US2883549A - Driven penetrant method of flaw detection - Google Patents

Description

United States Patent DRIVEN PENETRANT METHOD OF FLAW DETECTION Carl E. Betz', Chicago, 111., assignor, by mesne assignments, to Switzer Brothers, Inc., Cleveland, Ohio, a corporation of Ohio No Drawing. Application January 31, 1956 Serial No. 562,575

Claims. (Cl. 250-71) Although the v methods are limited to the detection of cracks, tears,

blow-holes, laps, and like flaws having surface openings,

i.e., surface discontinuities, the methods are generally relatively sensitive and convenient to practice. The reason for the increasing use of the penetrant methods of inspection is that a minute surface discontinuity, especially at or near areas of maximum stress, is often likely to be the source of failure in service whereas purely internal flaws, unless quite gross or in a part designed with a very small factor of safety, may be relatively harmless.

The penetrant methods of inspection, in general, comprise the steps of applying a penetrant liquid to the surface of the test body, allowing portions of the penetrant to enter the flaws, removing the remaining penetrant from the surface of the body, allowing the penetrant retained in the flaws to appear at or in flaw openings, and then inspecting the body for the indication made on the surface by the retained penetrant. Where the penetrant is fluorescent and inspection is carried out I under fluorescigenous radiations (e.g., black light) in accordance with the aforesaid Switzer process, very minute flaws are readily located. Since the introduction of the Switzer method, various and significant improvements have been made, as in methods of removing the penetrant from the surface of the article without removing appreciable quantities from the flaws, in aiding the development of the flaw indication, increasing the fluorescent efficiency of the penetrants, and the like. Also, various techniques in removing the penetrant from the surface and developing subsequent flaw indications at or around the flaw openings have permitted the use of non-fluorescent penetrants whose indications are observable in ordinary visible light.

With relatively few exceptions, such as in the U.S. patent to Richard A. Ward, No. 2,405,078, for Methods and Compositions for Locating Surface Discontinuities, which teaches the art that a proper sequence of heating and cooling the test body can aid the actual penetration of the flaws by the penetrant, those skilled in the art may seem to have generally overlooked the fact that, regardless of how the characteristics of the penetrant may be improved, or how precisely the penetrant may be removed from the surface of the test body without removal from the flaws, or how effectively the indication may be developed, the success of the operation Of any of the penetrant methods of flaw detection depends upon getting the penetrant into the flaws and subsequently having retained penetrant appear .at least at, and generally preferably around, the flaw opening.

Actually, the interest, in and development of aspects of penetrant inspection methods other than procedures to promote penetration and expulsion of the penetrant has not been due to an oversight by the art. Rather, heretofore, there did not appear to be much of anything which could be done to improve penetration and expulsion of the penetrant beyond careful formulation of the penetrant to obtain optimum penetrability with respect to the materials of the bodies to be tested and, of course, to thoroughly clean the test body of extraneous matter which might clog the flaw openings, as taught by the above Switzer and Ward patents. Such apparent expedients as application of pressure, lowering the viscosity of the penetrant with solvent, et cetera, made no appreciable improvement in penetration or expulsion of the penetrant. In the first place, lowering viscosity (10% not necessarily improve penetrability; in the second place, the surface areas of the flaw openings to be located are usually very small; thus, enormous pressures, in terms of pounds per square inch, become, at the flaw openings, very insignificant forces, especially in the case of flaws having very fine surface openings and very small volume, say, an incipient fatigue crack, a forging lap, or a cold shut.

In such tightly closed, small volume flaws (effectively measurable in two dimensions only, length and depth), surface-active forces between the penetrant liquid and the walls of the flaws would appear to account for penetration, rather than any mechanical force on the penetrant liquid, such as a pressure differential. At least this explanation is fairly well accepted in the art and may well be correct. The infiectiveness of pressure differentials in accelerating penetration of ultra-fine voids is borne out by leak detection procedures for treating vessel walls for leaks through the walls. Starting with leaks of somewhat less than pin-hole size through an aluminum gasoline tank wall of about 16 gauge metal, for example, the effect of pressure upon the rate of penetration through the wall becomes progressively less as the distance between opposite walls of the leak become less until the leak becomes an ultra-fine void, when pressure differentials have no discernible effect whatever on the rate of penetration of the leak. The ineffectiveness of mechanical forces or pressure differentials on the rate of penetration seem to appear particularly pronounced in the case of cracks of the type under consideration here, i.e., ultrafine voids having only one surface opening and a closed end. In the penetration of a leak, i.e., an open-ended void, if any fluid already in the leak is displaced by a penetrating liquid entering at one end, at least such displaced fluid may escape from the other; but, in a closed end void, no such escape seems possible, and the compression of any such fluid already within the void would appear to promptly counteract any mechanical force acting on the penetrating liquid.

