Title:
METHOD OF REDUCING NOTCH SENSITIVITY IN TUBULAR PRODUCTS
United States Patent 3805459
Abstract:
A method of reducing notch sensitivity in tubular products, to inhibit premature bursting on such being subjected to a high internal pressure. An interior circumferential region is uniformly roughened in the product together with removal of some material in the wall of the product. Such effectively eliminates notches as stress risers, and on a high internal pressure condition occurring, resultant stress is distributed throughout the uniformly roughened region.
US Patent References:
EXTRUSION APPARATUS AND METHOD
Moss - December 1970 - 3549735
Method of polishing hollow stem valves
Colwell - April 1945 - 2373459
Cleaning pipe
Malmros - July 1937 - 2087694
Process of treating metal frictional surfaces
Butler - December 1926 - 1610927
EXTRUSION APPARATUS AND METHOD
Moss - December 1970 - 3549735
Method of polishing hollow stem valves
Colwell - April 1945 - 2373459
Cleaning pipe
Malmros - July 1937 - 2087694
Process of treating metal frictional surfaces
Butler - December 1926 - 1610927
Application Number:
05/218472
Publication Date:
04/23/1974
Filing Date:
01/17/1972
View Patent Images:
Export Citation:
Assignee:
Zirconium Technology Corporation (Albany, OR)
Primary Class:
Other Classes:
451/54, 451/51
International Classes:
B24C1/10; C21D7/06; C21D7/00; B24B1/00
Field of Search:
51/319,320,321,290,323 252/142-151
Primary Examiner:
Kelly, Donald G.
Attorney, Agent or Firm:
Kolisch, Hartwell & Dickinson
Parent Case Data:
This is a continuation-in-part of my prior-filed application entitled, "Method of Reducing Notch Sensitivity in Tubular Products," filed Oct. 21, 1971, Ser. No. 191,201.
Claims:
I claim
1. A method of processing a tubular product to inhibit bursting of the product on its being subjected to a high internal pressure condition, the method comprising
2. The method of claim 1 wherein the blasting is performed from opposite ends of the tubular product to obtain a substantially uniform reduction in wall thickness throughout the entire length of the product.
3. The method of claim 1 which further comprises the step of pickling the internal circumferential region of the tubular product performed after blasting with particles of abrasive.
1. A method of processing a tubular product to inhibit bursting of the product on its being subjected to a high internal pressure condition, the method comprising
2. The method of claim 1 wherein the blasting is performed from opposite ends of the tubular product to obtain a substantially uniform reduction in wall thickness throughout the entire length of the product.
3. The method of claim 1 which further comprises the step of pickling the internal circumferential region of the tubular product performed after blasting with particles of abrasive.
Description:
This invention relates to a method of reducing notch sensitivity in tubular products.
In the manufacture of precisionly dimensioned seamless tubing, it is conventional to prepare a so-called tube hollow, usually through an extrusion process, and to pass such tube hollow through a tube reducer or other means whereby tube wall area reduction takes place. This reduction in tube wall area is produced both by reducing the diameter of the tube, as well as by reducing the thickness of the tube wall.
It is known that extrusion tool marks may leave longitudinal seams on the inner circumferential surface of a tube hollow. During subsequent reducing operations, these may be folded up into lap type defects that for the most part may be detected using non-destructive testing techniques, such as ultrasonic testing. By way of illustration, ultrasonic testing is capable of detecting a notch where such has a depth exceeding about 2 mils or more.
It is also known that pits and other types of discontinuities (embraced within the meaning of "notch" as used herein), not necessarily coming from the tube hollow, may also be present on the inner circumference of the finished tube.
Discontinuites of the type discussed may be of such small size that they are not detectable using nondesctructive testing prodecures. However, they introduce what is referred to herein as notch sensitivity, since their presence in the tube has been found to reduce seriously the circumferential ductility or fatique resistance of the product. These small surface discontinuities function as stress risers, whereby on destructive testing for such things as circumferential ductility (burst test), or for fatique resistance (pressure impulse test, etc.), premature failure is noted in the tube even though the product in all other respects has the appearance of meeting requisite standards.
