The ricochet of dumb-bell shaped projectiles

Retardations and rotations were measured for five double-ended or dumb-bell shaped (d.s.) projectiles having spheres separated by a light weight rod at specific distances. The projectiles were fired into sand at approximately 100 m/sec and at an impact angle of 10°. Experimental data indicate that for identical sizes of sphere, initial speed, impact angle and projectile mass, the forces of ricochet are independent of the interspherical distance and seem to depend only on the geometry of the leading face.Tests were also carried out with two spherical and five d.s. projectiles at various entry speeds with 5°, 10° and 15° impact angles and it was found that the crater volume in sand is proportional to the initial momentum of the projectile.     

The effect of spherical projectile speed in ricochet off water and sand

Experimental results are reported for the ricochet of steel and duralumin spheres ( and 1 in. dia.), from shallow depths of water and dry sand. The critical angle for ricochet off water is shown to increase with speed to approach the theoretical limit of . For ricochet off sand the critical angle decreases with speed but a cut-off angle exists, , for which no ricochet occurs at any speed.     

The trajectory of a projectile when fired parallel and near to the free surface of a plastic solid

Spherical and elongated projectiles were fired (in the range 150–350 m/sec) into modelling clay (Plasticine) parallel to the surface but from a position below it. The projectiles emerge from the surface if the depth (z₀) to diameter ratio is less than about 1·7. Results for the variation of exit speed, angle, length of plough and angular rotation with initial speed are given. The trajectory is seen to depend strongly on surface pressure.     

Mechanical behaviour of brittle materials under hydrostatic pressure (on torsion of chalk cylinders)

Torsion tests were carried out on black-board chalk under hydrostatic pressures of up to 4·0 kgf/mm² in order to investigate the pressure effect on the deformation and fracture of brittle materials. Chalk, which is very weak (the tensile strength is about 0·1 kgf/mm²), was chosen as test material because the ratio of the applied hydrostatic pressure to the tensile strength of the material could be easily raised to high values.It is shown that there is a distinct difference in the characteristics of deformation and fracture between specimens under a pressure which is higher than a critical value of 0·3 kgf/mm² and those under a pressure less than that value. It is also shown that under relatively high pressure a macroscopic crack is formed around the twist angle yielding the maximum torque and thereafter it propagates helically with increase of twist angle.     

A full factorial investigation of the erosion durability of automotive clearcoats

A full factorial experimental investigation has been carried out into factors affecting the resistance of a commercial acrylic/melamine automotive clearcoat to erosion by silica sand particles. The factor variables and their ranges were: particle size 125–425 μm; temperature 30°C–65°C; impact angle 30°–90°; particle velocity 35 m s⁻¹–55 m s⁻¹; and the baking process applied to the coating. An empirical linear regression model for the erosion response of the coating with R²_adj=97.5% was generated from the data. The regression coefficients of this model quantify the relative strengths of the effects of each of the factors. Several interactions between the factor variables were identified. In particular, the glass transition of the coating, which occurs at 40°C, has a significant effect on its response to erosion. The study has allowed the combinations of conditions that would be of most concern for automotive paint users to be identified.     

A method for achieving effective load transfer between the inner and outer layers of a two-layer braided high-pressure hydraulic hose

A method is proposed for accommodating the misfit between two layers of a braided hose. This is achieved by laying the inner and outer braids at angles that lie symmetrically on either side of the equilibrium helix angle of tan⁻¹ 2 for a single-layer hose. The theory of the method is developed and it is shown that a difference of helix angles of 4° or more will accommodate the fractional radial misfit of 3 × 10⁻³ existing in a particular sample of hose. An analysis of the wire tensions in a pressurised hose, assuming constant inter-layer volume, revealed that seriously unequal tensions exist in the two layers if the two helix angles lie asymmetrically about tan⁻¹ 2. The process of taking up misfit, now proposed, generates asymmetry. However this asymmetry and the associated inequality of wire tension decreases as the difference of braid angles increases. Guided by this generalisation, it is demonstrated that an acceptable hose design is achieved if the inner braid is laid at an angle of 59·74° and the outer braid at 49·74°. The geometrical changes that take up the misfit at low pressures produce an extension of the hose of 0·65% and reduce the braid angles by about . The ratio of final tensions in the two layers is 0·93. The analysis in the paper leads to the conclusion that manufacturing uncertainties in the angle at which a braid is laid can generate serious inequalities of wire tension. In the specific illustrative example analysed here an uncertainty of ±1° in braid angle will cause the ratio of wire tensions to vary, but it never falls below a value of 0·63.     