Although the ultrasonic vibration of liquid is regarded as essentially the application of high frequency alternating mechanical forces to create standing waves in the body of the liquid, with the resultant creation of alternating pressure differentials throughout the body of the liqhid, it is the object of this invention to improve penetrant inspection methods, and particularly the penetration of fine, tightly closed, small volume cracks and the like, by the application of ultrasonic vibrations to applied to the surface of the test body.

Due to the difliculty of ascertaining the existence of such very fine cracks except by penetrant inspection methods, it may be that the advantage of this invention is penetrants that it improves the sensitivity of penetrant inspection methods, particularly fluorescent penetrant inspection methods, permitting the detection of finer flaws than possible without the improvement afforded by this invention Actual experiments indicate, however, that, contrary to the expected ineffectiveness of mechanical forces on the rate of penetration of ultra-fine voids, the actual advan tage of this invention may be that it greatly accelerates the rate of penetration by the penetrant fluid. Also, contrary to prior experiences with the effect of the application of mechanical forces or pressures upon the rate or degree of penetration of voids by penetrating liquids, the advantages of this invention appear to decrease as the width of the crack, i.e., the distance between opposite walls, increases.

Other objects and advantages of this invention will be apparent from the following general and detailed disclosure, including the appended claims.

Since a penetrant method of inspection functions by the appearance or reappearance of retained penetrant at or around the flaw openings after penetrant has been applied to the body and removed from the unflawed surface, a fundamental requirement for the success of such a method manifestly is that some of the penetrant enter the flaw while the penetrant is applied to the body. This now simple and manifest fundamental has long been recognized by the art to the extent of recommending that, prior to application of the penetrant by coating the body or immersing it in the penetrant, the body be cleaned to remove any oils, tars, or other residues of manufacture or use which might cover or fill (partly or completely) the flaws and thereby block the essential penetration of the flaws through their surface openings.

Accordingly, the art has practiced various methods of cleaning the test body prior to application of the penetrant, as disclosed in the above Switzer patent or Ward patent, or by other processes used for cleaning parts during manufacturing and maintenance operations, such as vapor degreasing, ultrasonic cleaning, etc. Except as the effect of heating the body in a cleaning bath may become a step in the desirable sequence of heating and cooling steps, as disclosed in said Ward patent, the particular cleaning procedure employed prior to application of the penetrant ofiered no special advantage in subsequent penetrant inspection procedures; any advantages were simply those known to be obtained if the part were simply being cleaned to remove surface foreign matter, either merely for appearance or to avoid impairment of subsequent manufacturing operations. Thus, for example, in solvent cleaning parts by either the vapor degreasing or ultrasonic cleaning procedures mentioned above, one might use the same solvent in a vapor degreaser or in a tank in which the solvent is vibrated or driven ultransonically. If the part were relatively large, say, an aircraft engine connecting rod being cleaned for a maintenance inspection, the part could probably be cleaned more quickly and economically in a vapor degreaser than in an ultrasonic cleaning bath, although in time the latter would generally clean the part just as effectively. On the other hand, if the part were relatively small, such as a watch or fuse part preferably'to be cleaned en masse with a great number of like parts, ultrasonic cleaning would generally be quicker, although such parts could, with greater handling care and more time, probably be cleaned as effectively in a vapor degreaser. Aside from such considerations applicable to cleaning a part for any purpose, ultrasonic cleaning appeared to offer no apparent advantage, in preparing a part for a penetrant method of inspection, over other cleaning methods which could clean as effectively. That is to say, when the surfaces of a part appeared to be cleaned as effectively by one method as another, ultrasonic cleaning appeared to be no more eifective in removing possible residue or foreign matter which could close the surface opening of a flaw and thereby prevent successful detection of the flaws by a penetrant method of inspection. This apparent result was not unexpected in view of (a) the general assumption that ultrasonic cleaning functions because of the mechanical force by which the waves in the ultrasonically vibrated fluid agitate to force matter away from the surface being cleaned and thereafter suspend it in the fluid and (b) the presumably known ineffectiveness of mechanical forces or pressure differentials in accelerating penetration of fine flaws by liquid penetrants.

