COLLAPSING PRESSURES OF LAP- WELDED STEEL TUBES BY PROF. REID T. STEWART MEMBER OF THE SOCIETY Western University , Allegheny, Pa. A PAPER READ AT THE CHATTANOOGA MEETING, MAY, 1906 JUL 2 S 1924 UNIVERSITY OF ILLINOIS 1906 PUBLISHED BY THE SOCIETY 12 W. 31st Street, New York, N. Y. G%0 ' II EtlGlUtEKiUG UtiilMi 4-Cx COLLAPSING PRESSURES OF LAP-WELDED STEEL TUBES. 3 Wo. 1116.* COLLAPSING PRESSURES OF BESSEMER STEEL LAP- WELDED TUBES, THREE TO TEN INCHES IN DIAMETER. BY PROF. REID T. STEWART, PITTSBURG, PA. (Member of the Society.) Abstract. This research was undertaken for the purpose of supply- ing an urgent demand for reliable information on the behavior of modern wrought tubes when subjected to fluid collapsing pres- sure. Every means known to engineering science that could aid in the accomplishment of this undertaking has been used, and every possible effort made to get at the truth and have the research yield trustworthy data. It was planned and executed under the immediate direction of the author, at the McKeesport works of the National Tube Company, and has occupied for its completion, during a period of four years, the time of from one to six men. Series One . — This series of tests was made on tubes that were 8f inches outside diameter, for all the different commercial thick- nesses of wall, and in lengths of 2± 5, 10, 15 and 20 feet between transverse joints tending to hold the tube to a circular form. The chief purpose of this series of tests was to furnish data for deter- mining which of the existing formulae, if any, were applicable to modern lap-welded steel tubes, especially when used in com- paratively long lengths, such as well casing, boiler tubes and long plain flues. Series Two. — This series of tests was made on single lengths of 20 feet between end connections, tending to hold the tube to a cir- cular form. Seven sizes, from 3 to 10 inches outside diameter, and in all the commercial thicknesses obtainable, have been tested to date. The chief purpose of these tests was to obtain, for com- mercial tubes, the manner in which the collapsing pressure of a tube is related to both the diameter and thickness of wall. * Presented at the Chattanooga, Tennessee, Meeting (May, 1906) of the Ameri- can Society of Mechanical Engineers and forming part of Volume 27 of the Transactions. 4 COLLAPSING PRESSURES OF LAP-WELDED STEEL TUBES. Inapplicability of Previously Published Formulae. — Prepara- tory to entering upon the present research all existing published formulae that could be found were collected, and, after the com- pletion of Series One, were tested as to their applicability to mod- ern steel tubes. Among the formulae thus tested were two each by Eairbairn, Unwin, Wehage and Clark, and one each by Nys- trom, Grashof, Love, Belpaire, and the Board of Trade (British), all of which, with possibly two exceptions, appear to be based upon Fairbairn’s classical experiments made more than a half century ago, upon tubes wholly unlike the modern product. With- out exception, all of these formulae, when thus tested, proved to be inapplicable to the wide range of conditions found in modern practice. As an illustration of this, the very first tube tested in connection with this research failed under a pressure that exceeded by about 300 per cent, that calculated by means of Fairbairn’s for- mula. Results of Present Research. — The principal conclusions to be drawn from the results of the present research may be briefly stated as follows : 1. The length of tube, between transverse joints tending to hold it to a circular form, has no practical influence upon the collapsing pressure of a commercial lap-welded steel tube so long as this length is not less than about six diameters of tube. (Pp. 32, 40.) 2. The formulae, as based upon the present research, for the collapsing pressures of modern lap-welded Bessemer steel tubes, are as follows : P = 1,000 (l - Vl - 1,600 J) . . • • (A) P = 86,670 2 - 1,386 • • (B) Where P = collapsing pressure, pounds per sq. inch. d = outside diameter of tube in inches. t = thickness of wall in inches. Formula A is for values of P less than 581 pounds, or for values of 4 less than 0.023, while formula B is for values greater than CL these. These formulae, while strictly correct for tubes that are 20 feet in length between transverse joints tending to hold them to a cir- COLLAPSING PRESSURES OF LAP-WELDED STEEL TUBES. 5 cular form, are, at the same time, substantially correct for all lengths greater than about six diameters. They have been tested for seven sizes, ranging from 3 to 10 inches outside diameter, in all obtainable commercial thicknesses of wall, and are known to he correct for this range. For the convenience of those who wish to apply these formula to practice a table has been calculated, giving the collapsing pres- sures of all the commercial sizes of lap-welded tubes from 2 to 11 inches outside diameter. (See p. 84.) For those who prefer graphical methods charts have been con- structed, for use of which see pp. 89, 92. When applying these formulae, tables and charts to practice, it should be remembered that a suitable factor of safety must be applied, which should not he less than from 3 to 6, see p. 88. 3. The apparent fiber stress under which the different tubes failed varied from about 7,000 pounds for the relatively thinnest to 35,000 pounds per square inch for the relatively thickest walls. Since the average yield point of the material was 37,000 and the tensile strength 58,000 pounds per square inch, it would appear that the strength of a tube subjected to a fluid collapsing pressure is not dependent alone upon either the elastic limit or ultimate strength of the material constituting it. (See p. 73.) Introduction. The planning and execution of this research was rendered especially difficult because of the lack of any reliable data bearing upon the behavior of modern wrought tubes when subjected to a fluid collapsing pressure. Fairbairn’s experiments, made more than a half century ago on tubes unlike the modern product, were of such a character as not to furnish suitable data for the planning of a similar but much more elaborate research on modern tubes. Aside from the numerous formulse, some ten or twelve in number, based practically upon Fairbairn’s ex- periments, and therefore not to he seriously considered in this connection, the only available data consisted, so far as could be discovered, of a few isolated experiments on flues and several records of the condition under which tubes and flues have failed in service, together with a table of computed collapsing pressures published in a well-known handbook, whose origin could not be traced. As an illustration of the utter unreliability of ex- 6 COLLAPSING PRESSURES OF LAP-WELDED STEEL TUBES. i sting data on this subject, at the commencement of this research, the very first wrought tube tested failed at a pressure that ex- ceeded by about 300 per cent, that calculated by means of Fair- bairn’s formulae. The experimental part of this research was carried out under the immediate direction of the author at the National Department of the National Tube Company, McKeesport, Pa. He is greatly indebted to the officials of the National Department for the courtesy shown him, especially to the manager, Mr. G. G. Crawford, to the superintendent of the tube mills, Mr. A. M. Saunders, and to Mr. J. A. McCulloch, in whose department the special apparatus was con- structed and the experiments conducted. The great interest in the work shown by Mr. McCulloch and his many valuable suggestions as the work progressed were of inestimable value. All the author’s wishes in the matter have been cheerfully carried out, the Tube Company generously providing every needful facility for carrying on the research in a most thorough manner. The exceptional consistency of the results obtained, taking all things into consideration, are due in a large measure to the care with which the author’s assistants, Messrs. H. G. Wardale, H. E. Williams and J. N. Kinney, have done their work; and the value of the final conclusions are due largely to Messrs. E. E. Shanor and F. P. Kramer, who have, under his immediate direction, deduced the formulae representing the results of the experiments and prepared the tables, charts and drawings contained in the body of this paper. It is due Mr. Shanor to state that the greater part of this work has been done by him. The original Log of Tests comprises, in addition to what has been abstracted for this paper, a complete file of autographic calipering diagrams, photographs showing two views of each tube after being collapsed, impressions from the collapsed sections, and remarks on each individual test. The complete record fills two quarto volumes, each about five inches thick, and, in addition, the matter resulting from working up this data in order to get the final results obtained are sufficient to fill a third volume. All this matter has been carefully worked over for this paper and condensed into the form of tabulated results and charts show- ing the consistency of the results obtained, and at the same time revealing to the eye the laws involved. While much has been necessarily omitted, it is hoped that enough has been given to convince the engineer or artisan, who COLLAPSING PRESSURES OF LAP-WELDED STEEL TUBES. 7 may have use for them, of the trustworthiness of the final con- clusions. Preliminary Considerations. While planning this research it was assumed that the resistance offered by a tube to an external fluid pressure would depend upon the following five things, namely : 1. The diameter of the tube. 2. The length of tube between transverse joints or end con- nections tending to hold it to a circular form. 3. The thickness of the wall. 4. The deviation of the tube from perfect roundness. 5. The physical properties of the material of which the tube is made. Of these five things that may vary it was thought that, for the preliminary experiments, at least, Nos. 4 and 5 would be prac- tically constant; No. 4, because the tubes being all made by the same process, would probably run fairly uniform as to deviation from roundness, and No. 5, because the material in this case being Bessemer tube steel, is known to run fairly uniform in its physical properties. The physical tests would, of course, serve as a check upon this latter. The only variation, then, to be expected in Nos. 4 and 5 would be that due to the inability of the manufacturer to turn out a uniform product. It is recognized here that the physical properties of rolled steel depend in some measure, other things being equal, upon the thickness of the plate; or, in this case, upon the thick- ness of the wall of the tube. It is clear that any variation of this nature would be a function of the thickness, and would conse- quently be taken care of in an empirical formula by the quantity representing the thickness. All the published formulae bearing upon the subject indicated that the diameter and thickness of wall has each an important determining influence on the collapsing pressure of a tube; and since there were the best of theoretical reasons for believing this to be the case, it was of course decided to plan the research to dis- cover, if possible, the precise nature of this influence over a wide commercial range. The influence of length of tube, between transverse joints, or end connections tending to hold it to a circular form, upon the collapsing pressure, appeared, in the light of available data, to 8 COLLAPSING PRESSURES OF LAP-WELDED STEEL TUBES. be the most uncertain of all the variables entering the problem. It was therefore decided, first of all, to determine the precise na- ture of this influence. In order to do this the following apparatus was used, the greater part of which was especially constructed for this research : Hydraulic Test Apparatus. The production of a suitable apparatus in which to subject the tubes to an external fluid pressure, and at the same time handle with expedition the large number of tests contemplated, was a somewhat difficult problem to solve. After much consideration of the matter the scheme illustrated in Tig. 1 was adopted. It will be seen by reference to this figure that the scheme pro- vides for: 1. A test cylinder with one head removable for the reception of the tube to be tested, this cylinder being provided with means for creating an hydraulic pressure within, thus subjecting the tube under test to a fluid-collapsing pressure. 2. A low pressure water supply, L , of large volume to rapidly fill the space within the test cylinder not occupied by the tube under test. 3. A variable high pressure water supply, H, furnished by an hydraulic pressure pump, P , the purpose of which was to create a fluid pressure within the test cylinder, the tube under test by this means being subjected to a gradually increasing fluid-col- lapsing pressure. 4. A set of pressure gauges, B, C, T>, having a large range in capacity, connected so that they could be used either singly for indicating the fluid pressure within the test cylinder or in com- bination for comparison. 5. A vent pipe, V, leading from the interior of the tube under test through the head of the test cylinder to the atmosphere, in order to maintain constantly an atmospheric pressure within the tube being tested. 