The possession of a high degree of circumferential ductility is a requirement in certain types of tubing which may be expected during its service life to be subjected to a high internal pressure condition. Exemplary of such an application is the zirconium alloy tubing used as cladding for reactor fuel. Fuel swelling and the formation of gaseous fission products may produce a volume expansion within the tubing, and without circumferential ductility such results in bursting and radioactive contamination of the reactor system. With other types of tubes, such as tubes annealed to habe high strength levels, a high degree of fatique resistance, as determined by such tests as a pressure impulse test, is an important property. Exemplary of this type of tubing is the titanium alloy tubes that have been used extensively for hydraulic pressure lines in aircraft and the like. Over a period of time, sudden pressure surges, or accidental over pressures, produce failure in such lines if stress risers are present. By noting these specific areas of tube use, however, it is not intended to limit this invention to these particular uses.
A general object of this invention, therefore, it to provide a novel method of reducing notch sensitivity in tubular products.
More specifically, it is an object to provide such a method which is particularly applicable to the processing of tubing possessing small discontinuities which are ordinarily nondetectable using conventional ultrasonic, or other nondestructive techniques.
A more specific and further object of the invention is to provide such a method wherein steps in the method comprise uniformly roughening an interior circumferential region in the tubular product, and subsequently distributing stress within the product on a high internal pressure condition occuring.
A further and more specific object of the invention is provision of such a method wherein roughening is performed with reducing of tube wall thickness, wherein defect depth is reduced simultaneously with the introduction of a roughened surface.
These and other objects and advantages of the invention will become more fully apparent as the following description is read in conjunction with the accompanying drawings, wherein:
FIG. 1 is a cross section of a tube, on an enlarged scale, showing notch defects on the interior surface thereof; and
FIG. 2 is a cross section of such tube showing the tube after it has been roughened according to the present invention.
Describing an embodiment of the invention, a Zircaloy tube such as might be used to clad reactor fuel in a nuclear reactor conventionally may have an I.D. of approximately 1/2 inch, and a wall thickness ranging from 20 to 50 mils. Such may be prepared from a tube hollow by passing tube hollow in multiple passes through a tube reducer or other machine. Typically, for instance, the original tube hollow may have an O.D. of approximately 2 1/2 inches and a wall thickness of about 1/2 inch. During the multiple passes through the tube reducer, tube wall area reduction takes place, both by reason of the diameter of the tube hollow being reduced, and also through a reduction in wall thickness.
Finally prepared tubing of the type described may have an inner wall surface which gives the appearance of being perfectly smooth and free of any notches such as pits or other discontinuities. The tube may be subjected to ultrasonic tests which senses discontinuities of about two mil depth or more, without any such discontinuities being detected. However, when different specimens of such a tube are subjected to destructive testing, i.e., a burst test, a pressure impulse test, etc. to determine circumferential ductility or fatigue resistance, rather inconsistant results may be obtained.
Typically, circumferential ductility is determined in a burst test by calculating the total circumferential elongation occuring in the specimen before bursting, as a percent of the original outer circumference of the specimen, such calculation being referred to as percent TCE. Zircaloy tubing of the type described with not ultrasonically detectable defects, and without treatment as contemplated by the invention, may exhibit widely varying TCE values.
According to the invention, as interior circumferential region of the tube is uniformly roughened. With discontinuites of the type mentioned, i.e., having a depth of less than about 2 mils, the roughening imparted to the inside of the tube optimumly should lie within the range of about 20 to 50 rms. Further, the roughening should be uniform, in that from visual observation the roughening appears continuous over the inside of the tube.
In a preferred embodiment of the invention, roughening is performed by blasting the inside of the tube with particles of an abrasive material. Such blasting it effective not only to impart a uniform roughness to the tube's interior, but also to reduce to some extent the tube wall thickness. The tubes with which I have been concerned have ranged in length from 5 to 25 feet. In roughening such tubes by blasting with an abrasive, the blast nozzle is introduced to each of the opposite ends of the tube. The abrasive airborne particles are directed by the nozzle in a turbulent type of flow down the interior of the tube to the tube's opposite open end. With blasting done from both ends, wall reduction throughout the entire length of the tube is substantially uniform.
