RésuméCet article présente une étude expérimentale sur le comportement de l’écoulement radial entre une buse de soufflage et une paroi cylindrique. L’objectif du travail est d’avancer des solutions possibles au problème d’encrassement des buses de soufflage dans le domaine du contrôle dimensionnel pneumatique. Dans le cas des buses utilisées traditionnellement, les résultats indiquent qu’il existe une région dépressionnaire entre la surface frontale de la buse et la paroi cylindrique. Cette dépression est vraisemblablement associée à une zone de recirculation qui serait à l’origine du problème d’encrassement. Les résultats obtenus montrent qu’un simple changement dans la géométrie de la buse de soufflage (i.e. buse chanfreinée) permet l’élimination de la zone dépressionnaire. L’utilisation de telles géométries pourrait logiquement permettre l’élimination des dépôts de saletés sur la surface frontale des buses de soufflage,AbstractThis article presents an experimental investigation on the behaviour of the radial flow between an injection nozzle and a cylindrical wall. The objective of this work is to find possible solutions to nozzle fouling problems found in industrial pneumatic dimensional control applications. In the case of traditional nozzle geometries, results indicate the presence of low-pressure areas in the space between the nozzle and the cylindrical wall. These low-pressure areas seem to indicate the presence of separated flow regions which would likely be the cause of the nozzle fouling problems. Results show that simple changes in nozzle geometry (i.e. chamfered nozzles) can lead to the elimination of the low-pressure areas. The use of such nozzle geometries seems to be a logical solution for the nozzle fouling problems encountered in industrial applications.

This article presents an experimental investigation on the behaviour of the radial flow between an injection nozzle and a cylindrical wall. The objective of this work is to find possible solutions to nozzle fouling problems found in industrial pneumatic dimensional control applications. In the case of traditional nozzle geometries, results indicate the presence of low-pressure areas in the space between the nozzle and the cylindrical wall. These low-pressure areas seem to indicate the presence of separated flow regions which would likely be the cause of the nozzle fouling problems. Results show that simple changes in nozzle geometry (i.e. chamfered nozzles) can lead to the elimination of the low-pressure areas. The use of such nozzle geometries seems to be a logical solution for the nozzle fouling problems encountered in industrial applications.

The application of pneumatic metrology to control dimensional accuracy on machined parts is based on the measurement of gas flow resistance through a restricted section formed by a jet orifice placed at a small distance away from a machined surface. The backpressure, which is sensed and indicated by a pressure gauge, is calibrated to measure dimensional variations. It has been found that in some typical industrial applications, the nozzles are subject to fouling, e.g., dirt and oil deposits accumulate on their frontal areas, thus requiring more frequent calibration of the apparatus for reliable service. In this paper, a numerical and experimental analysis of the flow behavior in the region between an injection nozzle and a flat surface is presented. The analysis is based on the steady-state axisymmetric flow of an incompressible fluid. The governing equations, coupled with the appropriate boundary conditions, are solved using the SIMPLER algorithm. Results have shown that for the standard nozzle geometry used in industrial applications, an annular low-pressure separated flow area was found to exist near the frontal surface of the nozzle. The existence of this area is believed to be the cause of the nozzle fouling problem. A study of various alternate nozzle geometries has shown that this low-pressure recirculation area can be eliminated quite readily. Well-designed chamfered, rounded, and reduced frontal area nozzles have all reduced or eliminated the separated recirculation flow area. It has been noted, however, that rounded nozzles may adversely cause a reduction in apparatus sensitivity.