6. An air vent, E, connecting with the highest point of the in- terior of the test cylinder, in order to thoroughly free it from air while being filled with water, after the insertion of a tube to be tested. In addition to the above, while carrying out this scheme, de- vices were in use for manipulating the removable head, and for handling the tubes while being entered and withdrawn, but in COLLAPSING PRESSURES OF LAP-WELDED STEEL TUBES, -Sixteen-inch^Hydraulic Test Apparatus. Especially Designed and Constructed for Collapsing Tests on Tubes, Conducted by Prof. R. T. Stewart. 10 COLLAPSING PRESSURES OF LAP-WELDED STEEL TUBES. order not to encumber the paper with unessential details no men- tion of these will be made. Sixteen-Inch Test Cylinder . — This cylinder as originally con- structed was made up of three sections, whose aggregate length approximated 45 feet, the intention being to have it long enough to accommodate a string of well casing consisting of either two full lengths of 20 feet each, including three couplings, or one full length of 20 feet, with a half length coupled to each of its ends. It was soon discovered that the behavior of a tube in collapse was such that precisely the same results could be had from a single 20-foot length as from either of the above arrangements. Because of this the cylinder was shortened at the first opportunity, by the removal of the intermediate section, to a length of about 30 feet. The sections of this test cylinder were made from Bessemer steel lap-welded tubes, 16 inches outside diameter and three-quar- ters inch thick, to which steel flanges were welded for the inter- mediate joints. Thickening rings were welded to the ends in- tended to be threaded for the attaching of the heads. The highest fluid pressure reached in this test cylinder was in connection with the retest of No. 418, which failed under a fluid pressure of 2,890 pounds per square inch. This corresponds to a stress of 28,000 pounds per square inch in the wall of the cylin- der. This was about as near the yield point of the material con- stituting the cylinder as it was thought prudent to go, so that all tests at higher fluid pressures were made in the 8-inch test cylinder, which was relatively about twice as strong. The heads for the 16-inch test cylinder were made from cir- cular blanks punched from steel plates 2^ inches thick, and pressed into shape, while hot, by means of an hydraulic press. They were then fitted to the ends of the test cylinder, which had been re- enforced in the manner already described, by means of trapezoidal threads designed so as to best resist stress in one direction. In the retest on No. 418 these threads were subjected to a shearing stress of 666,000 pounds. The flange joints connecting the different sections of the test cylinder were tongued and grooved, and were made up with leather packing in the bottom of the grooves. These joints each con- tained eighteen lf-inch steel bolts, and were fully as strong against internal fluid pressure as the wall of the cylinder. Means of Filling Test Cylinder . — The rear end of the test cylin- der was connected, in the manner shown in Big. 1, to the low COLLAPSING PRESSURES OF LAP-WELDED STEEL TUBES. 11 pressure water supply, L , of the works, for the purpose of rapidly filling the space within the cylinder and surrounding the tube under test. By this means the test cylinder was quickly filled with water, the pressure within being maintained constantly an atmos- pheric pressure by means of the air vent, E , shown at the top of the left hand head. This vent also served the purpose of entirely freeing the cylinder from imprisoned air, thus reducing to a mini- mum the distortion of the tube under test when failure occurred, and also rendering a serious accident to the attendants impossible in case rupture of the cylinder wall should occur while making a test. Hydraulic Pressure Pum,p . — The pressure within the test cylin- der was created by means of an hydraulic pressure pump capable of working against a fluid pressure up to 3,000 pounds per square inch. Ordinarily this pump was operated, upon entering the region of expected collapse, so as to increase the fluid pressure at a rate of from about 2 to 10 pounds per second, depending upon the gauge used. At these rates of increase of pressure the con- ditions were favorable for the making of an exact determination of the fluid pressure under which the tube failed. Pressure Gauges . — The gauges used for indicating the pressure at instant of collapse were three Shaw differential-piston mer- cury gauges, having capacities of 1,000, 3,000 and 8,500 pounds per square inch. They were connected in the manner shown in Fig. 1, so that, by opening or closing suitable valves, any one or more of them could be connected to the test cylinder for the pur- pose of indicating the pressure therein. They could also he in- terconnected for the purpose of comparing their scale readings at different pressures. The matter of selecting a suitable type of gauge for this research was, at the start, given due consideration. Spring gauges, owing to their liability to become deranged when once calibrated, were not to he considered, and a mercury column for the high pressure expected was out of the question. After considering various forms of dead-weight testers and high-pressure manometers, it was de- cided to use the Shaw differential-piston mercury gauge. This gauge is in reality a mercury column shortened, for all pressures, to a length of about three feet, by the introduction of differential pistons. These pistons are very ingeniously provided with soft rubber disks, placed so as to render them absolutely fluid-tight, and at the same time practically frictionless. With clean pistons and 12 COLLAPSING PRESSURES OF LAP-WELDED STEEL TUBES. new rubber disks these gauges were sensitive and in every respect reliable. For the service required of them in connection with this research, these gauges were superior to the usual hydraulic spring gauge in three very important respects, namely: 1. The scale of the Shaw mercury gauge as compared with that of the hydraulic spring gauge can be read with about three times the accuracy, that is, the error of scale reading is only about one- third that of the ordinary spring gauge of the same capacity. 2. Since this gauge is in reality a shortened mercury column, it is, when properly constructed, as reliable as the latter. In this respect it bears the same relation to the spring gauge as the mer- curial barometer does to the aneroid. It lacks, of course, the closeness with which pressures may be read on a mercury column just in proportion to the relative lengths of their respective scales. 3. The Shaw gauge is practically free from the vibrations that are often so annoying when using a spring gauge. This property of the mercury gauge rendered it eminently serviceable in this con- nection, since it was necessary to create the fluid pressure by means of a plunger pump without an air cushion. ! Eight-Inch Test Cylinder . — This smaller cylinder was con- structed for the purpose of testing the 3 and 4-inch tubes, and all of these sizes were tested in it with the exception of Nos. 462 and 464-469. This test cylinder was made up from a single 20-foot length of 8-inch double extra strong pipe, 8f inches outside di- ameter, and J-inch wall. The details of one end of this cylinder, with tube under test in place, are shown in Fig. 2, the other end being an exact dupli- cate of the one shown. It will be observed that this apparatus is arranged so as to permit of testing a plain end tube, with the ends open to the atmosphere and the interior of the tube exposed to view while under test. In this way the tube while under test is entirely relieved of any longitudinal stress due to the fluid pressure surrounding it. The sectional view, Fig. 2, shows clearly the construction of the cylinder. It will be observed that the tube is held in place within the test cylinder by steel centering rings, A, one at each end, while the cup leather packing rings are being slipped in place over the ends of the tube to be tested. This leather packing ring, at each end of the test cylinder, is backed by a cast iron ring, B , that fills the space, as shown, between the inner sur- face of the end of the test cylinder and the outer surface of the end COLLAPSING PRESSURES OF LAP-WELDED STEEL TUBES. 13 a < £ B Ph OD w . . § fc ^ 2 £ a B o B Ph 5 h B ft W E-i U P Q £ O H ° B B tf O g g * W B 0 £ 0 £ B I ^ GO 3 h 3 Pf o s o 0 a* 3 s 02 o B p H i o 05 3 o B 6 £ 0 ^ £ H H W Q 0 W 14 COLLAPSING PRESSURES OF LAP-WELDED STEEL TUBES. of the tube under test. This latter ring is held in place by means of a steel sleeve, C , engaging its outer surface by means of an in- ternal flange, and which is attached to the end of the test cylinder by means of the trapezoidal threads shown. A plug, D , was inserted, as shown, near each end of the experi- mental tubes for the purpose of preventing the centering ring and packing from being damaged and the attendant difficulty of re- moval of tube that might result from the tube collapsing in the end connections. Since a commercial tube is more apt to collapse at or near one end than near the middle of its length, this simple expedient made it possible to conduct the experiments without the frequent delays that would have otherwise resulted from the jam- ming of the tube in the end connections. This smaller test cylinder was placed over and was supported by the larger one. It was connected to the same set of pressure gauges, and was operated, in every essential respect, precisely as was the larger apparatus. In order to get, with the apparatus available for the purpose, a fluid pressure equal to the greatest working capacity of this cylin- der, it became necessary to couple up in series two hydraulic pressure pumps, each of 3,000 pounds capacity, so that the second pump could, if desired, deliver water to the test cylinder under fluid pressures up to 6,000 pounds per square inch. The highest pressure attained in this apparatus was 5,625 pounds per square inch fluid pressure, which was had while testing Hos. 476 and 477. Test Heads , Supports and Vents . — The different styles of test heads used, the manner of supporting the tube in the test cylinder, and the vent pipes connecting the interior of the tube under test with the atmosphere, are clearly shown in Tigs. 3 to 5. The Coupled Test Heads , as shown in Tigs. 1 and 3, were made up from short lengths of tubing of the same diameter and thickness of wall as that of the tube placed under test. One end of this test head was threaded like the tube under test, the two being connected by means of a standard sleeve coupling, in pre- cisely the same manner as two sections of the same tubing would be connected in practice, as, for example, in the case of a string of well casing. The other end of each of these test heads was closed by having a steel disk inserted into its end and welded in place, the closed end of the left-hand test head being drilled and tapped for the reception of the end of the vent pipe for maintain- ing atmospheric pressure within the tube under test, as shown. Air Vent COLLAPSING PRESSURES OF LAP-WELDED STEEL TUBES. Fig. 3. — Experimental Tube with Coupled Test Head shown in Position in 16" Hydraulic Test Cylinder. This Style of Head is Marked “ a ” in the Tabular Statements. 16 COLLAPSING- PRESSURES OF LAP-WELDED STEEL TUBES. This style of test head was used for all the tests of Series One and for such portions of Series Two as contain the letter “ a ” opposite in column 33 of the tabular statement of principal results of tests. This method of closing the ends of the experimental tubes, aside from the annoyance of an occasional collapse of the head itself, while entirely satisfactory in other respects, proved to be both slow and expensive to carry out with the facilities at hand for making up the experimental tube before testing, and for the removal of the heads after failure had occurred. Until it was discovered that the influence upon the collapsing pressure due to the tendency of the end connections of a tube to hold it to a circular form, ceased to be measurable, for a commercial tube, at a distance along its length from either end of from 3 to 4 diam- eters, this style of test head apparently possessed the merit of subjecting the tube under test to the same kind of end support as that actually existing in a string of well casing. After this fact was fully established the less expensive and otherwise more satisfactory methods below described were used. The Bolted Test Head , Fig. 4, was suggested by the appliance commonly used by tube works for the testing of tubes. In the commercial testing of tubes it is invariably the practice to subject the tube to an internal or bursting pressure; whereas, in connec- tion with this research, an external or collapsing pressure was applied. This test head (Fig. 4) consisted of a casting with a circular groove cut into its face for the reception of the plain end of the experimental tube. At the bottom of this groove was inserted suitable packing for the production of a water-tight joint when the head is firmly pressed against the end of the tube. The two through bolts shown were intended merely to hold the two heads in place and create sufficient initial pressure to prevent leakage at the start of the test, the external fluid pressure being relied upon, during the continuance of the test, for maintaining a tight joint between the test head and the end of the tube. These test heads were each provided with two small rollers for the purpose of making easier the handling of the experimental tubes while being inserted and withdrawn from the hydraulic test cylin- der. The left-hand head was drilled and tapped, as shown, for the reception of the end of the vent tube. The vent tube for con- stantly maintaining an atmospheric pressure within the tube under COLLAPSING PRESSURES OF LAP-WELDED STEEL TUBES. 