Ordinarily, and with defects of about two mil penetration or less, the tube wall thickness is reduced at least about one third of such penetration. For instance, with defects expected to have a depth range of up to about two mils, material removal is performed to reduce the wall thickness on an average at least about one third of this amount and preferably about one half, which amounts to a wall reduction of one mil.
Ordinarily the amount of material removed during the blasting is less than about 5 percent of wall thickness. With tubes having a wall thickness greater than the range indicated, it is permissible to remove more material from the inside of the tube without exceeding this limitation, as long as specified tolerances on wall thickness are maintained.
In selecting the abrasive material used to roughen the tube interior, and to attain a roughness within the range indicated, the abrasive preferably should have a particle size within the range of about 40 to 150 mesh. An excellent abrasive to perform the roughening is silicon carbide, although it should be obvious, of course, that any of the well known abrasive materials would be suitable.
Describing a particular roughening operation, Zircaloy alloy tubes of 0.450 inch I.D. and free of ultrasonically detectable discontinuities were roughened on their inner surfaces using a silicon carbide abrasive. The tubes were blasted from both ends, using a conventional air pressure operated blasting nozzle inserted into each of opposite ends of the tubes. Air at 50 p.s.i. was used in the blasting, and blasting were performed for a period of about 1 minute at each of the respective ends of the tubes. The blasting produced material removal from the inner walls, whereby there was an average descrease in wall thickness of about one mil, plus or minus 0.3 mils. Surface roughness within the interior of the tube was determined to be within the range of about 30 to 40 RMS, and appeared uniform throughout.
Tubes so processed were evaulated in the laboratory for circumferential ductility together with specimens of untreated tubes. In such evaluation, the tubes were subjected to an internal pressure increased to a level sufficient to produce bursting. Percent TCE was calculated for the various untreaded specimens, as well as the tube specimens which had been roughened as described. The untreated specimens exhibited a percent TCE ranging from about 8 to 23, with the average percent TCE lying about midway between this range, or being approximately 15 percent. With the treated specimens, calculated percent TCE ranged from about 22 to 25 percent.
In other tests, titanium alloy tubes annealed to have high strength were similarly roughened. These tubes were subjected to pressure impulse tests to determine fatigue resistance. Essentially all tubes tested met commercial standards for aircraft hydraulic lines.
As illustrated in the accompanying drawings, a portion of a tube on a greatly enlarged scale is shown at 10. Such has an inner circumferential surface designated 10a. Notches in the surface are depicted at 12, 14, and 16. These notches, as is typical, vary in depth with notch 12 having a depth of less than one mil, notch 14 having a depth of about 1 mil, and notch 16 having a depth of almost 2 mils. All of such notched ordinarily would be nondetectable using nondestructive testing techniques.
With roughening of the tube interior, the tube appears as shown in Fig. 2. The roughened surface is uniformly roughened as indicated by the overlapping nature of the surfaces defining the depressions formed by the abrasive particles. With material removal reducing wall thickness by about 1 mil, the notches 12 and 14 are essentially completely removed. A trace of notch 16 remains, but metal around the edge of the notch has been removed, and because of this and, because the depth of the notch is decreased by the metal removal, the notch in effect becomes lost amongst the various depression produced by the roughening.
In some instances, to obtain desired cleanliness within the tube, it may be desirable to subject the tube after the roughening operation described to a pickling step which has the effect of removing some additional material from the inner surface of the tube wall. Describing a typical pickling operation, the tube may be immersed in an acid pickling solution consisting of 15 parts by volume of a 4 percent by weight hydrofluoric acid solution, and 80 parts by volume water. Exposure to the acid solution may not be long, usually not exceeding about a half a minute. After the acid treatment, the tube is subjected to a water rinse. Such a pickling removes approximately 0.5 mil of material from the surfaces exposed to the acid solution.
The appearance of the tube interior after pickling indiscates a smoothening of the tube interior by the pickling step. Thus the RMS is reduced somethat through this cleaning operation. Tubes which have been cleaned by pickling, after having been roughened as contemplated by this invention, exhibit the same improvement of reduction to notch sensitivity as tubes that are roughened without pickling.