A test rig for testing radial diffusers

A simple test rig for investigating the performance of vaneless diffusers with diverging walls for centrifugal compressors has been designed, fabricated and satisfactorily tested. The test rig was designed to give values of area ratio from 1·5 to 6 for an included angle of 10°. A uniform flow at the entry of the diffuser had been achieved by the use of a centre body and suction on both the walls. The swirl was zero in all the tests.A maximum Reynolds' number of 2·8 × 10⁵ can be obtained during the tests. Experimental values of the coefficients for losses, the pressure recovery and efficiency were obtained from the measurements of static and stagnation pressures.     

An experimental investigation into load relaxation in aluminium (HE30TB) and mild steel (EN1A)

Short-time room-temperature tensile load relaxation tests were conducted on aluminium alloy (HE30TB) and mild steel (EN1A) specimens using a closed-loop, electrohydraulic servo-controlled testing machine under strain control by means of an extensometer mounted directly on the parallel section of the tested specimens. The relaxation periods (usually 60 sec.) were interruptions at chosen points in constant strain rate tensile loadings. The loading strain rates in the different tests varied from 5 × 10⁻⁴ to 10⁻² sec. The transient relaxation behaviour was investigated for the purpose of testing the applicability of the most widely assumed viscoplastic constitutive models. This was achieved by comparing the plastic strain rate just after the beginning of load relaxation at constant total strain to the plastic strain rate during the tensile loading just before the start of the relaxation interval. All common viscoplastic theories predict that the plastic strain rate ratio should be unity. The experimental results for both materials indicate, however, that the plastic strain rate ratio varies from almost zero for relaxation periods early in the loading, to a maximum of around 0·2 for some relaxation periods beginning at relatively high loads and strains. This agrees with previously reported results on pure aluminium, which is not very rate sensitive, but the results for the more rate-sensitive mild steel may be surprising. Only if the actual relaxation rate drops by a factor of about 100 in 0.2 sec, could the findings of this experimental programme be reconciled with predictions of the usual viscoplastic theories. The experimental programme also included constant strain-rate tests at several rates and jump tests, in which the rate was switched back and forth between 10⁻⁴ and 10⁻²/sec.     

Experimental study on whirling, flying and tilting motions of a 3.5 in. FDB spindle system

This research develops an experimental method to measure the motion of a FDB spindle system with a 3.5 in. disk by using three capacitance probes fixed on the xyz-micrometers, and it shows that a FDB spindle system has whirling, flying and tilting motions. It also shows that the whirling, flying and tilting motion converge very quickly to the steady state at the same time when the rotor reaches the steady-state speed. However, they are quite large even at the steady state when they are compared with the 10 nm flying height of a magnetic head. For the FDB spindle system used in this experiment, the whirl radius and the peak-to-peak variations of flying height and tilting angle at the steady-state speed of 7200 rpm are 0.675 μm, 30 nm and 5.785°×10⁻³, respectively, so that the radial motion of the FDB spindle system exceeds a track pitch of a 3.5 in. HDD with 90,000 TPI.     

The high strain rate compression of 7039 aluminium

Aluminium alloy 7039 was compressed at strain rates between 2 × 10³ and 2·5 × 10⁴s⁻¹ using a modified Hopkinson bar. At strain rates between 2 × 10³ and 1·2 × 10⁴s⁻¹ there was a linear relationship between the flow stress and strain-rate with a slope corresponding to a macroscopic viscosity of 2·9 kPa s. At strain rates between 1·2 and 2·5 × 10⁴s⁻¹ there was a levelling out of the flow stress, but the data was too scattered to give a definite trend. Due to the opposing effects of linear work-hardening and adiabatic heating, at strains above 0·15 the specimens work-softened at a rate inversely proportional to the square root of the strain rate. At the higher strains, specimens cracked along the dominant adiabatic shear band formed during the compression.     

Force, torque and plastic flow analysis in rotary upsetting of ring shaped billets

The orbital upsetting of rings has been analysed for a Mises material by using an upper bound approach. Forging forces, rocking die torques and ring profiles are calculated at each step of the process. Experiments are described in which rings made of 1045 mild steel, 52100 chromium alloyed steel and 316 stainless steel are upset at room temperature on a 1.6 MN rotary press. The main parameters are: orbital oscillation angle, 2°, upper die oscillations, 200 min⁻¹ and lower die speed, 4.10 mm s⁻¹. An experimental rocking die placed on a conventional testing machine has been used for the rotary upsetting of rings made of Plasticine as model material. The simulation parameters are: oscillation, 2°, upper die oscillations, 40 min⁻¹, speed, 98.4 mm min⁻¹. The theoretical values of forging forces, rocking torques, and ring profiles are in keeping with the experiments. So the proposed upper bound approach may be considered as a good model for rotary upsetting.     

