisson's ratio of the model matrix is appreciably different from that of the prototype (0.5 compared to 0.35). One significant result was that the maximum stress occurs in the middle of the matrix section between fibers which is at variance with the theoretical prediction of maximum stress at the interface. Stress-concentration factors vary from 1.80 at the interface to 2.0 at the midpoint of the matrix section between fibers.     

Engineering Models for Numerical Analysis of Composite Bending Members∗

Abstract

ABSTRACT The two-step numerical analysis of a composite beam structure is presented in this paper. The first step, based on the idea of dividing the cross section into laminas, leads to the estimation of the moment-curvature relation for different types of cross sections used in composite beams. The second step adopts this constitutive relation, which is expressed in the space of generalized stresses and strains, into finite element nonlinear code. Some numerical examples are given, to show the agreement of numerical calculations with results of the authors' experiments, when the shrinkage of a concrete encasement and stresses due to welding processes in steel beams are considered. In addition, the numerical concept presented here seems to reduce the sensitivity of the final results obtained to finite element discretization error.     

Photoelastic investigation of a channel cover of bleed condenser

This paper describes the investigation on an epoxy model of a channel cover of bleed condenser, by three-dimensional photoelasticity. The channel cover is a thick plate with two nozzles which are connected on one side to pipes carrying water vapor/water under high pressure. A flat tube plate is connected on the other side by means of sixteen bolts around the periphery with high initial bolt forces. The photoelastic model of the channel cover was made by machining a cast epoxy block (Araldite CT200). The stress distribution along critical sections of the model and the regions of maximum stresses and their values were determined. The prototype stresses were calculated from experimentally obtained values of model stresses, using simiilitude laws.     

Analytical modelling of thermal stresses in anisotropic composites

This paper represents a continuation of the author's previous work which deals with an analytical model of thermal stresses which originate during a cooling process of an anisotropic solid elastic continuum. This continuum consists of anisotropic spherical particles which are periodically distributed in an anisotropic infinite matrix. The infinite matrix is imaginarily divided into identical cubic cells with central particles. This multi-particle–matrix system represents a model system which is applicable to two-component materials of the precipitate–matrix type. The thermal stresses, which originate due to different thermal expansion coefficients of components of the model system, are determined within the cubic cell. The analytical modelling results from fundamental equations of continuum mechanics for solid elastic continuum (Cauchy's, compatibility and equilibrium equations, Hooke's law). This paper presents suitable mathematical procedures which are applied to the fundamental equations. These mathematical procedures lead to such final formulae for the thermal stresses which are relatively simple in comparison with the final formulae presented in the author's previous work which are extremely extensive. Using these new final formulae, the numerical determination of the thermal stresses in real two-component materials with anisotropic components is not time-consuming.     

Measurements of mean and fluctuating wall shear stress beneath spanwise-invariant separation bubbles

 Pulsed-wire measurements of wall shear stress have been made beneath two separation bubbles. In one a cross flow was generated by means of a (25°) swept separation line. Fluctuating stresses in orthogonal “streamwise” and cross-flow directions are very nearly equal and independent of at least moderate cross flow velocity. These fluctuations are largely determined by large-scale motions in the outer flow, whereas the mean shear stresses are not. The pdf of the “streamwise” fluctuations is unchanged by the cross flow. When a cross flow is present the pdf of the cross-flow stresses is similar to the “streamwise” pdf. Dependence on Reynolds number is the same in both flows. Received: 10 April 1998/Accepted: 17 July 1998     

Measuring Inaccessible Residual Stresses Using Multiple Methods and Superposition

The traditional contour method maps a single component of residual stress by cutting a body carefully in two and measuring the contour of the cut surface. The cut also exposes previously inaccessible regions of the body to residual stress measurement using a variety of other techniques, but the stresses have been changed by the relaxation after cutting. In this paper, it is shown that superposition of stresses measured post-cutting with results from the contour method analysis can determine the original (pre-cut) residual stresses. The general superposition theory using Bueckner’s principle is developed and limitations are discussed. The procedure is experimentally demonstrated by determining the triaxial residual stress state on a cross section plane. The 2024-T351 aluminum alloy test specimen was a disk plastically indented to produce multiaxial residual stresses. After cutting the disk in half, the stresses on the cut surface of one half were determined with X-ray diffraction and with hole drilling on the other half. To determine the original residual stresses, the measured surface stresses were superimposed with the change stress calculated by the contour method. Within uncertainty, the results agreed with neutron diffraction measurements taken on an uncut disk.     