17 CD E-i rH & g go 2 a & & co W COLLAPSING PRESSURES OF LAP-WELDED STEEL TUBES. tion with this research on the collapsing pressures of tubes. Al- together about 6,000 of these autographic records have been made and are on file. Collapsing Tests, Series One, Showing the Influence of Length of Tube on the Collapsing Pressure. Since the influence of the length of tube, between transverse joints or end connections tending to hold it to a circular form, upon the collapsing pressure appeared to be the most uncertain element entering the problem, it was thought best, first of all, to determine the precise nature of this influence. Accordingly, it was decided to make a series of tests on a single diameter of tube for all the commercial thicknesses of wall obtainable, in five dif- ferent lengths of from 2J to 20 feet. Selection of Tubes for Testing . — The tubes used in making this series of tests, as well as the other series contained in this paper, were obtained from the National Department of the National Tube Company, McKeesport, Pa., on order issued by the Job Work Shop, in the usual commercial way. Those who filled these orders had no means of knowing for what purpose the tubes were to be used, and presumably, therefore, the tubes thus obtained for pur- poses of testing represent fair samples of the regular commercial product of the mills. Every tube thus obtained, without exception, was tested, the complete results of all tests being recorded in the Log, a summary of which appears in this paper. The results may therefore be accepted as indicating the strength to resist fluid collapsing pres- sures of this Company’s Bessemer steel lap-welded tubes, the tubes being taken just as they are found in stock. Diameter of Tube Tested . — For Series One it was decided to use 8J-inch well casing, which has a nominal outside diameter of 8f inches. This size was adopted because, taking all things into consideration, it seemed to afford the greatest opportunities for getting at the results desired. The various diameters of the individual tubes of this series are given in columns 2, 3, 4, 5 and 34 of the tabular statement of principle results of tests, Figs. 11-15. (See folders.) The nominal outside diameter , in inches, appears in column 2, and is for this series 8.625 inches for all tubes tested. COLLAPSING PRESSURES OF LAP-WELDED STEEL TUBES. The average outside diameter, as made up from measurements on each individual tube, at intervals of one foot along its entire length, are entered in column 3. These measurements were made by means of an especially constructed steel tape, the spacing of whose graduations bore the same relation to those of an ordinary scale divided into inches and hundredths as the length of the cir- cumference of a circle bears to its diameter. That is to say, each inch division on the tape was actually 3.1416 inches long. By this means diameters could be read directly from circumferential meas- urements, thus making it possible, by means of a single reading, to obtain an average of all the different diameters at any particular foot length of the tube. The advantage of this method will be appreciated when it is remembered that more than 5,000 deter- minations of mean diameters, at the different cross-sections, one foot apart, had to be made for Series One and Two of this investi- gation alone, the tubes tested in every case not being perfectly round at any of these sections. The greatest and least outside diameters, at the place of collapse, by which is meant that point of the length of tube where, after failure, the distortion was greatest (see Fig. 16, page 30), are en- tered respectively in columns 4 and 5. These entries were made up from the measurements made for out-of -roundness of tube, at each foot along its length, before being placed in the hydraulic test cylinder. Thickness of Wall. — There were five nominal thicknesses of wall tested in Series One, namely: 0.180, 0.229, 0.271, 0.281, and 0.322-inch, having nominal weights of respectively 16.07, 20.10, 24.38, 25.00 and 28.18 pounds per foot length for the outside diameter of 8f inches chosen for this series. The actual plain- end weights per foot corresponding to these nominal thicknesses were respectively 16.23, 20.53, 24.18, 25.04, and 28.55 pounds. The nominal thicknesses of wall of the different tubes tested appear in column 6 and the corresponding nominal weights in columns 13 and 34. The average thickness of wall of the tubes of this series appears in column 7. This average thickness for each tube was calculated from the plain-end weight, length, and average outside diameter, as given in column 3. In this way a more exact value could be arrived at for the average thickness than by any other practical means. 28 COLLAPSING PRESSURES OF LAP-WELDED STEEL TUBES. The greatest and least thickness of wall at the place of collapse are given respectively in columns 8 and 9. These were obtained from the tube, after collapse, by cutting it across at the point of its length where the distortion appeared to be greatest, after which the greatest and least thicknesses were measured by means of a micrometer caliper. Weights of Tubes. — The nominal weights, in pounds per foot length, of the tubes tested are given in columns 13 and 34. For this series, having the uniform nominal outside diameter of 8f inches, these nominal weights were 16.07, 20.10, 24.38, 25.00 and 28.18 pounds per foot length. The actual plain-end weights cor- responding to the nominal thicknesses of wall were respectively 16.23, 20.53, 24.18, 25.04, and 28.55 pounds per foot. The actual plain end weights per foot length of the individual tubes tested are entered in column 14. The entries in this column were made up by dividing the weight in pounds of each tube by its length in feet, as given in column 11. The weighing was done on a tested platform scale, and, for those tubes that were threaded before weighing, an allowance was applied for the loss of weight due to threading, this allowance being arrived at experimentally by weighing a number of pieces both before and after threading. In this way the corrections for the different styles of thread were arrived at. Lengths of Tubes . — The tubes for Series One were ordered in five lengths for each of the five thicknesses tested. These lengths are entered in column 10, and were 20, 15, 10, 5, and 2^ feet, including the threaded ends, for this series. For the tubes of all the other series contained in this report the length ordered in each case was 20 feet, for both plain and threaded ends. It will be noted that the groups 'Nos. 26 to 34 inclusive, and 77 to 79 in- clusive, were supplied in random lengths, presumably, because the stock did not contain tubes of sufficient length, at that date, to fill the order in 20-foot lengths for tubes of these particular weights. All the other tubes were supplied in substantially the lengths as ordered. The actual lengths of the tubes tested, to the nearest thousandth of a foot, as measured by means of a steel tape, are entered in column 11. These measurements for this series include both threaded ends, the coupling that is usually shipped as a part of every threaded tube, and which is ordinarily measured up as a part of its length, not being included in these measurements. The measure- COLLAPSING PRESSURES OF LAP-WELDED STEEL TUBES. 29 ments on the tubes of the other series that have threaded ends were made in the same manner. The unsupported lengths of the tubes are entered in column 12. These were arrived at by subtracting the lengths of the portions of the threaded ends that lay inside the couplings from the corre- sponding actual lengths as given in column 11. Column 12 then shows the actual length of tube exposed to a fluid collapsing pres- sure, and which, at the same time, received no direct supporting action from any outside source tending to hold it to a circular form. These were the lengths used in deducing the general con- clusions from the individual experiments, as shown in Fig. 21. Collapsing Pressure . — In and near the region of expected col- lapse the hydraulic pressure within the test cylinder and surround- ing the tube under test was increased at the rate, in pounds per second, shown in column 17. This rate in every case was low enough to permit of making accurate readings of the fluid pressure exerted upon the tube under test, and also allow for free elastic deformation of the material constituting the walls of the tubes. In no case was the elastic limit of the material exceeded until after failure had actually occurred. The apparent stress on the wall of the tube at instant of failure ranged from about 7,000 to 31,000 pounds per square inch, respectively, for the relatively thinnest and thickest walls in Series Two. (See page 73 and Fig. 51.) The fluid collapsing pressure , in pounds per square inch, are entered in column 15, the gauge from which the pressure was read being indicated opposite in column 16, where B, C and D designate respectively the 1,000, 3,000 and 8,500 pounds capacity Shaw differential-piston mercury gauges. These gauges were fre- quently compared, and the slight differences were adjusted so as to make the readings on B and D conform to those on C, the intention being to have gauge C calibrated after completion of the tests. Collapsed Portion. — The appearance of the tube after being col- lapsed in the hydraulic apparatus is clearly shown by the photo- graphs and by the collapsed sections, examples of which are shown in Figs. 16 and 17. These photographs of collapsed tubes , taken in conjunction with the collapse sections, show very clearly the precise nature of the distortion resulting from the subjection of the tube to an external fluid pressure sufficient to cause failure. Referring to Fig. 16, which is a reproduction of the photograph of Nos. 50 to 54 in- 30 COLLAPSING PRESSURES OF LAP-WELDED STEEL TUBES. Fig. 16 .— Photograph of Tubes, Nos. 50 to 54 , after being Collapsed. COLLAPSING PRESSURES OF LAP-WELDED STEEL TUBES. 31 32 COLLAPSING PRESSURES OF LAP-WELDED STEEL TUBES. elusive, it will be observed that two views of each tube are shown. One of these was taken looking in the direction of the axis of col- lapse, while the other view was taken after the tubes were rotated on their supports through an angle of 90 degrees. These tubes were all calipered for out-of -roundness, at distances of one foot apart along their entire lengths, before being placed in the hydraulic test apparatus, and, while doing this, the ends of the greatest and least diameters at each of these sections were marked on the tube by means of plus (+) and minus ( — ) signs, as shown on these photographs. Where noughts (0) appear the tube was so nearly round as to make it difficult to distinguish a greatest and least diameter at that section, even with the ex- ceedingly refined methods in use for making this determination. The length of collapsed portion of tube, in feet, is entered in column 18. This length was determined, after collapse, by meas- uring the length of the portion of the tube which showed a per- manent distortion — for example, referring to Fig. 16, test num- ber 54, it appears that this permanent distortion just becomes measurable at 4£ feet, is greatest at 7\ feet, and terminates at 10^ feet. In this case, then, the length of collapsed portion is 6 feet, which is somewhat less than one-third the length of the tube. More than two-thirds of the length of this tube has suf- fered no permanent distortion whatever. This localization of collapsed region is typical of all tubes tested in relatively long lengths. The lengths of the collapsed portions expressed in diameters of the tubes were obtained by dividing the length of collapsed portion of each tube by its outside diameter, both being expressed in inches. The distance of collapsed portion from the end of tube , column 20, was obtained by measuring the distance from the point of greatest distortion — for example, at 7\ feet for test number 54, Fig. 16, to the numbered end of tube, as shown in the photographs. For further discussion of this matter see page 75. The angular distance from weld , column 21, was obtained by measuring the angular distance from the axis of collapse (see Fig. 17) to the weld. Assuming that the observer is stationed at the numbered end of the tube, and looking in the direction of its length, angles measured in the direction in which the hands of a clock rotate are marked plus ( -f- ), while those measured in the opposite direction are marked minus ( — ) . COLLAPSING PRESSURES OF LAP-WELDED STEEL TUBES. 