It should be pointed out here that tubes that are subjected to a pickling, without roughening as by blasting, do not exhibit the same improvement with respect to reduced notch sensitivity. This may be explained by the fact that blasting, in a manner of speaking, is preferential, in that it removes material along the plane of the surface on which it is performed without removing material, for instance, from the base of a notch. Pickling, on the other hand, is nonpreferential, in that when such is performed it removes material both from the plane of the surface as well as from the notch base.
In reducing the tube wall area of the tube hollow to produce a tube, when a tube reducer is utilized, the hollow is pushed through dies contacting the hollow's outer surface which are in motion with the internal area of the hollow supported on a stationary mandril. In a swaging operation, on the other hand, the tube hollow is pushed through a stationary die contacting the hollow's outer surface, together with a stationary mandril which supports the inner surface of the tube hollow. So-called drawing is similar to swaging, save that the tube is pulled through the die and over the mandril, rather than pushed. In all these types of reduction operations, it is normal to expect a somewhat smoother outer surface of the tube, with less notch defects, than the inner surface, by reason of better lubrication during the reduction process.
The outer surface of the tube may be conditioned in a number of different ways to inhibit notch sensitivity therein. It should be pointed out, however, that the conditioning of the outer surface of a tube does not present the same problems as conditioning the tube's inner surface, by reason of the fact that the surface is exposed, and visually (aided by magnifiying means if necsssary), it is possible to determine the extent, distribution, and nature of any notches therealong.
In conditioning the outer surface of a tube, such may be done, for instance, through a polishing action using polishing equipment, which might typically remove about a mil of material. This may be followed by a pickling action, carried on to remove about two mils of material from the outer surface. Polishing alone is not satisfactory, ordinarily, and must be followed by a pickling, in order that any polishing marks be removed.
In another type of conditioning of the outer surface, the tube, for instance, may have its outer surface first detergent cleaned, then polished, followed by a pickling action, with a final peening of the surface. The peening produces a working of the surface but does not result in a material removal, as does the roughening action contemplated for the tube material by the present invention.
In performing roughening on an inner surface of a tube with an abrasive and by blasting, as described above, the nozzle ejects the abrasive particles at an acute angle against the inner surface of the tube, which particles then move in a turbulent motion down through the tube interior to produce the roughening desired. The process is particularly suitable for a tube interior, since the particles are funneled down through the tube by the tube wall in random motion, to produce a material removal which is uniform about the circumference of the tube. Material removable on the outer surface of the tube through a blasting operation is not as effectively performed, since the outer part of the tube does not operate to guide or channel abrasive particles, and as a consequence, uniform removal circumferentially about the tube through the abrasive action is not readily possible. According to the invention as preferred, therefore, it is contemplated that the tube outer surface be conditioned through steps such as polishing and pickling, polishing and peening, or cleaning, polishing, pickling and peening, carried on in such order, with the inner surface of the tube conditioned through the abrasive roughening described.
The abrasive roughening is distinguishabled from the type of process used on the outside of the tube in that, in one operation, there is both material removal and condition of the tube surface, the removal in a manner of speaking being preferential. Polishing, for example, performs a preferential material removal, but must be followed by some sort of conditioning step. Comparing a peening operation, this conditions the surface of a tube, but does not result in material removal.
The effect of the roughening, as contemplated by the invention, is to eliminate surface discontinuities in the tubes as stress risers. Stress becomes distributed throughout the roughened internal regions of the tube on a high pressure occurring.
While an embodiment of the invention has been described it should be obvious that modifications and variations are possible without departing therefrom.
In the manufacture of precisionly dimensioned seamless tubing, it is conventional to prepare a so-called tube hollow, usually through an extrusion process, and to pass such tube hollow through a tube reducer or other means whereby tube wall area reduction takes place. This reduction in tube wall area is produced both by reducing the diameter of the tube, as well as by reducing the thickness of the tube wall.
It is known that extrusion tool marks may leave longitudinal seams on the inner circumferential surface of a tube hollow. During subsequent reducing operations, these may be folded up into lap type defects that for the most part may be detected using non-destructive testing techniques, such as ultrasonic testing. By way of illustration, ultrasonic testing is capable of detecting a notch where such has a depth exceeding about 2 mils or more.