Stress dependence of cross slip energy barrier for face-centered cubic nickel

The energy barrier for the cross slip of screw dislocations in face-centered cubic (FCC) nickel as a function of multiple stress components is predicted by both continuum line tension and discrete atomistic models. Contrary to Escaig's claim that the Schmid stress component has a negligible effect on the energy barrier, we find that the line tension model, when solved numerically, predicts comparable effects from the Schmid stress and the Escaig stress on the cross slip plane. When the line tension model is compared against an atomistic model for FCC nickel, a good agreement is found for the effect of the Escaig stress on the glide plane. However, the atomistic model predicts a stronger effect than the line tension model for the two stress components on the cross slip plane. This discrepancy is larger at higher stresses and is also more severe for the Escaig stress component than for the Schmid stress component.     

Measuring Multiple Residual-Stress Components using the Contour Method and Multiple Cuts

The conventional contour method determines one component of residual stress over the cross section of a part. The part is cut into two, the contour (topographic shape) of the exposed surface is measured, and Bueckner’s superposition principle is analytically applied to calculate stresses. In this paper, the contour method is extended to the measurement of multiple residual-stress components by making multiple cuts with subsequent applications of superposition. The theory and limitations are described. The theory is experimentally tested on a 316L stainless steel disk with residual stresses induced by plastically indenting the central portion of the disk. The multiple-cut contour method results agree very well with independent measurements using neutron diffraction and with a computational, finite-element model of the indentation process.     

Nonlinear deformation and buckling of curvilinear pipes loaded by external pressure

The problem of loading of a thin-walled elastic pipe (a toroidal shell) by external pressure is examined in a geometrically nonlinear formulation. A numerical algorithm is used to study the nonlinear deformation of the shell and the stability of its equilibrium states when its cross section has undergone a significant change in shape. Results are presented from a determination of the critical stresses of curvilinear pipes with allowance for moments in the subcritical state. These results are compared with the approximate solution. Chaplygin Siberian Aviation Institute, Novosibirsk 630051. Translated from Prikladnaya Mekhanika i Tekhnicheskaya Fizika, Vol. 39, No. 4, pp. 162–166, July–August, 1998.     

Experimental study and mathematical simulation of the characteristics of a turbulent flow in a straight circular pipe rotating about its longitudinal axis

The vorticity formed in the cross section of a turbulent flow in a straight circular pipe rotating about its longitudinal axis decreases the values of the turbulent stresses, turbulence energy, and dissipation rate along the pipe. The results of laboratory experiments and calculations by the second-order closure model of turbulent transfer are presented. On the whole, the model using a system of transport equations yields better agreement with experimental data than the models with algebraic relations for second-order moments. Translated from Prikladnaya Mekhanika i Tekhnicheskaya Fizika, Vol. 39, No. 2, pp. 103–116, March–April, 1998.     

The Compatibility Constraint in Linear Elasticity

It is shown that the St.Venant–Beltrami equation of compatibility can be regarded as an internal constraint on the admissible strains, a constraint maintained by reactive stress fields that are collectively characterized in a global manner. It is also shown that such reactive stresses have an important role in a formal integration of the mixed boundary-value problem of linear elasticity. The main tools – Beltrami's map, Donati's theorem, and Cesàro's formula – have been available for a century or so; they are here used in a form as close to original as possible.     

Measurement of thermal stresses in the S.S. “Boulder Victory”

A lack of comprehensive experimental measurements of thermal stresses induced in a ship's hull structure by diurnal temperature changes prompted this study. Its essential purpose was to provide reliable prototype measurements of thermal-stress patterns around a complete transverse section of a ship. These results are compared with stresses computed by the theory of simple beams under arbitrary temperature distributions across their section. Bridge-type SR-4 strain-gage assemblies were developed for attachment directly to the structure. A variety of temperature conditions were observed and corresponding strain measurements taken. The results are consistent and give a reliable picture of thermal-stress conditions in a ship. They also verify the prediction of thermal stresses afforded by the simple-beam theory.     