33 These angular distances were measured by means of an espe- cially constructed tape. On this tape distances were laid off equal to the circumference of the different sizes of tubes to be tested. These distances were then divided into 360 equal parts, each of which would represent the length of one degree of arc of the cir- cumference of the tube. It is evident that a tape constructed after this mannner affords a most satisfactory means for measuring an- gular distances around tubes. In this way the original angular distance between two points on the surface of a tube, lying in the same transverse plane, can be measured just as readily after the tube has been distorted as before. When it is considered that all of these measurements for angular distance from axis of collapse to the weld had to be made at the place of greatest distortion, after collapse had taken place, the utility of this special tape will be apparent. Physical Properties of the Steel . — The physical properties en- tered in columns 22 to 25, inclusive, are the averages from three test specimens cut from each tube, after removal from the hy- draulic test apparatus, the test specimens in every case being cut from the undistorted portion, except for the cases, clearly noted in the tabular statement of Series Two, where these specimens were cut from the distorted portion. Tor these latter it will be observed that the yield point is raised and at the same time the elongation and reduction of area are lowered, more or less according as the portion from which the specimens were cut were more or less distorted. In working up the result for this paper it was assumed that the material of these tubes possessed the same average physical properties as the others of the same series. This seemed but a fair assumption to make under the cir- cumstances, since there was no apparent reason why the material constituting them should differ in respect to physical properties from that of the other tubes tested at the same time. All the specimens for the physical tests were cut lengthwise of the tube and were pulled in the testing machine without being previously subjected to any straightening action whatever. The test specimens were substantially of the form and dimensions as that adopted for plate metal by the American Section of the International Association for Testing Materials, namely: eight inches between extreme gauge marks, one and one-half inches wide throughout the gauged portion, and enlarged to two inches width at the ends where held in the grip of the testing machine. 34 COLLAPSING PRESSURES OF LAP-WELDED STEEL TUBES. The physical tests were made at the National Tube Company’s Laboratory, McKeesport, Pa., under the immediate direction of their metallurgist, Mr. G. M. Goodspeed. Chemical Analysis . — The chemical analyses, columns 26 to 31 inclusive, were also made at the Tube Company’s Laboratory, from drillings taken from the tube or from the ends of the physical test specimens. m Material . — The kind of material, whether Bessemer steel, open- hearth steel, or wrought iron, constituting the different experi- mental tubes, as entered in column 32, was determined for each case by means of the chemical analysis. It will be observed that the experimental tubes, with but few exceptions, were composed of Bessemer steel. In Series One, three of the tubes tested proved to be wrought iron and also three open- hearth steel, all the others of this series being Bessemer steel. In Series Two, one group of five tubes proved to be wrought iron. The Bessemer steel constituting the tubes had the following average physical properties: Tensile strength, pounds per sq. inch 58,000 Yield point 37,000 Elongation in 8 inches, per cent 22 Reduction of area, per cent 57 And the following average chemical analysis : Sulphur, per cent 069 Phosphorus, per cent 106 Manganese, “ “ 35 Carbon, “ “ 074 Deduction of Law Showing the Relation of Collapsing Pressure to the Length of Tube. Regrouping of Tests . — In order to arrive at the law expressing the collapsing pressures of Series One in terms of the length of tube and the different thicknesses of wall, for the diameter chosen, it was necessary,, first of all, to arrange the tests of this series in the manner shown in the table, Pigs. 18-20. In this table all the values entered in the different columns, except the last three of the group of four columns headed “ Collapsing Pressure,” are taken directly from the corresponding columns of the table of “ Principal Results of Collapsing Tests, Series One,” Pigs. 11-15. It will be observed that the tests have been regrouped so that, COLLAPSING PRESSURES OF LAP-WELDED STEEL TUBES, 35 EFFECT OF LENGTH ON COLLAPSING PRESSURE . — Abstract from Logof Tests Conducted by Prof. R.T.Stewart, /50Z-0+, on National Tube Co/s Lap- welded Bessemer Steel Tubes in Lenqths of 2'k,5,IO,lS,andZO Feet,? %" Outside Diameter, to Determine the. Effect of Length of Tube on the Collapsing Pressure; to which is Added a Comparison with Values Read from the Curve. M, Represent- ing the Average Results of these Tests. Mode by e.e.s. under direction of a. r s. /9os. rests Grouped and Arranged in Order of Length and Thickness of Tube. Actual Thickness of Wall, Unsupported Actual Collapsing Pressure Designation of Tube, Number Outside Inc hes. Length Plain End Pour, c/s per Sguare Inch. of Test. Diameter, Nominal Computed of Tube. Weight, Observed Corrected to From % Vo not/ on cts Reported. Inches. from W 9 f feet. Lbs. per Ft. Nom. Thk. Curve FI. from M. zz 8.057 0.108 2.213 15.23 8/5 990 - 3 29 8.0S5 0.170 2.2/2 IS.NI 1085 1/85 12 2 2/xPoet Lengths. 23 8. CNN 0.180 0.181 2.220 I0.NI 1685 /d 75 no +// 8 '/j,‘ Casing, 10.07 lbs. 25 8.00! 0.181 2.205 I0.N3 985 905 - 0.5 2/ 8.058 6.182 2.198 / 6.N5 9/5 895 - 7.5 Average, 8.050 0.170 2.2(0 15.99 977 101 2 t N.N ’ /9 8.005 O.S73 N. 7/8 /5.0N 525 005 t 6 /r 8.601 0.180 N.702 / 0.3N 5N0 SNO - 5 5 Foot Lengths. n 8.059 0.180 6.181 N. 768 /0.N2 575 SOS 570 - / u 8.050 6.182 N.08% /0.N3 6/5 590 t 3.5 8 '/jf Casing, 16.07 lbs. 20 8.05N 6.182 N.763 /0.N3 705 085 tzo Average 8.058 0.186 N.70N 10.25 592 597 t N.7 /9 8. CNN 0.171 9.095 15.5 / N55 S55 t 6 IS 8.006 0.177 9.09 S 10.03 590 025 i/9 / 0 Foot Lengths. is 8.05/ 0.180 0.178 9. 765 IC.II 550 575 525 t 9.5 !Z 8.05N 0.180 9.09 1 10.26 570 570 t 8.5 V'/tf Casing, 10.07 lbs. II 8.002 0./8N 9.7/5 10.05 575 535 t 2 30 8.070. 0.229 0.197 11.950 / 7. 79 050 NOS 520 -10.5 8'/h Casing, 20 . to lbs. /W«ra,« 8.059 0.181 / 0.075 10.39 505 559 t S7 7 8.05/ 0.I7N IN. 7 23 15.76 N25 N90 - 2 •? 8.055 0.183 IN. 7 62 / 0.5N 550 520 t N IS Foot Lengths. C 8.051 0.190 6.180 IN.708 / 0.8 2 575 510 500 t 2 10 8.052 6.188 IN. 700 10.95 S80 995 - / 8 ’/jy" Casing, 16.07 Uos. 9 8. OSS 6.19/ IN. 768 17.21 010 995 - / Ave rage 8.053 0. 189 IN. 7 68 10.00 sm 502 + 0,9 / 8.057 0./70 19.818 15.92 N50 990 t 3 9 8.0NO 6. 183 19.762 /C.SN N50 920 N75 -U.5 20 Foot Lengths. 3 8.CNI 0.120 6. no 19.095 / 0.77 535 9 75 0 S%" Casing, 16.07 tbs. Z 8.03 7 0.191 19.0 90 17.2 N 025 5/5 f 2S S 8.038 6.191 19.769 / 7. 23 02 0 505 t O.S Average 8.CN3 0.185 19.72 N U.7 9 530 981 + /.3 NS 8.050 0.2/0 2.188 18.93 I2N0 1935 - / N2 8.077 6.2/0 2 .200 19.02 1353 / 595 1950 t 0.5 2 Zz Foot Lengths. V? 8.055 0.229 0.2/2 2.2/8 19.10 /30S 198 0 t z NO 8.0N8 0.2/3 2.220 19.18 1330 1995 + 3 VZyCasinq, 20. /0 lbs. 97 8.057 0.2 IN 2,193 19.20 I3N0 1995 t 3 8.6S7 0.2/2 2.265 19.10 13/N i*no t 2.7 NN 8. UN 0.207 N.7II 18.00 910 1/05 f Z 9/ 8.053 0.208 9.080 /8.7N 97 0 1215 t 6 S Foot Lengths. 92 8.670 0.2 29 0.2/0 N.087 18.93 80S 1030 UN 5 -/o V/ii Casing, 20.10 lbs. NO 8.657 0.210 N. 7/3 / 9.N8 975 1125 - z N3 8.00/ 6.2/9 N.080 19.72 875 995 -13 Average 8.00/ 0.2.12 N.C95 19. // 90 7 1100 - 3.N 38 8.000 0.268 9.7/6 18.77 7 00 930 /07S -/ 3.S 32 8.000 6.20 9 l/.NSI 18.86 73 0 956 toss -•0 31 8.058 0.2/0 n.8/0 18.90 8/0 / 020 1050 - 3 35 8.005 6.2/2 9.701 19. n 886 107 0 /07S - 0.5 !0 Foot Lengths 30 8.003 0.229 0.2 IS 9.702 /9.3V 8NZ 1000 " - 7 . 39 8.088 0.217 9.705 19.59 895 920 " - 9 9/f“Ca si ng, ZO./O lbs. 37 8.005 0.220 9. 7 6/ 19.82 9C0 /600 " - /.5 33 8.0 ON 0.220 II.3LN 20.38 8N0 875 /OSS - / 7 39 8.009 0.232 11.300 20.89 900 925 “ - IZ5 8.006 0.2/7 10.502 19.51 89 1 979 -8T.Z ! 2 3 4 5 6 7 8 9/0 Fig. 18 , 36 COLLAPSING PRESSURES OF LAP-WELDED STEEL TUBES. EFFECT OF LENGTH ON COLLAPSING PRESSURE. Abstract from Log of Tests Conducted by Prof. R.T.Stewart, 1902-04, on National Tube Co.’s. Lap- welded Bessemer Steel Tubes in Lengths of Z'/x,5,IO,l5,and20 Feet^Vs'Outside, Diameter ; to Determine the Effect of Length of Tube on the Collapsing Pressure-, to which is Added a Comparison with Values Read from the Curve, M, Represent- ing the Average Results of these Tests. node by e.e. s.under direction of n.r.s., isos. rests Grouped on d 4 rranqad in Order of Langth arid Thick ness of Tube.. Number of Test. Actual Outside Diameter. Inches. Thickness of Wall, Inches. Unsupported Length Feet. Actual Plain End V/eight, Lbs. per Ft. Collapsing Pounds per Pressure,. Square Inch. Designation of Tube, as Report e d. Nominal Computed from Wgf. Observed Corrected to Nom. Thk. Fro no Curve M. % Variation from M. 29 5.659 0.2/3 12.636 n.23 150 10/5 - 25 29 9.6 50 0.229 0.2/3 12.372 19.1 6 7 SO 115 !0 80 -12.5 !5 Foot Lengths. 27 9.629 0.233 13.373 20.15 ms / 075 - 9 I'/j/" Casing , 20.10 /bs. A verage 9.696 0.220 12.792 H.79 90S 1 0 02 - 6.3 20 Foot Lengths 26 9.609 0.229 0.2/9 19.966 19.57 no 170 910 - / 15v"Cq si nq , Zo./O/hs. 70 9.673 0.27 1 0.253 2.197 22.72 1975 / (75 - / 3.5 r'/q Casing, 2 H.3* lbs. 93 9.65J 0.22/ 0.259 2 .156 23.23 ms 2/ 25 /// If" Lind Pipe, ZS. 00 tbs. 97 2.675 0.21/ 6.2 6/ 2./SI 23.95 11/5 2030 1 12 U .. .. it *f 16 9.656 0.Z9I 0.261 2/96 29.00 1190 19 75 t 2 /# » /* n «i 73 1.690 0.27/ 0.269 2.136 29.15 1150 19 70 /93S - 3 C asin g ,2*.3S lbs. 7/ 9.650 0.27/ 0.27/ 2.196 29.20 1930 1930 0 ft H II » 72 9.662 0.27/ 0.273 2.626 29.97 mo /1 55 - 5 .1 M «l M 79 1.699 0.271 0.279 2 .196 19.50 /71S 1750 - 9.5 *« M II M 9S 9.699 0.211 0.2 71 2.159 29.92 1765 /(IS -/3 Line Pipe, 25. JO lbs. 19 9.639 0.291 0.279 2.196 29.10 2200 2/10 t 1.5 Z'/ z Foot Lengths. Average 9.659 0.269 2.135 29.09 /172 noo - 1.0 Corrected to Nom. Thk $.27/ as 9. 69 9 0.27/ 0.257 9.631 22.91 1395 IS/ 0 - 9 X'/q" Casing, 2 *.3? lbs. cc 9.653 0.27 / 0.261 9.(9/ 23.17 1520 15(0 - 6 0 •• .1 ii £9 1.670 0.271 0.261 9.651 29.02 7725 IKS + r II 1. H || 6 7 9.65/ 0.271 0.269 9.69/ 29.07 1690 1670 t 0.5 II M I. |» 92 9.679 0.21/ 0.2 73 9.696 2 9.50 2030 zoos /(6S t2l IT Line Pipe,Zs.fO 1 bs. 9! 9.669 0.211 0.275 9.(99 2 9.6/ 1(25 / 570 - 5.5 i« i. i. i. i. £5 9.656 0.271 0.277 9.626 29. 16 / 750 1695 / / TfVii Casing , 2V.3* lbs. 11 1.65! 0.211 0.279 9.69/ 29.13 1615 1525 - 1.5 T" Line Pipe, 2S./0 lbs. 90 9.615 0.21 1 0.219 9.(96 25.95 noo 1556 - 7 ii ii ii i. ii 99 1.699 0.21/ 0.217 9.(36 25.75 / 615 /915 -to 5 Foot Lengths Average 9.669 0.279 9.691 2 9.5/ / 619 1653 - /.5 Corrected to Nom.Th k.0.21 1. 15 1.669 0.21/ 0.260 9.696 23.2/ / 575 1615 *■ 9.5 V" Line Pipe , ZS./O lbs. 69 9.657 0.27/ 0.260 9.666 23.2/ /950 / 570 t /S V'/S Casing , ZH.sv/hs. a 9.669 0.2 7/ 0.265 9.656 23.7 5 //SO 1215 -21.5 I# #. ii 63 9.662 0.2 7/ 0.267 9.69/ 23.19 1595 /625 t s ii ii // // 62 1.679 0.27/ 0.261 9.636 29.05 1621 1660 t 7 ii • ii /i // S3 9.669 0.211 0.270 9.(51 29.20 1990 /9S0 1550 - (.5 T "Line Pipe , ZS. to lbs. 97 1.669 0.211 0.272 9.(56 29.90 / 695 1635 t S.S n o, a ii ii *9 9.695 0.21/ 0.275 9.699 29.72 /500 I960 - 6 ii ii /i ii ii 60 2.673 0.2 7/ 0.276 9.596 29.75 1950 / 795 t/6 9'/S Casing , ZN.39 lbs. S <6 1.663 0. 21/ 0.271 9.(36 29.95 /S95 /970 - 5 ** Line Pipe,, ZS./O lbs. III 1.(92 0.322 0.217 9.(30 25.51 1725 /550 0 109 2.652 0.322 0.29/ 9.690 26.00 1105 /615 f 1 /0 Foot Lengths. Average 8 .666 0.272 9.692 29.90 1583 15(7 t 1.2 Corrected to Nom. Thk. 0.271 $7 9.695 0.27/ 0.263 19.6 56 23.59 1590 K7S + 13.5 9 /T Casing , 29.39 /hs. 129 1.699 0.29/ 0263 19.(91 2 3 .56 1250 1335 - 9.5 S'" Line Pipe, ZS. to ibs. *79 1.669 0.211 0.266 17.916 23.19 1925 1936 — — 90 9.656 0.25/ 0.266 19.(96 23.95 133 5 1990 - 2.5 » •• •< •• •• 8 Z 2.662 0.21/ 0. 27/ /9.136 2 9.27 /990 199 0 *- / •• S9 9.662 0.27/ 0.272 19.(16 29.33 mo 1700 1975 t/S I'/,’ Casing, 29.31 lbs. 55 2.690 0.27/ 0.273 19.(36 29.35 / 51 0 ' 15(5 F 6 ii •• •; •• 56 9.650 0.27/ 0.275 19.(96 29.55 1920 / 310 - 6.5 9! 9.666 0.29/ 0.276 19.(91 29.73 13 05 1255 -IS S'" Line. Pip e, 2 5.10 Has. 59 9.(65 0.27/ 0.271 19.656 29.17 139 0 13/5 -// 9/.," Casing, 29.31 /bs. Id/ 9.695 0.322 0.296 19.630 2 5.11 nzs 1525 t 3.5 1" Line Pi/oe, 29. H lbs. IS Lengths, to Nom.Thk.o.l7L Average. 1.653 0.273 19.690 29.3J 193 5 19(1 - 0.5 ■ >r "See 20 Foot Lengths. 1 2 3 4 5 b 7 8 9 !0 u Fig. 19. COLLAPSING PRESSURES OF LAP-WELDED STEEL TUBES. 