Nonetheless, because ultransonic vibrations also frequently change or appear to change the surface tension and viscosity of fluids subjected to them, the thought occurred that the ultrasonically-vibrated cleaning fluids might actually be entering the surface openings of fine cracks and displacing fluids or foreign matter from them. Once being penetrated into the cracks and having displaced foreign matter, they were retained themselves as foreign matter, blocking subsequent penetration by customary penetrants employed in penetrant inspection methods. Fluids customarily used for ultrasonic cleaning are light hydrocarbons or light chlorinated hydrocarbons which exhibit a substantial tendency to wet surfaces of the parts being cleaned. Although such cleaning fluids are normally considered relatively volatile, the surfaceactive forces on such liquids within the confines of a very fine crack might seem to prevent such fluids from escaping from the cracks by vaporization. Also, in extremely fine cracks, gases which form a film on and wet the walls of the cracks might constitute as much of an impediment to penetration by penetrating liquids as more viscous liquids. Thus, the vapors of the liquids normally used in ultrasonic cleaning procedures, if entrapped in the cracks, might constitute an impediment to penetration if the cleaning liquids themselves did not do so.

The thought then occurred that if the foregoing concepts were correct, many preferred penetrant liquids designed for penetrant inspection methods could be used as fluids for cleaning the flaws ultrasonically, which cleaning would simultaneously serve the purposes of both displacing foreign matter from the flaws and driving the flaw-revealing penetrant into the flaws. Most preferred penetrants constitute a light hydrocarbon, such as kerosene, a dissolved emulsifying agent, and a fluorescent dye dissolved in the kerosene and emulsifying agent. The emulsifying agent is provided primarily for the purpose of rendering the penetrant self-emulsifiable and thus readily removable from the surface of the test body by water. But such an emulsifying agent is also usually a surface active agent and could overcome the effect of any aqueous foreign matter in the flaws which might tend to repel displacement by the oily penetrant. In short, for purposes of penetrant inspection, it was conceived that, for cleaning fine flaws, a preferred type of pene' trant might serve as Well as or better than the customarily employed cleaning liquids used in ultrasonic cleaning.

The foregoing concepts are apparently correct, as may be demonstrated on test pieces having artificially created tightly closed cracks. A rod or bar of wire-drawing stock was grooved to provide a deep V-shaped groove extend ing toward its center. The bar was then drawn through successive wire-drawing dies to reduce the diameter of the bar and to close the groove into a more tightly closed crack after each draw. Following each such successive draw, a length of wire was cut off, and a narrow flat was ground down the length at the known location of the crack; this was done to eliminate any possible misleading irregularities at or adjacent the surface opening of the artificial crack due to burring, peening, or scratching caused by the die. With the length thus prepared, it was cut into still shorter lengths to provide samples for comparative tests. One sample, a control, was separately cleaned and inspected according to the preferred procedure disclosed in the above Ward patent. The other sample, the test piece, without preliminary cleansing other thanwiping, was placed in a tank of commercial fluorescent penetrant of the type disclosed in said Ward patent. The penetrant was vibrated ultrasonically at 40 kilocycles per second by a transducer immersed in the liquid. After a five-minute immersion in the ultrasonically vibrated penetrant, the test piece was removed, its surface was washed clean of fluorescent penetrant, and the test piece was then allowed to stand for the same period of time as the sample subjected to the control procedure. Thus, any dilference in crack indications between the test piece and control would be attributable only to the eifect of ultrasonically vibrating the pentrant liquid. The control and test piece were then inspected under black light for the indication of the crack.

Following the first draw, the crack was clearly visible under overhead visible lighting. The control piece had been immersed in penetrant for the same time as the test piece. No improvement in the indication was observable. Following the second draw, the crack was discernible under oblique light as an apparent fine scratch on the surface of the wire. A ten-minute immersion of the control piece was necessary to secure a satisfactory indication of the flaw, whereas, after an immersion of only five minutes, the test piece gave a very satisfactory indication of the flaw, indicating that the ultrasonic vibration of the penetrant had accelerated the penetration of the flaw. Following a third draw and further reduction and closing of the crack, a thirty-minute immersion of the control piece and five-minute immersion of the test piece both produced faint indications of the flaw, which had not been visible under direct or oblique white light. This further indicated an acceleration of penetration when the fluorescent penetrant was ultrasonically vibrated and also that the third drawing was nearly completely closing the crack.