It is also known that pits and other types of discontinuities (embraced within the meaning of "notch" as used herein), not necessarily coming from the tube hollow, may also be present on the inner circumference of the finished tube.
Discontinuites of the type discussed may be of such small size that they are not detectable using nondesctructive testing prodecures. However, they introduce what is referred to herein as notch sensitivity, since their presence in the tube has been found to reduce seriously the circumferential ductility or fatique resistance of the product. These small surface discontinuities function as stress risers, whereby on destructive testing for such things as circumferential ductility (burst test), or for fatique resistance (pressure impulse test, etc.), premature failure is noted in the tube even though the product in all other respects has the appearance of meeting requisite standards.
The possession of a high degree of circumferential ductility is a requirement in certain types of tubing which may be expected during its service life to be subjected to a high internal pressure condition. Exemplary of such an application is the zirconium alloy tubing used as cladding for reactor fuel. Fuel swelling and the formation of gaseous fission products may produce a volume expansion within the tubing, and without circumferential ductility such results in bursting and radioactive contamination of the reactor system. With other types of tubes, such as tubes annealed to habe high strength levels, a high degree of fatique resistance, as determined by such tests as a pressure impulse test, is an important property. Exemplary of this type of tubing is the titanium alloy tubes that have been used extensively for hydraulic pressure lines in aircraft and the like. Over a period of time, sudden pressure surges, or accidental over pressures, produce failure in such lines if stress risers are present. By noting these specific areas of tube use, however, it is not intended to limit this invention to these particular uses.
A general object of this invention, therefore, it to provide a novel method of reducing notch sensitivity in tubular products.
More specifically, it is an object to provide such a method which is particularly applicable to the processing of tubing possessing small discontinuities which are ordinarily nondetectable using conventional ultrasonic, or other nondestructive techniques.
A more specific and further object of the invention is to provide such a method wherein steps in the method comprise uniformly roughening an interior circumferential region in the tubular product, and subsequently distributing stress within the product on a high internal pressure condition occuring.
A further and more specific object of the invention is provision of such a method wherein roughening is performed with reducing of tube wall thickness, wherein defect depth is reduced simultaneously with the introduction of a roughened surface.
These and other objects and advantages of the invention will become more fully apparent as the following description is read in conjunction with the accompanying drawings, wherein:
FIG. 1 is a cross section of a tube, on an enlarged scale, showing notch defects on the interior surface thereof; and
FIG. 2 is a cross section of such tube showing the tube after it has been roughened according to the present invention.
Describing an embodiment of the invention, a Zircaloy tube such as might be used to clad reactor fuel in a nuclear reactor conventionally may have an I.D. of approximately 1/2 inch, and a wall thickness ranging from 20 to 50 mils. Such may be prepared from a tube hollow by passing tube hollow in multiple passes through a tube reducer or other machine. Typically, for instance, the original tube hollow may have an O.D. of approximately 2 1/2 inches and a wall thickness of about 1/2 inch. During the multiple passes through the tube reducer, tube wall area reduction takes place, both by reason of the diameter of the tube hollow being reduced, and also through a reduction in wall thickness.
Finally prepared tubing of the type described may have an inner wall surface which gives the appearance of being perfectly smooth and free of any notches such as pits or other discontinuities. The tube may be subjected to ultrasonic tests which senses discontinuities of about two mil depth or more, without any such discontinuities being detected. However, when different specimens of such a tube are subjected to destructive testing, i.e., a burst test, a pressure impulse test, etc. to determine circumferential ductility or fatigue resistance, rather inconsistant results may be obtained.
Typically, circumferential ductility is determined in a burst test by calculating the total circumferential elongation occuring in the specimen before bursting, as a percent of the original outer circumference of the specimen, such calculation being referred to as percent TCE. Zircaloy tubing of the type described with not ultrasonically detectable defects, and without treatment as contemplated by the invention, may exhibit widely varying TCE values.
According to the invention, as interior circumferential region of the tube is uniformly roughened. With discontinuites of the type mentioned, i.e., having a depth of less than about 2 mils, the roughening imparted to the inside of the tube optimumly should lie within the range of about 20 to 50 rms. Further, the roughening should be uniform, in that from visual observation the roughening appears continuous over the inside of the tube.