37 EFFECT OF LENGTH ON COLLAPSING PRESSURE— Abstract from Loq of Tests Conducted by Prof. R.T. Stewart, 1302-04, on National Tube Co.'s La/o- vv elded Bessemer Steel Tubes in Lenqths of ZZ*, 5, 10, IS, and 20 Feet, *%“ Outside Diameter, to Determine the Effect of Length of Tube on the Collapsing Pressure-, to which is Added a Comparison with Values Read from the Curve, M, Represent- ing the Average Results of these Tests. Made by r.e.s. under direction of R.r.s.jsos. Tests Grouped and Arranged in Order of Length and Thickness of Tubes. Number of Test Actual Outside Diameter, Thickness of Wall, Inches Unsupported Length of Tube Feet. Actual Plain End Weight, Lbs per Ft. Collapsing Pounds per Pressur e Square Inch. Designation of Tube, Q$ Reported. Nominal Computed from Wgjt. Observed Corrected to Nom.Thk. From Curve M %Variation from M. Si 7.004 0.27 f 0.259 19.049 23.13 1320 1450 t 2 7/y Casing, 24.39 ibs. 54 9.000 0.27/ 0.202 19.034 23.5Z 1495 1550 H/.5 •> i. j. 99 9.003 0.322 0.204 19.040 23.00 1375 1450 * 2 9" Line Pipe, 22.19 lbs. * 7 9 9.004 6.Z 9/ 0.200 17.990 23.94 1425 1496 t 2.5 •« • •• ,25.60 tbs. SO 9.000 0.27/ 0.27/ 14.040 24.29 1435 1435 t I 9/e Casing , 24.39 lbs. 5 3 9.000 0.Z7I 0.272 19.039 24.32 1520 1516 1420 t 6.5 „ It, 7 0 9.00 3 0.29/ 0.272 19.034 24.39 1410 1406 - 1.5 7" Line Pipe, 25.66 lbs. St 9.009 O.Z7 f 0.274 14.034 24.5 V 14 30 1395 - 2 V'/j'C a sing, 2 . 4 . 39 ibs. 7 S 9.040 0.291 0.274 19.044 24.44 1375 1345 - 5.5 7 " Line Pipe, 25-66/ bs. *7 7 9.000 0.29/ 0.274 19. SOI Z4.47 1275 1245 -12.5 1 , .. •• *7 9 9.000 0. 29/ 0.290 19.340 25.69 1256 1145 - 26 9.000 0.ZC9 19.042 24.05 IN 19 1440 / 1.2 Z6 Lenqths.CorrectedbThk.ZlI. Klron, not in overages. 119 9.045 4.29V 2.159 20.22 2450 2 7 96 tl 6.5 Hi 7.045 0.249 2.100 20.00 2305 20 55 t S.5 2/^Foot Lengths. HI 9.0S7 0.322 0.314 2.170 29.43 2240 2325 ISIS - 7.5 //? 9.04! 0.322 2.140 29.06 2490 2496 - / 7" Line Pijae,Z9./9 lbs. no 9.040 0.323 2.150 29.04 2340 2390 - S.S 9.040, 0.31 1 2.157 27.70 2347 2 520 t ■ 0.4 ns 9.0S9 0.247 4.045 2 0.45 1496 2270 - / 1 / 7 9.034 0X44 4.041 2 0.55 2325 2595 //3 S Foot Lengths, 113 9.043 0.322 0.307 4.040 27.36 1795 1470 22 90 ' 14 IIP 9.073 0.367 4.035 27.43 2046 2220 - 3 7“ Line Pipe, 29/9 lbs. I/O. 9.003 0.321 4.056 29.00 2225 2240 - 2 7.0S4 0.300 4.042 27.29 2473 2254 - 1.4 no 9.020 0.300 4.04b 27.20 2655 2235 - 4.5 tnz 9.07Z 0.3ZZ 0.314 9.045 29.04 1595 10 75 2/ 35 -Z/.5 16 Foot Lengths. t/OS 9.05Z 0.325 9. 040 29.96 1790 1750 * ■ 4 7"Line Pipe,29./9 lbs. 9.0ZC 0.300 4.040 27.26 2653' 2235 - 4.S tO.H. Steel, not in averages. 123 9.005 0.291 0.247 14.041 20.49 1075 1440 - 4.5 7" Line Pipe, ZS. 66 lbs. /OS 9.049 0.322 0.24 7 14.045 20.40 1526 1795 - 12.5 29. /Hbs. 10 0 9.072 0.322 0.309 14.035 27.49 1095 1935 2035 - 9.5 ., „ 1 , 103 9.04 1 0.322 0.3Z6 14. 040 29.39 2606 2625 - 6 .S • « *. 10 7 •/.coo 0.322 6.324 14.036 29.94 2025 2 005 - t.s 9.059 0.304 14.039 27.54 179/ 1919 - 5.0 IS Foot Lengths. 101 9.047 0.244 1 4.04 / 20.27 7 050 1935 - 1.5 too 9.040 0.297 14.045 2 0.44 mo 1970 t ■ 0.5 424 9.07 / 0.302 14.995 20.49 1575 1790 - 9.5 423 9.044 0.363 19.792 27.06 1930 2025 * 3 20 Foot Lengths. 42! 9.000 0.3ZZ 0.303 19.700 27.64 19 35 2135 1905 / 9.5 I0Z 9.050 0.363 14.055 27.66 1400 2106 t/6 9" Line Pipe, 29.15 ibs. 422 9.07Z 0.36 7 19.705 27.34 IC3S 1790 - 4 t9 9 9.000 6.309 19.030 27 .44 1735 1540 - V 420 9.059 0.3/6 19.975 Z7.C4 1905 1936 - 1.5 9.0 59 6.362 19.753 zo.?r I7CZ 1400 F 0.1 / 2 3 4 5 6 7 9 9 10 II Noter To obtain collapsing pressures for curve N , multiply corresponding tabular collapsing press- ures from curve M by 0.944 . Curve Mis based upon Series One only, wh He curve /V is based jointly upon both Series One and Series Two. Fig. 20 . 38 3000 2900 2800 2700 2600 2500 2400 2300 2200 2100 2000 1900 1800 1700 1600 1500 1400 1300 1200 1100 1000 900 800 700 GOO 500 400 300 200 100 1 Fig n t test all 1 COLLAPSING PRESSURES OF LAP-WELDED STEEL TUBES. 21. — Chart showing Relation of Length of Tube to Collapsing Pres- sure for National Tube Co.’s Lap-welded Bessemer Steel Tubes. Based upon Tests on Lengths of 2$, 5, 10, 15, and 20 Feet, Con- ducted by Prof. R. T. Stewart, 1902-4. le lines marked M show the relation of length to collapsing pressure, from } on tubes 8f" O.D., and the lines N, 5.6% below, this relation as based on ests on outside diameters from 3 to 10 inches. COLLAPSING PRESSURES OF LAP-WELDED STEEL TUBES. 39 for each of the five lengths of tube tested, the actual average thick- ness of wall of each experimental tube shall fall in the group having the nearest nominal thickness of wall. Collapsing Pressure Corrected to Nominal Thickness. — In the seventh column, or the first of those headed “Collapsing Pres- sure/’ Pigs. 18-20, are entered the observed collapsing pressures. These pressures, of course, in each case correspond to the actual thickness of the tube. Since the actual thicknesses of different commercial tubes of the same nominal thickness of wall vary somewhat in practice, as is evident from running the eye down column 4 of this table, it became necessary, in order to get strictly comparable results, to obtain from the observed collapsing pres- sure of each experimental tube, whose thickness of wall did not equal exactly the nominal thickness, a collapsing pressure that would correspond to this nominal thickness. That is to say, the observed collapsing pressures corresponding to the respective thick- nesses of wall tested were corrected, so that each would represent what the collapsing pressure would have been had the tube had the exact nominal thickness instead of that tabulated in column 4. These collapsing pressures, having been thus corrected to cor- respond to the nominal thickness of wall, are entered in column 8, or the second of those headed “ Collapsing Pressure.” This correction was made graphically by first plotting, to a ver- tical scale representing collapsing pressure and a horizontal scale representing thickness of wall, the results of the tests for all the different thicknesses of wall for each of the five lengths of tube; second, drawing the mean line representing average results ; and, third, drawing a line parallel to this mean line through the plotted value for each tube intersecting the ordinate corresponding to the thickness of that tube. The collapsing pressure corresponding to this point of intersection was then read and recorded in column 8. The Collapsing Pressure from Curve M for each nominal thickness, in the five different lengths tested, are entered in column 9. These were read from a chart similar to Fig. 21, but drawn to a much larger scale. The variation of the values given in column 8 from the corresponding values in column 9, in per cent., are given in column 10. This column shows at a glance how the individual tests corrected to nominal thickness differ from the mean values as read from curve M. When it is considered that these were the ordinary commercial lap-welded wrought-tubes, selected at random, and subjected to an external 40 COLLAPSING PRESSURES OF LAP-WELDED STEEL TUBES. fluid pressure, it is surprising that there should be such a slight variation. It will be observed that the greatest individual vari- ation does not exceed 22 per cent., while the greatest variation among the group averages is 8.2 per cent., the greater portion of these being less than 5 per cent. The average variation of the group averages is 0.2 per cent. Theoretically, of course, this should be zero, but the very small value of 0.2 per cent, obtained serves as a satisfactory check upon the accuracy of the mean values read from curve M. Relation of Collapsing Pressure to Length of Tube. — This re- lation is clearly shown in Fig. 21 for 8^-inch casing (8f inches outside diameter) in the four nominal thicknesses of 0.180, 0.229, 0.271, 0.322 inch, and for lengths of from 2 to 20 feet between the regular screwed couplings. On this chart the combined circles and crosses represent the dif- ferent plotted averages contained in column 8, Figs. 18-20. By means of these plotted values the curves marked M were con- structed. The spacing of these curves was adjusted so that when the values of the table, Figs. 18-20, were also plotted to collap- sing pressure and thickness, for each of the five lengths tested, the resulting curves were smooth. By this method of cross-plotting the individual experiments on 8f-inch outside diameter tubes, the four curves marked M were obtained. Each of these curves, then, is not based only upon the group averages belonging to it, but is also based upon the group averages belonging to the other three curves similarly marked. The curves marked N were obtained by adjusting the curves M so as to harmonize with the most probable values for collapsing pressure as based upon Series Two. The difference, it will be observed, is small, the different curves H being only 5.6 per cent, below the corresponding curves M. Discussion of Curves M and N. — An inspection of Fig. 21 dis- closes the fact that for this size of tube, especially for the thinner walls, there is a marked dropping off in collapsing pressure as the length of tube is increased up to about 4^ feet, or until a length equal to about 6 diameters is reached. Beyond this point there appears to be no further material decrease in the collapsing pressure. For example, from curve M, for a thickness of 0.180 inch, the collapsing pressures for lengths of 2, 4J and 20 feet are respectively 1,050, 570, 480 pounds. That is to say, an increase in length from 2 to 4J feet diminishes the collapsing pressure by COLLAPSING PRESSURES OF LAP-WELDED STEEL TUBES. 41 480 pounds, while a further increase from 4-J to 20 feet in length diminishes the collapsing pressure by only 90 pounds. As the thickness of wall is increased this disparity between the relative strength of long and short tubes becomes less prominent until for a thickness of 0.322 inch the difference is so small as to be of no practical importance. For example, assuming 20 feet to be the standard length of tube between end connections tending to hold it to a circular form, we find from curve N, for a thickness of 0.180 inch, that for lengths of 20, 15, 10, 5 and 2 feet, the respec- tive collapsing pressures would be 450, 470, 500, 540, and 980 pounds, which correspond to increasees of 4.5, 11, 20, and 118 per cent, respectively; while for a thickness of 0.322 inch the values of the collapsing pressures for the same lengths are 1,850, 1,915, 2,000, 2,140 and 2,390 pounds, which correspond to in- creases of 3.5, 8, 16, and 29 per cent, respectively. A study of these curves M and U, taken in connection with the photographs of the experimental tubes after being collapsed, and column 19 of the tabular statements of principal results of tests, will show conclusively that for lengths greater than about six diameters the strength of a tube when subjected to a fluid col- lapsing pressure is substantially constant. It must be remem- bered, while studying the photographs and column 19 of the tables referred to, that the inability to stop the hydraulic pressure pump instantly the experimental tube failed, together with the recoil of the hydraulic test cylinder, will ordinarily account for an ex- tension of the length of the collapsed portion by probably two or more diameters, after failure had actually occurred. Previously Published Formulae for the Collapsing Pres- sures of Tubes. Preparatory to entering upon the experimental investigation of which this is a report, an extensive search was made through the technical literature where one would expect to find matters re- lating to the collapsing pressures of tubes and flues. As a result of this search a number of formulae were collected and compared. Since completing the present research comparisons were made of the results of the actual tests and the corresponding values calculated by the different published formulae. These are shown in Figs. 22 to 33 inclusive. It will be observed that these com- parisons have been made for plain tubes, 8§ inches outside 42 COLLAPSING PRESSURES OF LAP-WELDED STEEL TUBES. diameter, in four commercial thicknesses, and for lengths of 2-J, 5, 10, 15, and 20 feet between end connnections tending to hold them to a circular form. In the table and charts above referred to and in the discussion that here follows : P = probable fluid collapsing pressure, in pounds per square inch, as based upon the present research, and for the conditions stated. p = collapsing pressure as calculated by the different published formulae for the same condition as for P. d = outside diameter of tube in inches. t = thickness of wall in inches. 1 = length of tube in inches. L = length of tube in feet. P/p = The relation of the actual collapsing pressure, for any stated conditions, to that calculated by the different pub- lished formulae for the same conditions. Fairbairris Formula. — From his own experiments Fairbairn established the following empirical formula : /2.19 /2.19 p = 9 , 676,000 -jj = 806,300 ^ • He states that “ the above is the general formula for the calcu- lation of the strength of wr ought-iron tubes subjected to external pressure, within the limits indicated by the experiments, that is, provided that the length is not less than 1.5 feet, and not greater than probably 10 feet.” It would appear that this upper limit was arbitrarily fixed, since none of the tubes tested by him ex- ceeded about five feet in length, and were held rigidly to a cir- cular form at the ends. Fig. 23 shows, plotted to scale, the values entered in columns 2 and 3 of Fig. 22. In this chart the vertical scale represents fluid collapsing pressure, in pounds per square inch, and the hori- zontal scale, length of tube, in feet, between end connections tend- ing to hold it to a circular form. The four curves shown in full lines are based upon the tests conducted, during the present research, on 8 f -inch outside diam- eter tubes in the four different thicknesses of wall shown and for the five lengths above stated. These lines are the same as curves H. (See page 40 and Fig. 21.) The broken and dotted lines represent the results obtained by COLLAPSING PEESSUEES OF LAP-WELDED STEEL TUBES. r>f/F 7 ) 000 * 04 * ‘ 2 MSS 3 JJ but)/ JO ft JOJ '■3pOJ± jopjoog bug hi n S3 60 V. of V) &■ n 3 •o ^ s ^ ^ * «•* i» si S3 a* 6* 01 ^ ^ ^ 5 ?3 J{? v» ss N to 0- 5 to S3 to K n. N N W Of Of m V» S »• > \3 0 te ^ v of K v 0 *i V v N «S rf) R. S Vd > ^ N) 0» V) W) N ^ V3 O ^ NO •O O' w> W> N > ^ n vj cs rv > rr) 0 ^ U 0% > c* O' S3 ^ 000 ‘ 0 U 7 S + gj0Q0'/.1fr£ - cl ‘o/nijujoj s t oj/ool/ag >A N, 0. >0 S' to < < Hj Sf \i CN< M K of ' W to to -N N H S' V, 1*1 < cf n> Vj K N N *>. V) VJ 5- N. S ^ N fl) N > »») \) N S) > U N nr) N v. N s o N \$ N 9* f* ^ ^ ^ <0 'c ^ < < < < CN Vs > « . V» < ■»■ N 5 V) to 1*) to to n V3 to Nl N to R. \9 m t* 0* <0 N <0 ff* >j to «0 «> Of Xi SO p ^ 2 ^ r: 60 V. 0 * fv \3 V • >• i*> *1 •> O) •Vj VS <0 V) O' to to > NS > 1*5 <0 -> n> i*$ R. !