Int. J. Mech. Sci. Vol. 21, pp. 409--416 Pergamon Press Ltd., 1979. Printed in Great Britain
DAMAGE STUDIES PLASTICINE
B. G. SINGER Mechanical Engineering Department, Sunderland Polytechnic, Sunderland, England
and S. J. HARVEY Combined Engineering Studies Department, Lanchester Polytechnic, Rugby, Warks., England (Received 21 Ju/y 1978; in revised form 27 January 1979)
Snmmery~Two mechanisms, namely, the spherical pressure wave and the microjet, have been used to account for the erosion of materials resulting from the collapse of cavitation bubbles. In recent years, however, high speed photography of collapsing bubbles has added support to the microjet mechanism. Experiments have been undertaken by the authors to examine the mechanism of the erosion of materials subjected to a cavitation environment. Stationary specimens of plasticine held in close proximity to the end of an ultrasonic horn have been damaged by cavitation in distilled water. By virtue of the features of the pits formed, as shown in the photographs in the paper, it is concluded that the cavitation erosion damage results from the impingement of high velocity microjets on the material surface during bubble collapse. INTRODUCTION
The harmful effects of cavitation have been recognised for many years with Rayleigh making one of the earliest studies of the phenomenon. The occurrence of cavitation extends to a wide range of engineering equipment and includes such items as hydraulic machinery, valves, bearings, pumps, engine liners and, nowadays, advanced power systems using liquid metals. One effect of cavitation is the erosion of materials when subjected to a cavitation environment, and it is this area which is the subject of this paper. The problem of erosion is very complex and has been the subject of a great amount of research which is still continuing. Very often in field situations, where other considerations are uppermost, the intensity of the cavitation attack cannot be reduced and consequently the only remedial action to counter the mechanical action of the cavitation is to employ more erosion resistant materials or to use surface coatings. (Chromium coatings on the waterside of Some cast iron diesel engine liners is an example of this.) For the erosion of materials to occur as a result of cavitation, it is now generally accepted that the erosive mechanism is mechanical in nature and capable of being very intense. No material is immune to the erosive influence of the phenomenon provided, of course, that the cavitation action is sufficiently severe. A measure of this severity is the fact that materials such as tool steels with yield and ultimate stresses of the order of 1 GN/m 2 (150,0001bf/in. 2) are susceptible to cavitation damage. Not surprisingly, weaker materials readily succumb to this severe mechanical loading with, in some cases, disastrous consequences. EROSION MECHANISMS Whilst the electro-chemical effects of corrosion will exacerbate the erosive effects of cavitation, the violent collapse of bubbles in close proximity to a material surface results in erosion which is a manifestation of surface loading. Two distinctly separate mechanisms have been suggested to account for this intense mechanical loading; these are: (1) Pressure wave model. This assumed that bubble collapse was spherically symmetric, which when arrested resulted in the propagation of a spherical pressure wave being imposed on any mater/a] surface close to the centre of bubble collapse. (2) Microjet mode/. This did not rely on the assumption of spherical coflapse of the bubble and postulated that dia'ing the non-spher/cal collapse a high velocity microjet was produced which impinged on any tli~teria] surface close to the centre of the bubble collapse. 4O9
B . G . SINGER and S. J. HARVEY
Until a few decades ago, the first mechanism which was given prominence by Rayleigh, was accepted as the more realistic explanation for the erosion of materials. Furthermore, the assumed spherical symmetry throughout the process allowed this model to be amenable to theoretical analysis which, in turn, predicted pressures to the order of 10~ atmospheres. Quite obviously, this order of magnitude was sufficiently high to reconcile this model with the necessary intensity of mechanical loading to promote erosion of materials. However, with the advent of high speed photography it became apparent that an alternative mechanism could apply and the photographic evidence[2-6] available gives weight to the microjet mechanism. Moreover, this evidence indicates that turbulence, pressure gradients and proximity of walls all militate against the possibility of symmetrically spherical collapse necessary for the pressure wave model. The potential for jets of water to erode materials is now well known and research[7-9] in this field adds further credibility to the microjet case. It is the purpose of this paper to describe some experiments using an ultrasonic vibrator to produce cavitation damage of plasticine specimens which gives additional support to the microjet model of cavitation erosion. DESCRIPTION OF APPARATUS Fig. 1 illustrates diagramatically the experimental arrangement used to expose plasticine specimens to a cavitation environment. Th~ main element of the rig was a 150 W ultrasonic drill unit with a stellite tip attached to the free end of the horn driven at 20kHz with a peak-to-peak amplitude of 2 rail (50.8 ftm).