High temperature tribology of ceramics

The friction and wear behaviour of self-mated couples of MgO---ZrO₂, Al₂O₃ and two types of SiSiC were studied under dry sliding conditions in a special pin-on-disc high temperature tribometer. The temperature was varied between 25 and 1000°C, and the sliding speed from 0.03 m s⁻¹ to 3 m s⁻¹. The morphology of the worn surfaces was studied by means of SEM, and their phase distribution by X-ray diffraction and TEM analyses. The results show that the wear coefficients of all couples mostly increase with increasing temperature and sliding velocity. The wear of MgO---ZrO₂ is influenced by tribo-induced phase transformations while α-Al₂O₃ retains its original structure for all test conditions. For SiSiC delamination and fatigue of the interface Si/ß-SiC predominate. At higher temperatures and sliding velocities tribo-oxidation is effective. The friction coefficients lie between 0.5 and 1.0 under steady-state conditions but for short test durations lower values can occur. The couple SiSiC/SiSiC has low friction coefficients at low sliding velocities and temperatures, even if the steady-state region is reached.     

Ceramic-ceramic composite materials with improved friction and wear properties

Dry friction and wear tests were performed with self-mated couples of SiC containing 50% TiC, Si₃N₄---BN, SiC---TiB₂ and Si₃N₄ with 32% TiN at room temperature and 400°C or 800°C.Under room temperature conditions, the friction coefficient of the couple SiC---TiC/SiC---TiC is only half of that of the couple SiC/SiC and the wear is one order of magnitude smaller. At 400°C, it exceeds the friction coefficient of SiC/SiC except at the highest sliding velocity of 3 m s⁻¹. At lower sliding velocities the wear coefficient of SiC---TiC/SiC---TiC is lower than that of SiC/SiC.The couple Si₃N₄---TiN/Si₃N₄---TiN exhibits high friction coefficients under all test conditions. At room temperature the wear volume of the self-mated couples of Si₃N₄ and Si₃N₄---TiN after a sliding distance of 1000 m is similar, but Si₃N₄---TiN shows a running-in behaviour. At 800°C the wear coefficient of Si₃N₄---TiN/Si₃N₄---TiN is approximately two orders of magnitude smaller than that of Si₃N₄/Si₃N₄, and equal to those at room temperature. At 22°C the addition of BN reduces the friction of Si₃N₄. The wear coefficient is independent of sliding velocity and the self-mated couples showing running-in. Friction and wear increase with increasing temperature. The wear coefficient of SiC---TiB₂ above 0.5 m s⁻¹ at 400°C is advantageously near 10⁻⁶ mm³ (Nm)⁻¹. With the other test conditions the wear behaviour is similar to SSiC.     

Erosive wear by silica sand on AISI H13 and 4140 steels

The erosive wear behaviour of AISI H13 tool steel and AISI 4140 steel has been investigated in this work using a sand blast-type rig. Samples of six different hardness levels (from annealed to 595 HV) were produced and subsequently tested using silica sand as the erodent material at impact angles ranging from 10° to 90°, air drag pressures of 0.689 and 1.38 bar (10 and 20 psi respectively), impact speeds ranging from 70 to 107 m s⁻¹ and various particle sizes. Results of erosion versus impact angle at different hardness levels showed three distinctive wear regions: (i) for impact angles of 10° and 20°, the amount of wear was higher at lower hardness values; (ii) for impact angles of 30° and 40° no significant changes were found in the amount of wear despite the increase in hardness; (iii) for impact angles of 60°, 75° and 90° the amount of wear was higher for higher hardness levels in the eroded material. Single curves showed typical ductile behaviour of these alloys, a transition towards brittle behaviour for the hardest specimens was also observed due to the formation of adiabatic shear bands. SEM analysis was conducted to identify the erosion mechanisms for each type of behaviour.     

In situ graphite lubrication of metallic sliding electrical contacts

Decoupled graphite lubrication of monolithic silver brushes on a copper rotor was studied in an ambient air environment under current varying from 0 A/cm² to 200 A/cm² using a custom designed electrical contact tribometer. Bifurcation of the positive and negative brush wear rates was observed at a current density of 200 A/cm². Energy dispersive spectroscopy showed transfer of copper from the rotor to the lower wear negative brush. Scanning electron microscopy of worn brush surface cross-sections created by focused ion beam milling revealed a fine-grained metallic layer below the graphite transfer layer on the negative brush surface; no such layer was found on the positive brush surface. At 40 A/cm², steady-state brush wear rates were very low (<10⁻¹¹ m/m). Friction coefficient at steady state was measured to be 0.15 ± 0.02 and was independent of current direction. Using a scanning white light interferometer, the thickness of the graphite transfer layer on the rotor surface was estimated to be 5 μm. Ultimately, the goal is to model lubricant buildup and removal as a competitive rates problem.     