\ N N C\ N N N N N N {S js fe ft > & & > ^0 k) Ift V) V) ¥feZoorzsz=ct pu/of-jjng joj o/nu/joj $ obot/oft S* 60 v> ^ -to fv Sj n «* n >0 n 0^ ^ <\| CT) ^ ^ ^ -to no 1 r V* § ^ oj ^ W) * «*) ^ «o ^ IV N 4) N f n Q «0 S) W) «o n >o 0 W> Vo O. s> 0i c\4 OO ^ — O (Jn •n M «n o n ^ n > ^ * <0 >0 \> tl ^ ^ Of «s> »• vv <0 JS <0 > ; V> >o N N ® t> - v N 10 ■k* N O N. to to v, sj to in to V) vi vi vi R. > 0* -0 ^ ^ N > «<3 > * ^ s in n q fS ^ ^ ^ ^ ^ ^ ^ S3 ^ O D 5 3 ? *° i n ^ Jc > > to > »0 S3 to CO tj fS 0- fto ^ ^ ^ •© N fr> N <0 ‘o’O b»^ if 5h ") N N N ooo‘W/*d /got! ‘o/dujjoj S t JOl/SOjC) vA NO > V» « (») N W> Jo bo < < oi ei o# Of V> N Of m v» t» < to to to to R. s M b* ^ jo jn ^ to jo ^ •N » NS O' INS ^ « N MV fN * m m to »* NS to In f. 9> to l*J <0 V 9 V) V) > sssss NNJ toCNN, 099 ’m’i-ct ‘jujof-o/ 07 JOJ o/nuuoj S'U/MUf) A> fi ■* N ^ Nd >8 N t N N O N ^ < *i *$ *6 Sj O V) N N \| < < «>« »j ^ Si IN > to N. NO 4 > V, in to to vi < It si j} t ' ft. rr) ^ OJ > <> ^ 5 =aL ‘tutor- ung joj o/dujjoj $ t u/M up £ 50 »• O' O' O' *) i>0 ^ < o« rJ i») > > <0 -N »s < N nj >' Of sfl s Sd Os O t» > ill n 2 ^ to ^ in I*) no to > V> to ^ no nj R. NS t» N o> "1 ^ » v) N V> > n n > j* N \> n oo n > ^ N N ^ ^ *0 ^ ^0 "1 ^ 00 no *0 Of of s» -R> so ^ s? 00 ^0 ^ jjp . . -ffQOJS'Zb? =cL ‘o/dujjoj s f ujOJtsAtj >A V s ^ 1*) ^ «o ^ > >0 ^ N W N of N» to *V rt> NS fN O- NJ nr, ■n •* « nj s*i tto N > s> ^ <0 Of '•0^0--' *o < Of rs »> Of nr) Nj Vo O' NS V> * nT to to t*) 9) R. v» V) M i. > s m m f. s, in . o ^ n sa ^ > o» ^ fto * * *) •V V) to. to CN Cs S3 N ro 4 ) Q N S) V) 'to j£ooo'?/P&=a f ‘ofnuuoj xojo/o/p s UJ/OCfJ/OJ to* o» l* V* 6, k SJ «f ^ n< t» ns n? o~ *} r£ < ri H Cj N ^ >3 f 0 >o ^ Ti 0 - fs. < oi oi t»i N Of S3 to. -to -Q ff* ta < rf £t m tv» ; Q & «n *v ^ 'n. ^5 "O V> ci - "iV N» bo (► N ^ > > CO Of ^ » 1C to n, to to 1 *) »toto>o» to no NS ^ £) 6» 0. > sj <0 N tn S3 ^ 0* to O' O S3 5p ^ ^ * * 900W7'b =ot * 0 / nut jo j/ SUJitoqj/oj 0 e» \> ^ rs ^ m ri W> tm > N U 0 * * < ri if) ^ N Of tm n >) n 9 * o» t* > ^ ^ to! cs r») ^ of Of nn H* V» V) to nj so no to no V, ns to ^ to ei <0 ^ R. ^ > Q * * *V » 2 J s > g ^ t '"’ 2* ro *to S3 N t* 55C5SS M " * * n n « ' > toNStotoo- to v> in ' * * S/S3J uo posog V 3AJOQ UiOJJ SOn/t?A % s "V JJ *0 «< (X 0- Of Of 5> 5 5 ot ^ s» > "to ^ ^ K fN SO S3 t» *> t* *0 > rf) fO rN VI •0 ^ co <0 F II) $ < N V) njN^Ob) of of (S ^ ^ V a ay u t 90 r n J. J°14t>u37 * C* Vo ^ *0 04 v> <5 W% ^ ^ ^ N is fo ^ ^ N N Of es ^ ^ Vt Q ^ X Of 43 Beyond Foirboirn's i'mlt of Unyth. f Beyond Unwins limit of Un 9 th. Fig. 22. — Comparison op Values for the Collapsing Pressures op 8f" 0. D. Tubes, Calculated by the dif- ferent Published Formulae, with Corresponding Values Based upon Tests by Prof. R. T. Stewart. 44 COLLAPSING PRESSURES OF LAP-WELDED STEEL TUBES. Fig. 23. — FAIRBAIRN. Chart showing Comparison of Values for Col- lapsing Pressures of 8f" O.D. Tubes, obtained by use of Fairbairn’s Formula, with values based on Tests on National Tube Co.’s Besse- mer Steel Lap-welded Tubes. Broken lines show values obtained by use of Fairbaim’s Formula, full lines those based on Prof. Stewart’s experiments. Fairbairn states that his formula is applicable to lengths from 1$ to 10 feet. COLLAPSING PRESSURES OF LAP-WELDED STEEL TUBES. 45 plotting to the same scales the corresponding values for collapsing pressure as calculated by Fairbairn’s formula. For both sets of curves the thickness of wall in decimals of an inch is written at both ends of the curve representing the tube. It should be observed that the portions of the curves consisting of long dashes represent the application of Fairbairn’s fomula within the limits of length assigned by him, while the portions consisting of short dashes are beyond this limit. These latter have been drawn for the purpose of showing how utterly inapplicable is this formula to modern commercial tubes of comparatively long lengths between joints or end supports tending to hold them to a circular form, such as well casing, boiler tubes and long, plain boiler flues. Applicability of Fairbairns Formula . — It would appear from this chart that Fairbairn’s formula is applicable only to tubes having a comparatively thin wall and, at the same time, a rela- tively short length between joints tending to hold the tube to a circular form. That this should be the case is quite natural, since the experiments furnishing the data upon which this formula was based involved these conditions of relatively thin walls and short lengths. This is quite apparent from an inspection of the chart which, it will be observed, is for tubes having outside diameters of 8f inches in lengths of from 2-| to 20 feet, and for thicknesses of wall of 0.180, 0.229, 0.271 and 0.322 inch. The only place on the chart where values calculated by Fair- bairn’s formula agree substantially, over any appreciable length of tube, with those obtained by the present research on modern commercial tubes is for the tubes having a thickness of wall equal 0.180 inch, which were the thinnest tested, and then only for a length up to about 4.5 feet, or six diameters of tube. The chart shows that at Fairbairn’s limit of length of 10 feet a tube having an outside diameter of 8§ inches and a thickness of wall of 0.180 inch would probably collapse at 500 pounds per square inch. For these same conditions Fairbairn’s formula gives a pressure of 220 pounds, or 44 per cent, of this value, the actual collapsing pressure being thus 2.3 times that calculated by this formula. Again, for a plain tube of twice this length, or 20 feet between end connections tending to hold it to a circular form, and for the same diameter and thickness of wall, we get 450 pounds for the former and 110 pounds for the latter. For this case it appears that the value calculated by use of Fairbairn’s formula is only 46 COLLAPSING PRESSURES OF LAP-WELDED STEEL TUBES. 24 per cent, of that obtained by experiment, the latter for this case being 4.1 times the former. It is thus seen that when applied to an 8f-inch tube, 0.180 inch thick, and 20 feet long between joints, Fairbairn’s formula gives a result for the collapsing pres- sure that is a trifle over 300 per cent, in error. It will be observed that Fairbairn’s formula is even less ap- plicable to a tube having a thicker wall. For example, Fig. 23 shows that for the same diameter of tube as before, but having a thickness of wall equal 0.271 instead of 0.180 inch, the true values for probable collapsing pressures at Fairbairn’s limit and for lengths of 20 feet are, respectively, 2.7 and 5.0 times that obtained by use of his formula. In other words, this formula for these con- ditions gives results that are apparently in error by respectively 170 and 400 per cent. For 8f-inch commercial tubes having thicker walls than 0.180 inch it will be observed that Fairbairn’s formula gives the correct collapsing pressure for but one length, namely, that at which the broken line cuts the corresponding full line. The chart shows clearly that Fairbairn’s formula does not apply to modern lap- welded tubes having either relatively thick walls or long lengths be- tween joints or end connections tending to hold them to a circular form. Material of Fairbairn s Tubes . — Ho attempt has been made, in this report, to modify the constant of Fairbairn’s formula so as to adapt it to steel tubes. This was because of two reasons: First, no determination appears to have been made for the physical prop- erties of the wrought iron constituting Fairbairn’s experimental tubes, at least an examination of the records of his research has not disclosed any information on this point. Second, even had we a record of the physical properties of the material of his tubes, there appears to be no simple relation between the collapsing pres- sure of a tube and the physical properties of the material of which it is formed. For a discussion of this, see page 73. However, since Fairbairn’s experimental tubes were made from rolled plates, Ho. 19 B.W.G., or 0.042 inch thick, of presumably high-grade iron, the flat plates being rolled cold into tubular form, then riveted and brazed, it would appear that, for these conditions, the material of these tubes would not differ greatly in physical properties from those of the very soft steel used in modern com- mercial lap-welded tubes. Fairbairn' s Approximate Formula . — This formula was obtained COLLAPSING PRESSURES OF LAP-WELDED STEEL TUBES. 47 Fig. 24. — FAIRBAIRN. Chart showing comparison of values for col- lapsing PRESSURES OF 8f" O.D. TUBES, OBTAINED BY USE OF FAIRBAIRN’S Approximate Formula, with values based on Tests on National Tube Co.’s Bessemer Steel Lap-welded Tubes. Broken lines show values obtained by use of Fairb aim’s Approximate Formula full lines those based on Prof. Stewart’s experiments. Fairbaim states that his formula is applicable to lengths from 1J to 10 feet. 48 COLLAPSING PRESSURES OE LAP-WELDED STEEL TUBES. from the preceding more exact one by changing the factor t 2 ‘ 19 to t 2 , thus giving rise to a formula that could be more readily handled in the making of calculations. The formula thus modi- fied is p = 9,676,000 ~ = 806,300 ~ • The precise manner in which this change affects the results can best be had by making a comparison of Fig. 24 with Fig. 23. Also by comparing columns 5 and 6 with 3 and 4 of Fig. 22. Such a comparison will show that, while not representing the conditions of Fairbairn’s experiments, namely, relatively thin walls and short lengths, as well as the more exact formula, it is also quite inapplicable to modern wrought tubes in relatively long lengths. Grashofs Formula. — Grashof selected from Fairbairn’s ex- periments 17 of those having walls 0.042-inch thick and 4 having walls to J inch thick and from these 21 experiments he deduced the following formula : v 2.315 V 2.315 ^ = 24,480,000^ = 2 , 040, 000 . . . .(A) As this formula was found to represent the results of Fair- bairn’s experiments on the tubes having walls 0.042-inch thick bet- ter than those having walls to J inch thick, he derived the follow- ing formula for the latter, namely : P = 1,033,600 J ^ m = 86,130 . • . -(B) This formula has been applied to tubes having an outside di- ameter equal 8f inches, in lengths from 2|- to 20 feet and for four commercial thicknesses of wall from 0.180 to 0.322 inch. The precise manner in which these calculated values differ from the true probable collapsing pressures is clearly shown in Fig. 25 and in column 14 of the table, Fig. 22. Ny strom’s Formula. — Nystrom also used Fairbairn’s experi- ments for the deduction of his formula for the collapsing strength of flues. CO 1 I APSING PRESSURE OF LAP-WELDED STEEL TUBES. 49 Fig. 25. — GRASHOF. Chart showing comparison of values for col- lapsing PRESSURES OF 8f" O.D. TUBES, OBTAINED BY USE OF GrASHOF’s Formula B, with values based on Tests on National Tube Co.’s Bessemer Steel Lap-welded Tubes. Broken lines show values obtained by use of Grashof’s Formula, full lines those based on Prof. Stewart’s experiments. 50 COLLAPSING PRESSURES OF LAP-WELDED STEEL TUBES. Fig. 26. — NYSTROM. Chart showing comparison of value for collaps- ing PRESSURES OF 81" O.D. TUBES, OBTAINED BY USE OF NYSTROM’S Formula, with values based on Tests on National Tube Co.’s Bes- semer Steel Lap-welded Tubes. Broken lines show values obtained by use of Nystrom’s Formula, full lines those based on Prof. Stewart’s experiments. COLLAPSING PRESSURES OF LAP -WELDED STEEL TUBES. 51 Where T is the tensile strength of the material. 50,000 for T and for L in this formula gives p = 692,800 f dVl Substituting FTystrom considered 4 a sufficient factor of safety for use with his formula. The customary value of 50,000 for T has been used in this formula for two reasons: (1) In order that the results obtained might be comparable with the other heretofore published formulas, all of which are presumably for wrought iron; and (2) because, since the collapsing strength of a tube, in the light of the present research, appears to be quite independent of the tensile strength of the material constituting it, it was thought best not to attempt any modification of the formula. (See page 73.) A comparison of pressures obtained from Eystronfis fomula and the probable collapsing pressures of modern lap-welded tubes is shown in Fig. 26 and column 8 of Fig. 22. Unwin’s Formulae. — Unwin, who had been associated with Fair- bairn when he made his collapsing tests on tubes, has derived the following formulae for thick-walled tubes, namely, walls ^ to £ inch thick. For tubes with a longitudinal butt-joint : p = 9,614,000 t 221 Z 0 - 9 d 1 . 