In a preferred embodiment of the invention, roughening is performed by blasting the inside of the tube with particles of an abrasive material. Such blasting it effective not only to impart a uniform roughness to the tube's interior, but also to reduce to some extent the tube wall thickness. The tubes with which I have been concerned have ranged in length from 5 to 25 feet. In roughening such tubes by blasting with an abrasive, the blast nozzle is introduced to each of the opposite ends of the tube. The abrasive airborne particles are directed by the nozzle in a turbulent type of flow down the interior of the tube to the tube's opposite open end. With blasting done from both ends, wall reduction throughout the entire length of the tube is substantially uniform.
Ordinarily, and with defects of about two mil penetration or less, the tube wall thickness is reduced at least about one third of such penetration. For instance, with defects expected to have a depth range of up to about two mils, material removal is performed to reduce the wall thickness on an average at least about one third of this amount and preferably about one half, which amounts to a wall reduction of one mil.
Ordinarily the amount of material removed during the blasting is less than about 5 percent of wall thickness. With tubes having a wall thickness greater than the range indicated, it is permissible to remove more material from the inside of the tube without exceeding this limitation, as long as specified tolerances on wall thickness are maintained.
In selecting the abrasive material used to roughen the tube interior, and to attain a roughness within the range indicated, the abrasive preferably should have a particle size within the range of about 40 to 150 mesh. An excellent abrasive to perform the roughening is silicon carbide, although it should be obvious, of course, that any of the well known abrasive materials would be suitable.
Describing a particular roughening operation, Zircaloy alloy tubes of 0.450 inch I.D. and free of ultrasonically detectable discontinuities were roughened on their inner surfaces using a silicon carbide abrasive. The tubes were blasted from both ends, using a conventional air pressure operated blasting nozzle inserted into each of opposite ends of the tubes. Air at 50 p.s.i. was used in the blasting, and blasting were performed for a period of about 1 minute at each of the respective ends of the tubes. The blasting produced material removal from the inner walls, whereby there was an average descrease in wall thickness of about one mil, plus or minus 0.3 mils. Surface roughness within the interior of the tube was determined to be within the range of about 30 to 40 RMS, and appeared uniform throughout.
Tubes so processed were evaulated in the laboratory for circumferential ductility together with specimens of untreated tubes. In such evaluation, the tubes were subjected to an internal pressure increased to a level sufficient to produce bursting. Percent TCE was calculated for the various untreaded specimens, as well as the tube specimens which had been roughened as described. The untreated specimens exhibited a percent TCE ranging from about 8 to 23, with the average percent TCE lying about midway between this range, or being approximately 15 percent. With the treated specimens, calculated percent TCE ranged from about 22 to 25 percent.
In other tests, titanium alloy tubes annealed to have high strength were similarly roughened. These tubes were subjected to pressure impulse tests to determine fatigue resistance. Essentially all tubes tested met commercial standards for aircraft hydraulic lines.
As illustrated in the accompanying drawings, a portion of a tube on a greatly enlarged scale is shown at 10. Such has an inner circumferential surface designated 10a. Notches in the surface are depicted at 12, 14, and 16. These notches, as is typical, vary in depth with notch 12 having a depth of less than one mil, notch 14 having a depth of about 1 mil, and notch 16 having a depth of almost 2 mils. All of such notched ordinarily would be nondetectable using nondestructive testing techniques.
With roughening of the tube interior, the tube appears as shown in Fig. 2. The roughened surface is uniformly roughened as indicated by the overlapping nature of the surfaces defining the depressions formed by the abrasive particles. With material removal reducing wall thickness by about 1 mil, the notches 12 and 14 are essentially completely removed. A trace of notch 16 remains, but metal around the edge of the notch has been removed, and because of this and, because the depth of the notch is decreased by the metal removal, the notch in effect becomes lost amongst the various depression produced by the roughening.