""--,,. PIEZOt ECTRIC / CRYST
C ~ " PLASTICINE DISTILLED WATER
FIG. 1. Diagrammatic arrangement of experimental equipment.
FIG. 2. Details of stationary plasticine holder.
Cavitation damage studies using plasticine
FIG. 3. Typical damage patterns obtained using plasticine: (a) separation distance 45 mil, exposure time 45 sec; (b) separation distance 55 mil, exposure time 30 sec ( × 2.5).
FIG. 4. Damage pattern produced on.~tationary stainless steel specimen after 6 hr exposure to cavitation (x5),
SINGER and S. J. HARVEY
(d) FIG. 5(a)-(d). P h o t o m a c r o g r a p h s of damage areas on plasticine subjected to cavitation ( x 25)
Cavitation damage studies using plasticine
FIG. 6. Photomacrograph of plasticine surface before exposure to cavitation (x50).
B . G . SINGER and S. J. HARVEY
iii~ ~ii ~i I
FIG. 8(a)--(c). Sections through pits formed in plasticine ( x 30)
Cavitation damage studies using plasticine
The acoustic pressure field generated resulted in the formation of a cavitation bubble field in the film separating the tip from the stationary specimen. The separation distance between the tip and the specimen surface was adjusted using feeler gauges before the specimen holder was locked in place. The specimen and the end of the stepped horn were immersed in a beaker containing approx. 1 I. of distilled water at 25°C. Fig. 2 shows how a brass stationary specimen was bored out and filled with plasticine. The excess was removed with a straight edge so that it was flush with the surface and then lightly smoothed with the finger. After the plasticine had been damaged the surface was photographed with 5 x 4 in. plates in association with a projection microscope. For the damage conditions reported the plasticine was not exposed to a cavitation environment for more than 60 sec and in some cases for a much shorter period. Overexposure to cavitation would result in the break-up of the surface of the plasticine with relatively large pieces being dislodged. For this series of tests plasticine was a convenient material insofar that its essentially plastic nature permitted the form of the surface deformation resulting from exposure to cavitation to occur with virtually no elastic recovery. Furthermore, with plasticine being a relatively soft material, the results of any surface loading were likely to be pronounced thus facilitating the examination of the surface. RESULTS AND DISCUSSION Figs. 3(a) and 3(b) are typical damage patterns obtained using plasticine specimens, with deep pits evident both at the periphery of the damage zone and in the central area. At sites such as those marked A, the plasticine has broken away as a result of the cavitation. The distinct ring of pits is characteristic of the erosion patterns normally obtained for the magnitude of the separation between the stationary surface and the vibrating stellite tip used in the tests. Similar erosion patterns have been obtained for metallic specimens and Fig. 4 is for a stainless steel specimen at a separation distance of 35 rail (0.89 ram) after 6 hr exposure to cavitation. Once a eaitt, there is a deep annular ring of erosion at the periphery of the damage zone which indicates that the occurrence of this feature is probably independent of the stationary specimen material properties and is a reflection of the hydro-dynamic regime existing within the film for a particular separation distance. Endo et al. in their paper describing tests with stationary specimens using a fig similar to that of the authors also mentions the formation of a deep ring of pits. However, the reason for this distinct ring of pits is not, as yet, fully understood and will not be discussed further since the overall damage characteristics of materials and the associated erosion patterns are not the subject of this paper. Figs. 5(a)--(d) are photographic enlargements of damage areas similar to those seen in Figs. 3(a) and Co); Fig. 6 is a photograph to the same m _agnificati0n of the plasticine surface before being exposed to cavitation. The dark specks are inclusions which are to be found in plasticine. The reason for including Fig. 6 is to allow a visual comparison to be made between the damaged and undamaged plasticine surfaces. The pits shown in Figs. 5(a)--(d) fall into two categories, namely, those which are deep holes with the depth > diameter, and, those which appear as craters. Some of the former category have been marked B on the photographs whilst some of the latter have been marked C. The form of the " B " pits strongly suggests that they are the result of microjets impin~ng on the surface of the plasticine. From the damaged surface of the plasticine it is not possible to conclude whether the damage has been caused by a single cavitation event or by repeated impact at the same area. Brunton's experiments with single bubble collapse and his studies of bubble collapse in ultrasonic cavitation led to the conclusion that cavitation erosion resulted from repeated collapse of relatively large bubbles anchored to sites on the surface which had been previously indented. From the photographs, the range of hole diameter at the surface associated with the "B" pits was found to be approx. 