The above test was re-run, but with the penetrant ultrasonically vibrated at 150 kc./sec., with the same results except that, following the third draw, no indication was noted on either the test piece or the control. Fresh samples of wire cut from the third draw where then retested at 40 kc./sec., and no indications were evident in either the controls or test pieces, even after longer immersion. Still other fresh samples of the wire subjected to the third draw were tested at 250 kc./sec., and no indications were observable, apparently regardless of the time of immersion in the ultrasonically-driven penetrant. However, employing samples taken from the second draw, acceleration of penetration was noted at this last and highest frequency, but to no apparent degree over that obtained at lower ultrasonic frequencies.

The failures to achieve indications in the samples taken from the wire subjected to the third draw (excepting in the first instance) are not attributed to the failure of the process to accelerate penetration where penetration is possible. Rather, it is suspected that, except for the portion from which were taken the samples used in the tests which did give indications, the third draw efiected a cold weld or peening which completely shut the crack, at least to a point below the grind. The samples which did give the faint indications were apparently just under a critical .point at which a crack with a surface opening could no longer be said to exist; any number of factors in the conditions of the die or localized conditions in the wire could have caused these slight difierences.

In production testing requiring the utmost sensltivity, the size of the part to be tested appears to be limited only by the capacity of the tank and transducers available for vibrating the penetrant ultrasonically. Whether the penetrant is vibrated by immersed transducers or transducers carried in or on the walls of the tank is an obvious matter of choice. Further, in production testing, it is generally advisable to first subject the test bodies to any suitable cleaning process to remove excessive surface soil before immersing the bodies in the ultrasonically-vibrated liquid penetrant in order to minimize contamination of the penetrant. While the process does appear to cause the penetrant to effect the ultimate cleaning of foreign matters from flaws and flaw openings, this ultrasonic cleansing by the penetrant is only incidental to the accelerated penetration obtained. And while acceleration of penetration is important to assure that optimum sensitivity is obtained during the period of immersion, the process will prove practical where such ultimate sensitivity is not necessarily required. By accelerating penetration, not only is the time for the entire procedure shortened but, for a given capacity of equipment, a correspondingly greater number of parts may be inspected, with the consequent saving in the bulk of equipment and floor space required.

While the specific examples given above employed a penetrant containing a fluorescent dye to render the penetrant highly fluorescent under black light, or to enhance any natural fluorescence of the penetrant, it is to he understood that in some instances, it may be desirable to use non-fluorescent penetrants which provide merely a color contrast between the surface and penetrant appearing at and around the flaw opening under ordinary visible light or visible light of wavelengths selected to enhance such contrast. Unless otherwise specified, in specific claims, this invention is not to be construed to be restricted to the use of fluorescent penetrants.

This invention is not limited to the specific examples set forth above, either as to parts tested, penetrant materials employed, or equipment used. Rather, it is limited only by the scope of the following claims.

What is claimed is:

1. The pentrant method of inspecting test bodies for flaws having surface openings comprising the steps of applying a penetrant liquid to the surface of the test body, subjecting the liquid to ultrasonic vibration during the application of said liquid to the surface of said body to aid the penetration of said liquid into any such flaws through the surface openings thereof and to cleanse said flaw openings of removable foreign matter which may have been blocking said openings; removing the penetrant from the surface of said body while permitting penetrant which has penetrated said flaw openings to remain, and subsequently inspecting said test body for the appearance of retained penetrant at and around said flaw openings to reveal the existence and extent of said flaw openings.

2. The method according to claim 1 in which said penetrant contains a dye to enhance the visible contrast between the retained penetrant and the surface of the test body at the time of inspection.

3. The method according to claim 2 in which inspection is conducted under wavelengths of light selected to enhance a color contrast between the retained penetrant and the surface of the test body at the time of inspection.

4. The method according to claim 2 in which said dye is a fluorescent dye and inspection is conducted under fluorescigenous radiation to cause portions of said penetrant appearing at and around said flaw openings to emit visible light.

5. The method according to claim 1 in which said ultrasonic vibrations range from approximately 25 kilocycles per second to 250 kilocycles per second.

References Cited in the file of this patent UNITED STATES PATENTS 2,259,400 Switzer Oct. 14, 1941 2,657,668 Meier Nov. 3, 1953 2,678,420 De Forest et al May 11, 1954