A feeling of freshness is obtained by ventilators that create a movement in the ambient air. The aim of a ventilation system is to obtain a velocity distribution as uniform as possible on a surface as large as possible. To ensure the "comfort" conditions, the air velocity must be in the range 0.3 to 0.6 m/s. This should be obtained as close to the system as possible, and in a large volume. Using single simple or swirling jets leads to bad velocity distribution on small areas. To ensure large flow rates, it should be necessary to use numerous distributed jets. It is proposed here to study a system that consists of a central swirling jet surrounded by inclined swirling jets. In this study, the influences of the main parameters are presented. These parameters are:the relative rotation directions (central jet versus surrounding jets), the relative flow rates (central jet versus surrounding jets), the difference between the angle of the central jet velocity and the angle of the surrounding jets velocity, the angle between the axis of the central jet and the axis of the surrounding jets, the distance between the system and the roof, the distance between the surrounding jets and the central jet. It is shown that a good combination of these parameters leads to a good velocity distribution.

RésuméUne sensation de fraîcheur s’obtient par des ventilateurs qui assurent à l’air une vitesse favorisant cette sensation. Le but est d’obtenir une vitesse d’air la plus uniforme possible sur une surface importante assurant la notion dite “de confort”, soit par des vitesses d’air comprises entre 0.3 et 0.6 m/s à une distance proche de l’origine du soufflage. L’utilisation d’un jet simple ou même hélicoïdal a pour inconvénient de souffler sur une surface faible et surtout sans uniformité. Pour garantir des débits importants de renouvellement d’air il faudrait multiplier le nombre de ces jets. Le système étudié ici est un système multi-jets qui est constitué d’une couronne de jets hélicoïdaux entourant un jet hélicoïdal central. On étudie dans cet article les principaux paramètres influençant les résultats et qui sont: le sens de rotation des jets les uns par rapport aux autres, le rapport du débit du jet central vis-à-vis du débit des jets latéraux, la différence entre l’angle de la vitesse initiale du jet central et celui des jets latéraux, l’angle entre l’axe du jet central et l’axe des jets latéraux, la distance entre le plafond et l’ensemble du système réalisé, la position relative des jets latéraux par rapport au jet central. Il est montré que le choix judicieux de l’ensemble des paramètres conduit à une bonne homogénéité de soufflage.AbstractA feeling of freshness is obtained by ventilators that create a movement in the ambient air. The aim of a ventilation system is to obtain a velocity distribution as uniform as possible on a surface as large as possible. To ensure the “comfort” conditions, the air velocity must be in the range 0.3 to 0.6 m/s. This should be obtained as close to the system as possible, and in a large volume. Using single simple or swirling jets leads to bad velocity distribution on small areas. To ensure large flow rates, it should be necessary to use numerous distributed jets. It is proposed here to study a system that consists of a central swirling jet surrounded by inclined swirling jets. In this study, the influences of the main parameters are presented. These parameters are:the relative rotation directions (central jet versus surrounding jets), the relative flow rates (central jet versus surrounding jets), the difference between the angle of the central jet velocity and the angle of the surrounding jets velocity, the angle between the axis of the central jet and the axis of the surrounding jets, the distance between the system and the roof, the distance between the surrounding jets and the central jet. It is shown that a good combination of these parameters leads to a good velocity distribution.

The present article analyzes different definitions of the swirl number and their application to control the mixing process in the cylinder of direct injection diesel engines. We first of all discuss the origins of the existing methods used to control swirl, which are usually based on the presence of a stirring device, or rotometer, in the cylinder head. A new possibility is then proposed for determining the swirl number, based on equivalent angular velocity, and a comparison made of results obtained using the different techniques available for measuring swirl in motor cylinders.

Based on the study of gross flow maldistribution in an experimental electrical heater, this paper presents the effect of flow nonuniformity on the performance of heat exchangers. First, it is shown that it is much more important to understand maldistributions for electrical heaters than for two-fluid heat exchangers. The study of the flow distribution in a particular heater shows that reverse flows may occur for poor inlet header design. Suggested here is a simple way to homogenize the flow distribution and a simple law to calculate, with good accuracy, the velocity ratio (ratio of the highest velocity in the tubes to the lowest velocity). The original fluid distribution is applied to heat exchangers (condensers, counterflow and crossflow heat exchangers), and it is shown that gross flow maldistribution leads to a loss of effectiveness of about 7% for condensers and counterflow heat exchangers, and up to 25% for crossflow exchangers, for velocity ratios up to 15.

