Liquid jets are widely used in cleaning operations in the food sector. Morison and Thorpe (2002) reported an experimental investigation of the flow patterns and cleaning behaviour of horizontal jets impinging on vertical walls. The Wilson et al. (2012) model, which described Morison and Thorpe's flow pattern data well, is extended to describe the flow pattern generated by a liquid jet, approaching a surface at a given angle to the horizontal, impinging on a plate inclined at a known angle to the vertical. The results are compared with experimental data collected for horizontal water jets impinging on inclined Perspex and glass plates. Tests employed nozzle diameters of 1, 2 and 3 mm at room temperature, using flow rates of 0.78-2.23 g s(-1), 3.7-9.9 gs(-1) and 7.1-17.3 gs(-1) (0.025-0.062 m(3) h(-1)) respectively. These are lower than industrial cleaning flow rates. The angle at which the horizontal jet impinged on the plate was varied from 30 degrees to 120 degrees. Two important dimensions are evaluated: (i) the width of the fast moving radial flow zone on the plate (the region bounded by the film jump, the feature similar to a hydraulic jump) at the plane of impingement; (ii) the distance on the plate to which the radial flow zone extends above the point of impingement. Both are described reasonably well by the model. Empirical relationships are reported for the width of the wetted region at the level of impingement, and the maximum width of the draining film. A short study of cleaning of layers of washable paint on glass, similar to the tests reported by Morison and Thorpe, show that the cleaning model recently developed by Wilson et al. (2014) gives a good description of the initial cleaning of such layers using an impinging stationary coherent water jet. (C) 2014 The Institution of Chemical Engineers. Published by Elsevier B.V. All rights reserved.

The cleaning action of stationary coherent liquid jets impinging (a) vertically downwards on horizontal plates, and (b) horizontally on vertical plates, was investigated using three soft-solid model soil layers: (i) PVA glue on glass and polymethylmethacrylate (Perspex) substrates; (ii) Xanthan gum on stainless steel; and (iii) petroleum jelly on glass. The liquid stream nozzle sizes, mass and volumetric flow rates and mean jet velocities investigated were: PVA, 2mm, 17–50gs−1 (0.06–0.139m3h−1), 5.3–15.9ms−1; Xanthan gum, 0.39–3.3mm, 2.1–148gs−1 (0.008–0.53m3h−1); 4.5–31.7ms−1; petroleum jelly, 2mm, 7.8–50gs−1 (0.06–0.139m3h−1); 2.5–15.9ms−1. For all three soils, rapid initial removal of soil from the jet footprint was followed by the growth of a nearly circular, clean region centred at the point of jet impingement. The rate of removal of soil decreased sharply when the cleaning front reached the hydraulic or film jump. The data for the radial growth removal stage were compared with a mathematical model describing removal of the adhesive soil layer, where the force on the cleaning front was evaluated using the result reported by Wilson et al. (2012): their theory gave the momentum of the liquid film; this momentum was balanced against the soil strength, giving a simple relation between the cleaned radius and time. All three soils showed reasonable agreement with the model, across the range of flow rates and temperatures studied. The kinetic constant in the model was sensitive to soil layer thickness and the nature of the soil. Cleaning tests on the petroleum jelly soils at different temperatures, and separate rheological measurements, showed that the kinetic time constant for coating removal was proportional to the (critical shear stress)−1.8. There was good agreement between results obtained with vertical and horizontal plates for the PVA and Xanthan gum soil layers. The petroleum jelly results differed, which is partly attributed to differences in preparing the layers of this rheologically complex material.

The flow behaviour and wetting patterns generated by a liquid jet impinging on a surface is important for cleaning and coating operations. The flow patterns generated by a static, coherent horizontal jet impinging on a vertical surface were studied for three nozzle sizes (1 mm, 2 mm and 3 mm diameter) on surfaces with different wetting characteristics, namely borosilicate glass and polymethylmethacrylate (Perspex). Experiments were performed using water and three different aqueous solutions of an anionic detergent (Tween 20) across the temperature range 20-60 degrees C. At the flow rates studied, which lie at the lower end of the range employed in industrial static spray ball systems (corresponding to jet mean velocities of 0.8-3.3 m s(-1)), an important series of phenomena are observed.After impingement, the liquid flows radially outwards over the vertical surface. The liquid flowing upwards from the point of impingement forms a feature resembling a hydraulic jump. The location of this film jump is sensitive to surface tension and contact angle at lower flow rates 0.65-10 g s(-1)), but at higher flow rates (greater than 11 g s(-1)) the jump location is insensitive to substrate nature. The model of Wilson et al. (2011), predicting the film jump location, is here modified and gives good agreement with the experimental data at the lower flow rates, using the contact angle for water on each substrate. At the higher flow rates the modified model suggests an effective contact angle of 90 degrees.The draining films exhibit the two forms reported by Wilson et al., namely rivulet and gravity flows, and the transition between these is described well by the criterion of Hartley and Murgatroyd (1964) across the range of temperatures and surfactant concentrations (above and below the critical micelle concentration) studied. Unlike the film jump location, the shape of the draining film was strongly affected by surfactant, which promoted gravity flow. (C) 2012 Elsevier Ltd. All rights reserved.