2--4rail (50-100;tm). Hammitt[ll] has suggested that the diameter of typical microjets resulting from cavitation bubble collapse was in the range 1-25/~m, which is significantly smaller than the hole diameters mentioned above. A possible explanation of this difference may arise from the radial flow of the test liquid on impact with the surface and during the formation of the hole itself as illustrated diagramatically in Fig. 7. However, until further controlled tests are undertaken, it is not possible to draw any definite conclusions about the microjet dimensions from the diameters of the holes formed in the plasticine. The plasticine was frozen and removed from the brass holders and sections were taken through the pits with a sharp razor blade. Figs. 8(a)-(c) are examples of these, which show clearly that the axis of some of these pits is as m~ch as 35° from the vertical. The ragged sides of these holes and the fact that in Figs. 8('o) and (c) the hole has been bridged is probably attributable to the smearing effect of the razor blade cutting through the plasticine. The maximum depth of these holes was of the order of 20 rail (0.5 ram). The depth of these pits and the angle at which they have been formed are factors which are inconsistent with the spherical pressure wave model of erosion although they do strongly support the contention that micr0jets are the dominant erosion mechanism. The diameter of the "C" type pits is in the range of 1-5-8 rail (38-200/tin) and these values are compatible with the pit diameters observed by Knapp using annealed aluminium. Pit diameters of
F~G. 7. Diagrammatic representation of microjet penetration in plasticine. M S Vol. 21. No. 7 - - C
B . G . SINGER and S. J. HARVEY
0.25 mm obtained by Brunton using piasticine covered with a very thin metal foil were of the same order of magnitude although, in general, l~rger than those found by the authors. Compared with the "B" pits, the craters associated with the "C" pits are not, by the .mselves, very conclusive evidence in support of the microjet mechanism. However, the rim associated with the crater C, in Fig. 5(d) is not symmetrical and this suggests that it could well have been formed by a micro jet (and its associated radial flow) impinging at an angle other than 90~ to the surface. Hence it is probable that the craters are formed by microjets which are less intense than those necessary to form the deep holes, if these are formed by a single cavitation event. On the other hand, if they are formed by the repeated impact at the same site then these "C" type pits could be the result of a single impact or a small number of impacts and merely represent the initial stage of the " B " type holes. Nevertheless, in either case, bearing in mind the possible fatigue nature of cavitation erosion of materials, such microjets are likely to contribute to the overall erosion of a material surface. Assuming that the damage is caused by microjets, it is clear from the photographs that the collapse of cavitation bubbles results in a wide range of microjet intensities as described by their diameters, although the threshold level at which a microjet could be classified as being potentially damaging will vary from one material to another. CONCLUSIONS
(1) Whilst the photographic evidence included in this paper does not exclude the possibility of the spherical pressure wave mechanism causing cavitation erosion, it does strongly indicate that microjets are the dominant cause of damage. (2) Using an ultrasonic vibrator, a wide range of microjet intensities result from the collapse of cavitation bubbles generated by the acoustic pressure field.
1. 2. 3. 4. 5. 6. 7. 8. 9. 10. 11. 12.
REFERENCES LORD RAYLEIGH, Phil. Mag. 34, 94 (1917). C. F. NAUDE and A. T. ELLIS, Trans. ASME, J. Bas. Engng 83, 648-56 (1961). T. B. BENJAMINand A. T. ELLIS, Phil. Trans. R. Soc. A260(1110), 221 (1966). N. D. SHtYrLERand R. B. MESLE~, Trans. ASME, J. Bas. Engng ff7D, 511 (1965). C. L. KL~O and F. G. HAMMrrr, Trans. ASME, J. Bas. Engng 825 (1972). J. H. BRUt~3N, Proc. 3rd Int. Conf. Rain Erosion and Allied Phenomena 821--846 (1970). F. P. BOWDgNand J. E. F~LD, Phil. Trans. R. Soc. A282, 331-352 (1964). G. W. VICKERSand W. JOHNSON,Int. J. Mech. Sci. 14, 765-777 (1972). G. HOFF, G. LANOeF.n~and H. REIGER, ASTM-STP No. 408, 42-69 (1966). K. ENDO, T. OKADAand M. NAKASHI~A,ASME Paper 66-Lub 9 (1966). F. G. HAMMITT,Proc. I. Mech. E. 183I, No. 2, 31-50 (1968-69). R. T. KNAPI', Trans. ASME 77, 1045 (1955).