Incremental deformation analysis of shell and corrugated diaphragm based on arbitrary configuration

With respect to an arbitrary configuration of a deformed structure, two sets of incremental equations are proposed for the deformation analysis of revolution shells and diaphragms loaded by both lateral pressures and the initial stresses produced in manufacturing. These general equations can be reduced to the simplified Koiter's Reissner-Meissner-Reissner (RMR) equations and the simplified Reissner's equations, when the initial stresses are set to zero. They can also be deduced to the total Lagrange form or the updated Lagrange form, respectively, as the structure is specified as the un-deformed or the former-deformed configurations. These incremental equations can be easily transformed into finite difference forms and solved by common numerical solvers of ordinary differential equations. Some numerical examples are presented to show the applications of the incremental equations to the deep shell of revolution and the corrugated diaphragms used in microelectronical mechanical system (MEMS). The results are in good agreement with those from finite element method (FEM).     

Fatigue-crack-tip plastic strains by the stereoimaging technique

The stereoimaging technique is an accurate, high-resolution means of measuring the in-plane displacements resulting from the deformation of a specimen so that the corresponding components of the strain tensor can be computed independently of the stresses. The example used in this paper is a fatigue-cracked specimen of a microscopically homogeneous experimental powder-metallurgy aluminum alloy, analyzed to determine the displacement and strain fields accompanying the opening of the fatigue crack. The displacement measurements are processed by a computer program which compensates for measurement fluctuations in the displacement data by smoothing, and derives the strain magnitudes. The principal strains and the maximum shear strain are determined using Mohr's circle, and the latter strain is then used to estimate the plastic-zone size. The crack-opening mode may be inferred from the displacement map, and the state of stress (plane stress or plane strain) inferred by applying the in-plane compatibility equation.     

Experimental methods of X-ray stress analysis

A technique is described for measuring residual stresses in steels and composite materials by X-ray diffraction. Specimen preparation, X-ray diffractometer alignment, diffraction-peak location, and the determination of the elastic modulus, Poisson's ratio and stress factor are covered. Application examples include measurement of heat-treating and shot-peening stresses in steels and grinding and temperature stresses in WC-Co composites.     

Three-dimensional Green's functions in anisotropic trimaterials

Three-dimensional elastostatic Green's functions in anisotropic trimaterials are derived, for the first time, by applying the generalized Stroh's formalism and Fourier transforms. The Green's functions are expressed as a series summation with the first term corresponding to the full-space solution and other terms to the image solutions due to the interfaces. The most remarkable feature of the present solution is that the image solutions can be expressed by a simple line integral over a finite interval [0,2π]. By partitioning the trimaterial Green's function into a full-space solution and a complementary part, the line integral involves only regular functions if the singularity is within one of the three materials, being treated analytically owning to the explicit expression of the full-space solution. When the singularity is on the interface, which occurs if the field and source points are both on the same interface, the involved singularity is handled with the interfacial Green's functions.A numerical example is presented for a trimaterial system made of two anisotropic half spaces bonded perfectly by an isotropic adhesive layer, showing clearly the effect of material layering on the Green's displacements and stresses. Furthermore, by comparing the present Green's solution to the direct (two-dimensional) 2D integral expression which is also derived in this paper, it is shown that, the computational time for the calculation of the Green's function can be substantially reduced using the present solution, instead of the direct 2D integral method.     

The warping functions of regular prismatic bars under torsion evaluated by caustics

Reflection of a bundle of coherent light on the warped cross section of a prismatic bar submitted to torsion forms a caustic on a receiver plane. From the mathematical expression of this curve and the theory of reflected caustics, it is possible to evaluate accurately the warping function of the cross section. Using this idea, it was possible to study the torsion problem in prismatic bars with sections which were equilateral triangles and squares. It was observed that the shape of the caustic is an hypocycloid curve with three or four cusps respectively. By evaluating the warping function by using elements from the respective caustics it was possible to find out that, for the triangular cross section, the expression for the warping function coincided exactly with the expression given by the exact solution of the problem. For the square cross section, a closed-form solution for its warping function was readily derived, to which the series approximation solution differed only by a few percent at maximum for the shear stresses. Since the method can be readily extended to any canonical polygonic cross section, it constitutes a general solution for the torsion of prismatic bars, which approximates their exact deformations better than the solutions based on the Saint-Vénant assumptions.