In some instances, to obtain desired cleanliness within the tube, it may be desirable to subject the tube after the roughening operation described to a pickling step which has the effect of removing some additional material from the inner surface of the tube wall. Describing a typical pickling operation, the tube may be immersed in an acid pickling solution consisting of 15 parts by volume of a 4 percent by weight hydrofluoric acid solution, and 80 parts by volume water. Exposure to the acid solution may not be long, usually not exceeding about a half a minute. After the acid treatment, the tube is subjected to a water rinse. Such a pickling removes approximately 0.5 mil of material from the surfaces exposed to the acid solution.
The appearance of the tube interior after pickling indiscates a smoothening of the tube interior by the pickling step. Thus the RMS is reduced somethat through this cleaning operation. Tubes which have been cleaned by pickling, after having been roughened as contemplated by this invention, exhibit the same improvement of reduction to notch sensitivity as tubes that are roughened without pickling.
It should be pointed out here that tubes that are subjected to a pickling, without roughening as by blasting, do not exhibit the same improvement with respect to reduced notch sensitivity. This may be explained by the fact that blasting, in a manner of speaking, is preferential, in that it removes material along the plane of the surface on which it is performed without removing material, for instance, from the base of a notch. Pickling, on the other hand, is nonpreferential, in that when such is performed it removes material both from the plane of the surface as well as from the notch base.
In reducing the tube wall area of the tube hollow to produce a tube, when a tube reducer is utilized, the hollow is pushed through dies contacting the hollow's outer surface which are in motion with the internal area of the hollow supported on a stationary mandril. In a swaging operation, on the other hand, the tube hollow is pushed through a stationary die contacting the hollow's outer surface, together with a stationary mandril which supports the inner surface of the tube hollow. So-called drawing is similar to swaging, save that the tube is pulled through the die and over the mandril, rather than pushed. In all these types of reduction operations, it is normal to expect a somewhat smoother outer surface of the tube, with less notch defects, than the inner surface, by reason of better lubrication during the reduction process.
The outer surface of the tube may be conditioned in a number of different ways to inhibit notch sensitivity therein. It should be pointed out, however, that the conditioning of the outer surface of a tube does not present the same problems as conditioning the tube's inner surface, by reason of the fact that the surface is exposed, and visually (aided by magnifiying means if necsssary), it is possible to determine the extent, distribution, and nature of any notches therealong.
In conditioning the outer surface of a tube, such may be done, for instance, through a polishing action using polishing equipment, which might typically remove about a mil of material. This may be followed by a pickling action, carried on to remove about two mils of material from the outer surface. Polishing alone is not satisfactory, ordinarily, and must be followed by a pickling, in order that any polishing marks be removed.
In another type of conditioning of the outer surface, the tube, for instance, may have its outer surface first detergent cleaned, then polished, followed by a pickling action, with a final peening of the surface. The peening produces a working of the surface but does not result in a material removal, as does the roughening action contemplated for the tube material by the present invention.
In performing roughening on an inner surface of a tube with an abrasive and by blasting, as described above, the nozzle ejects the abrasive particles at an acute angle against the inner surface of the tube, which particles then move in a turbulent motion down through the tube interior to produce the roughening desired. The process is particularly suitable for a tube interior, since the particles are funneled down through the tube by the tube wall in random motion, to produce a material removal which is uniform about the circumference of the tube. Material removable on the outer surface of the tube through a blasting operation is not as effectively performed, since the outer part of the tube does not operate to guide or channel abrasive particles, and as a consequence, uniform removal circumferentially about the tube through the abrasive action is not readily possible. According to the invention as preferred, therefore, it is contemplated that the tube outer surface be conditioned through steps such as polishing and pickling, polishing and peening, or cleaning, polishing, pickling and peening, carried on in such order, with the inner surface of the tube conditioned through the abrasive roughening described.
The abrasive roughening is distinguishabled from the type of process used on the outside of the tube in that, in one operation, there is both material removal and condition of the tube surface, the removal in a manner of speaking being preferential. Polishing, for example, performs a preferential material removal, but must be followed by some sort of conditioning step. Comparing a peening operation, this conditions the surface of a tube, but does not result in material removal.
The effect of the roughening, as contemplated by the invention, is to eliminate surface discontinuities in the tubes as stress risers. Stress becomes distributed throughout the roughened internal regions of the tube on a high pressure occurring.
While an embodiment of the invention has been described it should be obvious that modifications and variations are possible without departing therefrom.