16 * . . . (A) For tubes with a longitudinal lap-joint: V — 7,363,000 £ 0 9 ^r ltl6 « (B) Unwin states that when the length of tube between end con- nections or transverse joints tending to hold it to a circular form is at least 10 or 12 diameters, the strength does not decrease with further increase of length. A comparison of values from these formulae with the probable collapsing pressures of modern tubes is given in Figs. 27 and 28, and columns 10 and 12 of Fig 22. Clark’s Formulae . — For the derivation of his formula A, Clark selected, from the reports of the Manchester Steam-Users Associa- tion, the dimensions of six boiler flues which collapsed while in 52 COLLAPSING PRESSURES OE LAP-WELDED STEEL TUBES. Fig. 27. — UNWIN. Chart showing comparison of Values for Collapsing Pressures of 8§" O.D. Tubes, obtained by use of Unwin’s Formula A, WITH VALUES BASED ON TESTS ON NATIONAL TUBE Co.’s BESSEMER Stee . Lap-welded Tubes. Broken lines show values obtained by use of Unwin’s Formula “for tubes with a longitudinal butt-joint,” full lines those based on Prof. Stewart’s experi- ments. Unwin states that when the length is at least 10 or 12 diameters the strength does not decrease with further increase of length. 2700 2600 2500 2400 2300 2200 2100 2000 1900 1800 1700 1600 1500 1400 1300 1200 1100 1000 900 800 700 600 500 400 300 200 100 0 It Fig. I COLLAPSING PRESSURES OF LAP-WELDED STEEL TUBES. 53 28. — UNWIN. Chart showing comparison of values for collapsing PRESSURES OF 81" O.D. TUBES, OBTAINED BY USE OF UNWIN’S FORMULA B, WITH VALUES BASED ON TESTS ON NATIONAL TUBE Co.’s BESSEMER Steel Lap-welded Tubes. broken lines show values obtained by use of Unwin’s Formula “for tubes a longitudinal lap-joint,” full lines those based on Prof. Stewart’s experi- Inwin states that when the length is at least 10 or 12 diameters the strength not decrease with further increase of length. 54 COLLAPSING PRESSURES OF LAP-WELDED STEEL TUBES. Fig. 29. — CLARK. Chart showing comparison of values for collapsing Pressures of 8f" O.D. Tubes, obtained by use of Clark’s Formulas, WITH VALUES BASED ON TESTS ON NATIONAL TUBE Co.’s BESSEMER Steel Lap-welded Tubes. Broken lines show values obtained by use of Clark’s Formulae, A & B, full lines those based on Prof. Stewart’s experiments. COLLAPSING PRESSURES OF LAP-WELDED STEEL TUBES. 55 Fig. 30. — LOVE. Chart showing comparison of values for collapsing fit Pressures of 8£" O.D. Tubes, obtained by use of Love’s Formula, with VALUES BASED ON TESTS ON NATIONAL TUBE Co.’s BESSEMER STEEL Lap-welded Tubes. Broken lines show values obtained by use of Love’s Formula, full lines those base d on Prof. Stewart’s experiments. 2400 2300 2200 2100 2000 1900 1300 1700 1600 1500 1400 1300 1200 1100 1000 900 800 700 600 500 400 300 200 100 0 COLLAPSING PRESSURES OF LAP-WELDED STEEL TUBES. >. 31. — BELPAIRE. Chart showing Comparison of Values for Col- lapsing Pressure of 8f" O. D. Tubes obtained by use of Belpaire’s Formula, with values based on tests on Natio a . Tube Co.’s Bessemer Steel Lap-welded Tubes. Broken lines show values obtained by use of Belpaire’s Formula, full lines se based on Prof. Stewart’s experiments. COLLAPSING PRESSURES OF LAP-WELDED STEEL TUBES. 57 Fig. 32. — WEHAGE. Chart Showing Comparison of Values for Collaps- ing Pressures of 8f" O.D. Tubes, obtained by Use of Wehage’s Formula, with Values Based on Tests on National Tube Co.’s Besse- mer Steel Lap-welded Tubes. Broken lines show values obtained by use of Wehage’s Formula for welded or butt-joints, from Dingler’s Journal, Vol. 242, 1881, page 236, and the 4th German ed. Reuleaux’s Const., page 1084. The full lines show values based on P of. Stewart’s experiments. 58 COLLAPSING PRESSURES OF LAP-WELDED STEEL TUBES. Fig. 33. — WEHAGE. Chart showing comparison of values for co laps- ing PRESSURES OF 8f" O.D. TUBES OBTAINED BY USE OF WeHAGE’S Formula, with values based on tests on National Tube Co.’s Besse- mer Steel Lap-welded Tubes. Broken lines show values obtained by use of Weliage’s Formula for welded or butt-joints, from Reuleaux’s Constructor translated by Suplee, 1893, page 269. The full lines show values based on Prof. Stewart’s experiments. COLLAPSING PRESSURES OF LAP-WELDED STEEL TUBES. 59 actual use and in boilers under known pressures, formula was , = f ( 52 g“ - 500) The resulting ... (A) For the collapsing pressure of plain riveted boiler flues, Clark gives the following formula : 200,000 t 2 P - d l.-,5 (B) For a comparison of values from these formulae with the prob- able collapsing pressures of lap-welded tubes, see Fig. 29 and column 20 of Fig. 22. Love's Formula . — M. Love’s formula, which was also deduced from Fairbairn’s experiments, is as follows : p = 5,358,000 ^ + 41,900 -j + 1,320 For a comparison of values obtained by its use with the collaps- ing pressures of modern tubes, see Fig. 30 and column 22 of Fig. 22. Belpaire's Formula . — From Fairbairn’s experiments Belpaire has deduced the following formula : p = 3,427,000 ^ + 56,890,000 For a comparison of values from this formula with the probable collapsing pressures of modern tubes, see Fig. 31 and column 24 of Fig. 22. Dr. Wehage deduced a formula for flues with butt-joints which he states is also applicable to lap-welded tubes. This for- mula was apparently based upon Fairbairn’s three experiments on tubes thicker than -J inch together with three isolated tests on boiler flues and two failures of flues while in service. In metric units, as published in Dingler’s Journal, vol. 242, 1881, page 236, and in the 4th German ed. of Beuleaux’s Con- structor, this formula is — S a L = 120,000 -g where the collapsing pressure, a 1 , is expressed in kilograms per sq. 60 COLLAPSING PRESSURES OF LAP-WELDED STEEL TUBES. centimeter ; while the diameter, D , the thickness, S, and length l, are in millimeters. Reduced to the same British units in which the other formulae are expressed, this formula becomes p = 253,600 2\J x2' A comparison of values obtained by use of this formula with the results of the present research is shown in Fig. 32 and in column 18 of Fig. 22. Suplee’s translation of Reuleaux’s Constructor, 1893, page 269, states this formula in British units to be ^ = 490,000 For a comparison of results by this formula with the ’-esults of the present research see column 16 of Fig. 22, and also Fig. 33. English Board of Trade s Formula, —The following formula from the English Board of Trade is rbtained from Fairbairn’s ap- proximate formula by using a factor of safety of about 9, and substituting L -f- 1 for L, namely. _ 90,000 t 2 P ~ (Z + l)y 1 ^Variation. of Tube, as Reported. from Wgt Lbs. Ft Formula. from Gated. 2U 2.02V 0. 152 0./55 7.75 760 V73 - 9.V Cosing, tO.*tl tbs. 222 2.037 0./52 0./56 7.79 YVO VSV t 3./ .. u •• 2/3 2.037 0./52 0./67 10.30 SS6 76V ~/Z.Z '• A* •« <• 227 2.022 0 .156 0./66 / 0.70 mo /002 + /0.V M St <* '«#. 22S 2.03V 0./S6 0. 166 16.37 7/5 797 - 2 V. 3 II II M M 22/ 2.637 0./56 0. 166 10.70 760 97V -23.V M «• H •< 2/2 2. 0/ / 0./S6 0. 166 / 0.35 750 ZOOS' - 5.V 1* II •• it 2/7 2.02/ 0./S6 0./66 70.37 7V0 / 003 ~ Z.3 M M II tS 2/0 2.000 2.007 0 .152 0./67 10.77 / 025 /02V + 0./ • • M II M 23V 2.037 0 .152 0.167 / 0.75 775 /on -23.3 «• M II M 27/ 2.037 0/52 0./6V / 0.50 /OSO /027 / 2.2 II II II .« 2/t 2.00V 0.152 0./6V /O.SO V60 / 037 -/ 7. / • I II U II 2/S 2.033 0./S2 O.UV / O.JS to/o /02V • 1.7 .1 1 M II 2// 2.027 0./S6 6. 167 / 6.57 730 7076 -//./ *« (l |i M 223 2.022 0./5C 0./67 10.57 /070 7075 f 2.7 • 1 M II U 2*3 2.037 0.220 0/70 10.62 V75 . i 056 -77./ •« •• IH.Z6 » 2/7 2.03/ 0. 152 0.171 10.67 / 096 107/ + t.r •• '* /O.HG " 2/7 2.03V 0./S2 0./7S 10.72 1010 / 126 -/0.3 Average 2.02 V 0./67 10.72 92V /OOV - 7.9 220 2.02V 0 . 152 0./7V U./7 /72S / 173 72/. S sf Casing, 10.76 lbs. 2*0 2.055 0.220 0./79 n.22 /o/o 1176 -77./ •• •• 17.26 *' Z70 2.035 0.220 0./77 //./? / 095 //VS -7.6 257 2.027 0.203 0JV2 / / .35 / too 7233 -/ O.V •• /2 X 07 •• zsr 2.0/3 0.203 0./V2 / 1.36 /3S0 1237 + 9/ •• •• «• a 253 2.020 0.203 0./V7 / /.75 9 56 / 263 -27. V I. •• •* 272 2.000 2.035 0:220 O./VS n.s7 1272 127/ to./ «• •• IH.19 M 256 2.02/ 0.203 0./V5 //.ss / 750 1277 t/3.5 •• tZ.C't M 25 7 2.025 0.203 0./V7 11.77 1256 7333 - £.2 .1 .1 255 2.02 V 0.20 3 0./72 /1.7V 790 7 37S -72.5 II 1. 1. 250 2.0/2 0.203 0. 172 t/,76 / 750 /3V6 t S.7 II* •# II (1 275 2.0/0 0.220 0./73 / / .77 7375 /377 - /.6 .. .. W.IQ •• 277 5.77V 0.220 0./73 / / .77 1750 /703 717.7 .. Average 2.027 0. 176 H.57 iZS! /ZVS - 2.6 25/ 2.0/2 0.203 0.2/7 13.25 /7S0 / 677 t 3./ Sf" Casing, /z.67 tbs. 27/ 2.021 0.220 0.2/7 13.57 / 606 1767 - 9.7 •• « /7.20 •• 257 2.005 0.203 0.220 13.57 /5S0 nv9 -73.7 .. .. /2.07 •• 252 2.000 2.020 0.203 0.222 13.73 2675 /V/0 f/7.6 •• *277 2.032 0.220 0.22V 17.15 /ZOO /VVV - 36.7 •• •• 17.20 » 272 2.00/ 0.220 0.230 /7./7 /700 1736 - 7.9 27V 2.067 0.220 0.232 17.30 /VO 6 /963 - t.3 Average £.0// 0.222 13.77 1779 /VZ7 - 2 5 * Defective, not in averages. 7/5 2.072 0.27/ 0.250 15.53 2220 2/V3 t /.7 Jf" Cosing , 16.70 ibs. 2 2/ 2.027 0.27/ 0.25/ 15.77 / 750 2ZZ3 -Z/.3 177 2.072 0.27/ 0.257 15.77 2360 2Z70 t S.7 272 2.033 0.27/ 0.257 15.77 2775 2335 t 6. 0 * 220 2.057 0.27/ 0.260 16.06 1755 2336 -27.9 .. 227 2.022 0.27/ 0.260 15.77 2750 2356 + 7.0 7/2 2.072 0 27/ 0.262 16.16 2555 237Z t 7.7 .. V f 22Z 6.033 0.27/ 0.26 3 16.20 2600 t 2392 t V.7 2X2 2.027 0.27/ 0.267 / 6.77 2250 2756 - V.7 772 2.000 5.770 0.27 0.267 16.75 2590 Z506 t 3.7 6 ' O.D. Special , tl./X the. 77/ 5.773 0.2 V 0.267 16.75 2/50 2507 -17.1 27 3 2.02/ 0.27/ 0.270 16.59 2270 250/ - <7.2 S-a Casing , 16.10 ibs. its £.022 0.27/ 0.Z76 16.55 2550 2500 + 20 773 5.77X 0.2V 0.27/ / 6.55 2760 2530 - 2.V 6" O.D. Special , n./Z /bs. 7/7 £.072 0.27/ 0.27/ 16.71 2700 250/ - 7.0 5§‘ Casing, 16.16 / bs . 7/7 2.077 0.27/ 0.27/ 16.73 2575 2500 / 3.0 770 5.77/ 0.27 0.273 / 6.6V 27X0 2563 + V.5 6" 0.0. Special , /T.tZ/bs. 777 5.773 O.ZV 0.277 16.73 2755 2576 - 7.7 .. .. ;. .. .. ^ 7/X 2.077 0.2 7/ 0.277 /7.6V 2V70 ISVi + //.V ■5§" Casi ng ./ 6.1 0 /bs. 223 2.025 0.27/ 0.2V6 17.17 2600 267 Z - 1.6 ” Average £022 0.266 /6 3V 247/ 2776 - O.Z Defective, not tn averages. ? Did not co/taps e. / 2 3 4 S' 6 7 9 7 IO Fig. 44. 64 COLLAPSING PRESSURES OF LAP-WELDED STEEL TUBES, COLLAPSING PRESSURES . — Abstract from Log of Tests Conducted by Prof R.T. Stewart, I30Z-64, on National Tube Co.’s Lap-welded Bessemer Steel Tubes, 20 Foot Lengths, to which is added a Com parison with Calculated Values, loy Formulae A&B. Made by e.e.s. under direction of r.t. s ., nos. Tests Grouped according to Outside Diameter and Arranged in Order of Thicknesses of Tubes. Number of Test. Outside, Diameter fn cbes. Thickness of VJaii, In c htt j. Actual Plain End Weight, Lbs.p*rFt. Collapsing Pressure Pounds per Square, inch. Commercial Designation Of Tube, as Reported. Nomina/ Actual Nomina/ Computed from Wgt. Observed Calc'd by Formula. % Variation from Cok'd 0.005 O./S 3/ 52 13.57 5*5 59/ - /.6 6 §" O.D. Special, 13.39 lbs. hoh 0057 a./s 0/53 / 0.S9 520 606 -19.2 HO 3 i.ilS 0.00/ 3. IS 0./5H / 3.07 563 67X - 9.H *900 £.003 3./ 5 3. / 5H /6.0X H33 6/7 -35.1 3 16 £.£S7 0./72 3. 157 / 0. *7 7/0 CSX - 7.9 6+" Casing, II. 5* tbs. HOI £.£Str 3. !5 6. IS7 / 0. *7 5*5 OSX -//./ Of O.D. Special, 10.39 lbs. A veroge £.000 3./5S / 0.7 i 59 2 026 - X.7 * Defective , not in averages. See tog. 303 £.053 3/ OH II .35 733 753 - 2.7 3 OH £.025 0.052 3.172 a. /os U .HZ 030 70H -2/.S 6 f Casing, n.sx/bs. 30/ 0.057 0./05 II.H7 7 23 76Z - 5.5 301 0.053 O./CX / 1.02 030 X 33 -2/. 5 Average O.OSH 3.760 /■/ .H7 07 0 770 -/2.X 3 OS 0.0X0 0.176 /3.S7 7/00 7 157 - H.9 30? £.070 0./90 i 3.57 1375 US 9 -9.2 300 0.025 0.093 3.233 3.233 / 33/5 /Z05 /Z09 HO./ of Casing, 13.32 tbs. 307 0.0*2 0.232 / 3.7 9 1275 /Z3H F 3.3 30 * 0.0*7 3.235 /H./7 /ZOS 1271 - 3.5 Average 0.0*9 0.200 73X3 77*9 HOS -/.* HI 0 o.cvo O.ZH/ /0.S9 7 6*3 /73X - 3.3 HI3 0.0 79 3.ZHX 10.79 /7/0 7X32 - 6.7 H/H 0.0*2 3.2HX 17.0/ /6/S 7X37 -//.S' * 3/3 0.057 6.250 /7.D? 1*20 7X69 - 2.6 3IH £.025 0.07 / 3.23 r 3.25/ 17.20 2/75 1X75 -t/6.0 6f 'casing , 17.02 tbs. HU 0.0* H 3.25/ /7.2H !9 HO 7X69 F 3* 3/2 0.009 0.252 t 7.25 1975 7XX9 F 9.0 3/1 0.07H 0.2SH 77.3? 2/60 79/3 7/2.9 HIZ 0.075 3. 2 SO I7.SI 17*6 193* - X. 2 Average 0.077 3.253 17.15 /V79 7X0l t 0.9 *lron. not in averages. 310 0.00/ 3.23 X 3.203 /7.7V 2275 7997 7/3.9 6+ Casing, 17.02 tbs. HO* 0.0 5/ 0.2X3 3.20/ I7.XI 2303 23/5 F 2.2 6 " Full Weight, /V. 76 lbs. HOS 0.025 o.ose 3.2*3 0.202 /7.X 5 2/35 232 X F 5.3 .. H30 0.053 3.2*3 0.202 /7.X 5 / 9 75 2327 - 2.0 »• H 07 0.055 0.2*0 3.27X 72.99 2560 2239 F/H.6 .. .# HO 9 O.OSH 0.2*0 3.2XH 19.35 2393 2313 F / .2 Average 0.052 3.2£* /V.26 222V 2/02 7- 5.* r H39 7.539 0./9X 3. 