Solid particle erosion of CFRP composite with different laminate orientations

The solid particle erosion behavior of epoxy base unidirectional and multidirectional carbon fiber reinforced plastic composites was investigated. The erosion rates of these composites were evaluated at various impingement angles (15–90°) with a particle velocity of 70 m/s. Irregular SiC particles with an average diameter of 80 μm was used. The dependence of impingement angle on the erosive wear resembled the conventional ductile behavior with maximum erosion rate at 15–30° impingement angle. The erosion rate of unidirectional composites at acute impingement angle was higher for [90] than for [45] and [0] while the difference disappeared at normal impingement angle (90°). On the other hand, the erosion rates of multidirectional laminated composites ([0/90], [45/−45], [90/30/−30] and [0/60/−60]) were not much influenced by the fiber orientation except for 15° impingement angle.     

Accurate conductance measurements of a pinhole orifice using a constant-pressure flowmeter

A pinhole orifice with a known conductance can be used as a secondary flow standard. Commercially available laser-drilled pinhole orifices with diameters ranging from 1.0 μm to 50 μm can have molecular-flow conductances ranging from about 0.1 μL/s to 200 μL/s for N₂ at 23 °C. Gas flows of 10⁻¹¹–10⁻⁶ mol/s can easily be produced by applying an upstream pressure in the range of 1–10⁵ Pa. Accurate measurements of the orifice conductance as a function of pressure are required to use the pinhole orifice as a basis of a flowmeter. We use a constant-pressure flowmeter to make accurate measurements of the conductance of a 20 μm orifice as a function of pressure for gas flows of Ar and N₂ into vacuum. We present results of these conductance measurements for an orifice with a nominal diameter of 20 μm. The N₂ conductance of this orifice ranged from 30 μL/s to 60 μL/s over the range of pressures investigated, and was measured with an uncertainty of better than 0.2% (k = 2) for upstream pressures greater than 10 Pa.     

Brush-on-disc simulation tribotesting of materials for gas turbine-compliant seal components

A brush-on-disc tribometer was used for high-speed, high-temperature sliding experiments on materials simulating conditions in turbine flexible element gas path seals. Of the polymeric, ceramic and metallic material pairs tested only superalloy wires sliding against ceramic surfaces were promising candidates. After extensive running-in, nickel-based wires wore at a rate close to 10⁻⁸ mm³/N m when sliding under moderate loads at speeds to 100 m/s and temperatures to 500°C against a Cr₃C₂/NiCr coated counterface whose wear was also very low. Thus, useful seal life under engine operating conditions may result, depending upon allowable initial and minimum effective brush-rotor interferences.     

Dry sliding wear characteristics of some industrial polymers against steel counterface

In this investigation, the influence of test speed and applied pressure values on the friction and wear behaviour of polyamide 66 (PA 66), polyoxymethylene (POM), ultrahigh molecular weight polyethylene (UHMWPE), 30% glass fibre reinforced polyphenylene-sulfide (PPS+30%GFR) and aliphatic polyketone (APK) polymers were studied. Friction and wear tests of PA 66, POM, UHMWPE, PPS+30%GFR and APK versus AISI D2 steel were carried out at dry condition on a pin-on-disc arrangement. Tribological tests were performed at room temperature at different pressures (0.35–1.05 MPa) and sliding speeds (0.5–2.0 m/s). The results showed that, for all polymers used in this investigation, the coefficient of friction decreases linearly with the increase in pressure. The specific wear rate for UHMWPE, PPS+30%GFR and APK were in the order of 10⁻⁵ mm³/N m, while the wear rate value for PA 66 was in the order of 10⁻⁶ mm³/N m. In addition to this, the wear rate value for POM was in the order of 10⁻³ mm³/N m. Furthermore, as the results of this investigation, the wear rate showed very little sensitivity to the applied pressures and test speed.     

The effect of oil films on the performance of bell provers

Domestic gas meters are tested for accuracy of registration using bell provers. Oil has a tendency to form films on the vertical surfaces of the bell, and this paper investigates its effect on the accuracy of the prover. The geometry of the prover and the manner of its operation are both shown to affect the magnitude of the error. The most important physical property of the oil is the kinematic viscosity, ν, and the error is found to be proportional to (ν). Calculations for an oil having a value for ν of 2·5 × 10⁻⁵m²s⁻¹ indicate a maximum error of about 0·2% under normal operating conditions. The magnitude of the maximum error can be reduced by suitable design of the prover.