The basic study of heat exchangers is based on commonly accepted assumptions. Among these, one deals with the fluid distribution in the exchangers; it is assumed that the fluid is well distributed. But it is certain that, depending on the headers design, we can observe maldistributions. This paper presents the study of the fluid distribution in heat exchangers with headers of square cross section. The study is divided in two complementary parts: the first one is an experimental part, the second part is carried out using software. The experimental data, obtained using hot-wire anemometry, allows us to propose a dimensionless analytical expression of the fluid distribution in heat exchangers. Using static pressure probes, we have also been able to put to the fore reverse flows. This happens when the inlet tube diameter is too small compared to the height of the inlet header. In the second part of the study, the experimental data is compared to computed results. This comparison shows that a simple turbulence model gives accurate results when the model implemented in the software is finely meshed, the turbulence model and its parameters are well adjusted, and the solver is adapted. It is shown that two-dimensional simulations are sufficient to help to understand why backward flows may occur, but are not representative of three-dimensional flows. Transactions on Modelling and Simulation vol 21, © 1999 WIT Press, www.witpress.com, ISSN 1743-355X 350 Computational Methods and Experimental Measurements

This paper focuses on the study of flow distribution in electrical heaters and in two-fluid heat exchangers. First, it is shown that it is mon important to understand flow maldistributions for electrical heaters than for two-fluid heat exchangers. The study of the flow distribution in a particular heater shows that backward flows may occur for certain inlet headers. A simple method to homogenize the flow distribution and a simple correlation to predict the amelioration are suggested. The velocity ratio (ratio of the highest velocity in the heater to the lowest velocity), as well as the heater pressure loss, can be calculated with good accuracy. We then apply the original fluid distribution to condensers and counterflow heat exchangers, and show that fluid maldistribution leads to a loss of effectiveness of only 7% for a velocity ratio of 15.

The present paper describes an experimental investigation of the various parameters affecting the operation of industrial pneumatic controllers based on the jet nozzle principle. A test rig was built to monitor supply pressure, air temperature, airflow characteristics, and the static pressure distribution over the flat plate on which the jet impinges. The results demonstrate the existence of a low pressure, separated flow zone, subject to fouling, which subsequently was eliminated by appropriate changes of the injection nozzle geometry. The previous experimental findings were also confirmed by numerical simulation of the flow. Experimental results also show that the internal diameter of the regulator, situated inside the measuring branch, has an important influence on the sensitivity of the apparatus, as well as influencing its range.

An experimental investigation of the frequency response of electrochemical probes to transversal velocity fluctuation is reported. The study has been focused on wall shear stress for a modulated flow due to an oscillating cylinder along its own axes since the wall shear stress has been first numerically determined for this flow. Results indicate that a frequency response correction to probe outputs is required and that this correction depends on the amplitude of fluctuations.

The problem of obtaining a numerical solution for the steady flow between two coaxial infinite disks, one fixed and porous, the other rotating, is reduced by von Kámán's hypothesis to solution of a system of nonlinear equations. A Newton-type iteration results in several solutions to these equations, as a number of authors have already indicated. Nevertheless, an interval in which only one solution is found exists for small values of the Reynolds number based on the angular velocity of the rotating disk, the distance between the disks and the kinematic viscosity of the fluid. At large values of this Reynolds number, two solutions appear and have been the subject of intense controversy.In this paper, both physical and numerical arguments are presented which support a Batchelor-type solution for the flow between infinite disks, in which part of the fluid rotates as a solid body. The other solution, following Stewartson, assumes that the velocity of the fluid outside the boundary layers is entirely axial. This only seems to be verified experimentally when the distance between the disks is large compared with the (finite) radius of the disks.