The flow patterns created by coherent water jets created by solid stream nozzles impinging on vertical polymethylmethacrylate (Perspex) and glass surfaces were studied for nozzles with diameters 2–4mm at angles up to ±45° from the horizontal. The flow rates studied ranged from 7.1 to 133gs−1 (26–480Lh−1; jet velocities 2.6–10.6ms−1). The width and height of the film jump marking the limit of the radial flow zone were compared with models based on that developed by Wilson et al. (2011), modified to include the effect of gravity and the angle of inclination for non-horizontal jets (incorporating the flow distribution model reported by Kate et al. (2007. Journal of Fluid Mechanics 573, 247–263)). The location of the film jump and the flow pattern around the impingement point were sensitive to the nature of the substrate at low flow rates, but insensitive to substrate nature at higher flow rates. The models predicted the film jump location with reasonable accuracy, and the width of the wetted region at the mid-plane was found to follow a simple relationship to the film jump width there. A first-order model for the width of the rope of liquid draining around the film jump gave a lower bound estimate of this dimension. The falling film generated below the impingement point exhibited three forms of behaviour: a wide film, termed gravity flow; a narrowing film, termed rivulet flow, and a wide film which split into two with the formation of a dry patch. The transition to form a dry patch was found to obey the minimum wetting rate criterion reported by Hartley and Murgatroyd (1964), once loss of liquid due to splashback was accounted for. Dry patch formation within the falling film was only observed with upwardly impinging jets, and the tendency to form dry patches was predicted with some success by a simple two-stream model.

The shot peen forming process model presented herein can be used to properly simulate the progressive deformed plastic layer , using a notional temperature gradient to appropriate macroscopic deformation. Based on that, a nonlinear relationship between peening parameters specified for each discrete peening area and deflection at node points was set up,and then using a constrained optimization procedure to find an optimal solution subject to specific conditions. This simulation sharply reduces computation expenses,therefore it can quickly estimate peen forming process’s time and result, and find an optimal solution subject to specific conditions.

In the development of a flow test rig as a tool for investigating manufacturing variation, we encountered various problems. We suspect that we were not unique and that our observations would be of interest to others. In this paper we concentrate on the hydraulic engineering aspects of the rig, referring to the uncertainty only as required to understand the discussion. In particular we consider the consequences of selecting a standing start and stop system, and the errors which may result. We review the likely effects of entrained air, and we consider the error which may result from the ramp up and down at the start and end of each run. Further aspects will be discussed in subsequent papers. We aim to provide a source of guidance for manufacturers and others who need to install a calibration rig for their manufacture or research.

A finite element analysis of iuultiple random impacts involved in shot peen forming is presented. An explicit dynamic algorithm for modelling up to 1000 impacts and a static algorithm for spring-back simulatioii are combined to achieve a final curved shape after shot peening on a small sized aluminium 2024-T35 1 sample. The development of the plastic zones, residual stresses, and the final deflection is given by simulating the experimental conditions. The comparison between simulation results and experiments shows that the finite element impact modelling is able to investigate the ~nacroscopic effects (e.g. curvature) of shot pccning as well as the microscopic effects (e.g. local plasticity and residual stresses).

An equivalent loading unit that can produce a plastic layer was used to model the macroscopic forming effects of shot peening. Because a specific plastic coverage exists in which the individual peen impacts can be regarded as acting independently, their cumulative effects can be assumed to be distributed in the plastic layer as the result of a static load. The loading unit corresponds to the macroscopic stretching strain and can be conveniently calibrated from the peening parameters, such as shot radius, mass flow rate and air pressure. For more intensive peening, the number of loading cycles is in direct proportion to the peening time, which makes the model applicable to practical applications. Compared with experiments, this model gives a good prediction of the peen-formed deflection.

A finite element analysis of multiple random impacts involved in shot peen forming is presented. An explicit dynamic algorithm for modelling up to 1000 impacts and a static algorithm for spring-back simulation are combined to achieve a final curved shape after shot peening on a small sized aluminium 2024-T351 sample. The development of the plastic zones, residual stresses, and the final deflection is given by simulating the experimental conditions. The comparison between simulation results and experiments shows that the finite element impact modelling is able to investigate the macroscopic effects (e.g. curvature) of shot peening as well as the microscopic effects (e.g. local plasticity and residual stresses).