153 / / .2/ S/S 5/3 F 3.9 7" Converse Joint, /e.os lbs. 3/7 7.393 O./XO 3./5H 11.31 5 75 5/5 F//.0 Off Casing, / 2.3H lbs . H37 7.369 O./HX O./SS / / .3/ 6 35 533 F/3.5 7" Converse Joint, /e. OS lbs. 3/0 7. OHO O./XO o./si /I.SX 550 559 - /.0 6§~ Casing, 12.39 tbs. 3/S 7.000 7.0HC O./XO 6./5X 11.01 570 55* F 2.2 H3S 7.00 * O./HX 0/0/ n .75 £65 6 0S F 9.9 7 Converse Joint, to. 05 tbs. H3X 7.6/5 O./HX 6./02 U.X5 £75 0/5 F 9.X .. .. ,. .. .. 3/9 7.05 9 O./XO 3. 103 H.70 593 675 -I2.2 Of'Casing , 12.39 tbs. 3/r 7.333 6./X0 O./OO U./3 5*3 003 -12.1 H3i 7.3/5 0. !HX 0./6X / 2.25 CHS 690 - 0.5 7 Converse Joint , 10.65 tbs. Average 7.32* 0./60 / / .70 592. 3X6 F /. 5 322 7.3 57 0.233 70.97 1525 7976 F 3.3 32/ 7.050 6.ZHI n.so 1575 7577 - 3./ 320 7.000 7.0 S3 0.2HX 6.ZHS 17.70 1775 1626 F 9.2 0-f Casing, 17.51 tbs. 323 7.3H5 0.2 H 5 n.77 U75 162 * F 2.9 32H 7.3H7 0.2HX /X.02 7X53 7669 F//.2 Average 7.353 3.292 n.oo 76X0 7599 F S3 H3Z 7.3/* 0.20* 79.33 7*35 7929 - H.O H3H 0.9*H 3.209 19. 3H 1975 7 952 7 / .2 H3/ 7.000 0.975 3.2X 3.2X3 23.23 2300 2/30 7 X.O 7 O.D. Special . 2 . 0.12 lbs. H33 0.97H 3.1X7 Z3.H7 Z/X6 21*1 - 0.0 H30 0.9X9 3.290 23.7H 2995 22/3 7/3.5 A veroge 6.9*7 0. 279 10.02 2JH7 20*3 F 3.3 ) 2 3 A 5 £ 7 8 9 70 Fig. 45, COLLAPSING PRESSURES OF LAP-WELDED STEEL TUBES. 65 COLLAPSING PRESSURES . — Abstract from Log of Tests Conducted by Prof R. T. Stewart, /902-04, on National Tube Co.'s Lap-welded Bessemer Steel Tubes, 20 Foot Len qths, to which is a d d ed a Com parison with Calculated Values, by Formulae A&B. Made by e.l. 5. under direction of r. r. s., /90s. Tests Grouped according to Outside Diameter and Arranged in Order of Thickness es of Tubes. Humber of Test Outside Diameter; /netted. Thickness of Via//, inch as. Actual P/ainEnd CoJ/ap Po u n c/s sing Pressure., per- Square. fnc,h. Commercial D KSignation of Tube, as Reported. Nomina/. Actual. Nominal Computed from Wqt. Weight, L6s.pT ft Observed Ca/c'd by Formula. %1/ar/otion from Calc'cL / f.657 6. / 7( /S.92 956 9 / y f 7.7 9 7.696 6/73 / (.51 950 9(1 - 9.6 3 ? .625 7.69! O. 170 0./7( / 6.77 535 91/ 9 9.0 ?F Casing, 7 (.07 lbs. 2 7.637 6. m n.zi 625 539 9/7.6 S 7.637 6./1/ 7 7.23 (26 533 9/0.3 Averaqe 7.613 6/75 76.79 536 971 9 1.2 26 ? .625 7.661 0.221 6.2/1 11.57 776 126 9 6.1 i 'ff Casing, 26. /0 lbs. £2 7.661 0.27/ 6.257 23.13 /320 7/95 9/6.5 l£" Casing, 29.31 7bs. 51 7.666 6.27/ 6.2(2 23.52 7975 72'3( 920.7 95 7.663 6.322 6.2(9 13.6( 7375 7255 9 1C 1 L i ne Pipe., 27. H lbs. Si 7.660 6.27/ 6.27/ 29.27 7935 7326 9 1.2 If Casing, 29.3? lbs. S3 ? .625 7.666 6.27/ 6.Z72 29.32 7526 7336 9/3.7 70 7.(63 6.27/ 6.272 29.31 IHIO 7335 9 3.6 1" Line Pipe , 25.06 tbs. 75 7.610 0.27/ 0.279 29.99 7375 73(3 9 0.9 5/ 7.667 6.7.7/ 6.279 29.59 /*t30 7359 9 S.C ?jf Casing, 29.31 ibs. *77 7.666 4. 2*7 0.279 29.97 1275 7356 -6.0 1" Line Pipe, 25.66 lbs. *77 7.666 6.27/ 0.270 25.61 1256 79/ C -II. 7 Average 7.666 6.2(7 29.03 /-y/9 7366 9 1 . 3 *lror>, not in averages, 7 0 / 7.697 6.322 0.219 26.2 7 7(56 756/ 9 5.7 ?" Line Pipe, 21.71 7bs. 106 7.696 6.322 6.217 26.11 77/0 1591 9 7.5 111 7.671 6.322 6.362 2C.11 7575 7(33 - 3.5 1" Full Weight, 27.71 7bS. IOZ 7.656 6.322 6.363 27.66 71(6 7(97 971.1 7" Line pipe, Zl./t ibs. 923 7.625 7.691 6.322 6.363 2 7.66 7736 7(50 970.1 1" Full Weight, 21. /? Ibs. 92! 7.666 6.322 6.363 27.61 7 935 7(99 977. 7 « 922 7.672 6.322 6.367 27.31 7(35 7(12 - 2.1 tn 7.666 6.322 6.361 27.99 7735 7(99 9 2.9 ST" Line Pipe. 21. H tbS. 926 7.(57 6.322 6.3/6 27.(9 nos 777 7 9 S.7 SC" Full Weight. 21. nibs. Average. 7.657 6.362 ZC.17 77(2 7(1/ 9 7.9 T 0. H. Steel, not in overages. *927 S .677 6.3K 36.12 H36 2670 -77.5 * 925 7.666 6.391 3/. 66 2/16 2/69 9 3.6 *92? s.us 7.669 6.3(3 6.353 3/.3C 2695 2795 - 9.7 ST" Oil Well Tubing, 32.66 ibs. *921 7.673 0.3S6 31.(2 1136 2 7(7 -t6.1 *126 7.672 6.3(9 32.36 2/55 2252 - 9.3 Average 7.673 6.359 3/.9Z 2627 2/ 91 - S.( */ron , group rejected. 199 16.655 6.157 / (.55 2/0 2/1 - 9.1 997 10.695 6./6C 17.56 296 256 -9.0 115 /0.660 16.637 6. /St, 6.UJ 17.57 2/6 259 -77.3 /O" Converse Joint, /(./? /bs. 996 16.631 6/67 n.si 225 259 -II.9 99? 16.635 6. no 17.91 296 2(5 - 1.9 Average / 6.69! 6/(5 77.93 225 291 - 1.2 952 76.665 6. 175 / 1.93 365 327 - 5.7 953 76.633 6./16 19.11 315 397 9/3.1 95 1 16.666 76.629 6.263 6. Ill 26.35 396 3(7 9 (3 Id" Boi/er Tu bing, Z/.66 Ibs . 951 16.637 6/ 15 20.59 966 37/ 9 7.V 950 76.627 6.26 ( 2/. 57 925 936 - / .2 Average 76.626 6./19 26.37 373 3(1 9 9.6 956 9.196 0.3/2 32.27 7356 1321 9 2 2 9S9 70.623 6.3/9 32.(1 7315 1329 9 9.2 957 16.666 1.179 6.30 6.3/7 32. 7/ 7275 /3C9 - 6.5 /O' O.D. Special, 3/ 67 lbs. 95 1 7 6.663 6.3/7 32 11 1365 13(6 - 9.6 955 76.660 6.3/1 33.6/ /Z16 1379 - 7.2 Averager 16.601 6.3/6 32.(1 73/9 735/ - 2.3 / 2 3 4 5 7 1 7 / o Formula P , P- 1000 ( 1-Vi-itcoD J . Formula 3 , P = 36670 j - 13 S 6. P% Collapsing fluid pressure, in lbs. per sq. in., t • thickness of trail In inches, d ~ Outside dia. of tube in injches. Formula d applies only to values of £ greater than 0.227 , Or pressures greater than 581 pounds / trhUe for — ipiula P applies only to Values less then these- Fig. 46. 66 COLLAPSING PRESSURES OF LAP-WELDED STEEL TUBES. After a fruitless effort to derive a satisfactory formula on the basis of three variables, which, when plotted, would, of course, be a surface in space, the thought happily presented itself that two of these variables, t and d, could be replaced by their quotient, or ~ which, of course, might be treated as a single variable. By the adoption of this expedient, matters were greatly simplified, since it thus became possible to plot the results of the tests for all di- ameters and thicknesses of wall on a plane surface. Big. 47 shows the group averages of Figs. 43-46 plotted in this manner, that is to say, to a vertical scale representing fluid- collapsing pressures in pounds per square inch, and a horizontal scale representing the quotient arising from dividing the thickness of wall by the outside diameter, both being expressed in inches. Formula B . — By an inspection of Fig. 47 it became apparent that the bulk of the group averages could be represented by a straight-line formula, indeed all of them could be thus represented with the exception of the few having values of thickness divided by outside diameter less than 0.023. In other words, about 93 per cent, of the group averages of Figs. 43-46 can be thus repre- sented. On this basis then, for values of greater than 0.023, formula B was deduced, it being as follows : P = 86,670 - 1386 (B) Where P = collapsing pressure, lbs. per sq. inch, d = outside diameter of tube in inches, t = thickness of wall in inches. Remembering that this same formula might also have been ar- rived at by the substitution of proper empirical constants in a similar formula for a theoretically perfect tube, and further, since Fig. 47 shows no apparent deviation from straightness on the upward course, it was not thought necessary to set an upper limit to the value of in the application of this formula, believing that it will give substantially correct results for all commercial lap- welded Bessemer-steel tubes whose thickness divided by the out- side diameter is greater than 0.023. Formula A. — This formula for values of -4- less than 0.023 d COLLAPSING PRESSURES OF LAP-WELDED STEEL TUBES. 67 R.T. Stewart "Pulliams Enz. Co., N.T. Fig. 47. — Chart showing Actual and Calculated Collapsing Pressures of National Tube Co.'s Bessemer Steel Lap- welded Tubes Plotted to Thickness -t- Outside Diameter, or % / d . Based on Tests by Prof Stewart on 20-foot lengths. Note that group averages are represented by crosses ( + ), the attached figures indicating outside diameter, while values calculated by means of formulae A and B are represented by circles (O). 68 COLLAPSING PRESSURES OF LAP-WELDED STEEL TUBES. was derived upon the assumption that when plotted upon Fig. 47 the resulting curve should be tangent to the straight line represent- ing formula B, and be also tangent to the horizontal axis at the origin 0. This arbitrary assumption gave a formula which repre- sented very satisfactorily the few experiments in which ~ was CL less than 0.023. The formula thus obtained is P = 1000^1 - ^/i _ 1600^ • • • • (A) Where P, d and t are the same as for formula B. This formula should be used for values of and for P greater than-^j / ^ less than 0.023 Since constructing the charts and tables contained in this paper, \t was discovered that a formula having a rational form with em- uirical constants could be substituted for the purely empirical formula A. This formula, in addition to involving theoretical considerations of elasticity, is much the simpler of the two. It is applicable only to tubes having relatively thin walls, that is to say, to those having values of less than 0.023, and is P = 50,2l0,00()Qy (G) Where P } d and t are the same as for formula A. Either formula A or G represents satisfactorily the results of the experiments made on thin-walled tubes, that is, those in which A is less than 0.023, but probably formula G will permit of the greater exterpolation. The following formulae are meant for application in case the outside diameter and plain-end weight are given. They were derived irom formulae A and B and are p = 1000 ^1 - j/ 149.8 ^ - 799 4- 800 |/l - 0.375 ^.(O) 41,950 - 26,520 j/ 2.67 (D) Where w = the plain-end weight of tube in pounds per foot, while P and d are the same as for formulae A and B. COLLAPSING PRESSURES OF LAP-WELDED STEEL TUBES. 69 Formula C is lor values of less than 0.237 and P less a z than 581 lbs., while formula D is for values greater than these. Charts of Actual and Calculated Collapsing Pressures . — Figs. 48-50 show a comparison of the results obtained by actual test with the corresponding calculated values, plotted to a vertical scale representing collapsing pressures in pounds per square inch, and a horizontal scale representing the thickness of wall in decimals of an inch. It will he noted that Fig. 48 is for the ex- perimental tubes having outside diameters of 3, 6 and 10 inches, the diameter being written in each case on the margin at the right- hand end of the line representing the tube. Similarly, Figs. 49 and 50 were constructed for the tubes having outside diameters of respectively 4 and 7 inches, and 6f and 8f inches. The lines on these Charts were plotted from values calculated by means of formulae A and B, representing the most probable values for the collapsing pressures of lap-welded Bessemer-steel tubes in lengths of 20 feet between transverse joints tending to hold the tube to a circular form. The center of each small circle lying on these lines represents a plotted calculated value. The actual collapsing pressures of the different tubes tested, plotted to the same scales as the calculated values, are represented by crosses (+J for those having outside diameters of 3, 10, 4, 7 and 8f inch, and by tees (T) for the 6 and the 6| inch. In a number of instances, in order to avoid confusion, the characters be- ing very close together, it became necessary to omit a part of the cross (+), in which case it became a tee (T), and likewise a part of the tee, it thus appearing as an angle or ell (L). Group averages are represented on these charts by means of combined crosses and circles (-o-). It will be observed that the group averages of the actual collaps- ing pressures lie very close to the corresponding values calculated by means of formulae A and B. For the actual variation in per cent., see Figs. 43-46, column 9. which gives the variation for the individual tests as well as for the group averages. Formula? A and B being based upon the results of all the ex- periments on the 20-foot lengths of the lap-welded Bessemer-steel tubes tested, excepting the three that proved to be defective, it is clear that the curves plotted on these charts represent average values for the extreme range in thickness of wall for each of the seven diameters tested. 70 6000 5800 5600 5400 5200 5000 4800 4600 4400 4200 4000 8800 3600 3400 3200 3000 2800 2600 2400 2200 2000 1800 1600 1400 1200 1000 800 600 400 200 Fig. COLLAPSING PRESSURES OF LAP-WELDED STEEL TUBES. .02 .04 .06 .08 .10 .12 .14 .16 .18 .20 .22 .24 .26 .28 .30 .32 .34 .36 .38 .40 r. Stewart. Williams Eng. Co. N.Y. 48. — Chart showing Actual and Calculated Collapsing Pressures )f National Tube Co.’s Bessemer-Steel Lap-welded Tubes Plotted ro Thickness of Wall. For Outside Diameters of 3, 6 and 10 NCHES, IN 20-FOOT LENGTHS. BASED ON TESTS BY PROF. STEWART, L902-4. ote that individual experiments are represented by crosses (+), calculated :s by circles (o), and group averages by combined circles and crosses (-6-) 3800 3600 3400 3200 3000 2800 2600 24C0 2203 2000 1800 1600 1400 1200 1000 800 600 400 200 ( £ Fig, COLLAPSING PRESSURES OF TAP-WELDED STEEL TUBES. 71 4 " 49. — Chart showing Actual and Calculated Collapsing Pressures of National Tube Co.’s Bessemer-Steel Lap-welded Tubes Plotted to Thickness of Wall. For Outside Diameters of 4 and -7 inches, in 20-foot lengths. Based on Tests by Prof. Stewart, 1902-4. sTote that individual experiments are represented by crosses ( + ) calculated es by circles (o) and group averages by combined circles and crosses(-o-). 4000 3800 3600 3400 3200 3000 2800 2600 2400 2200 2000 1800 1600 1400 1200 ,1000 800 600 400 200 C li '’IG. COLLAPSING PRESSURES OF LAP-WELDED STEEL TUBES, .02 .04 06 .08 .10 .12 .14 .16 .18 .20 .22 .24 .26 .28 .30 .32 .34 36 .38 ,40 T. Stewart Williams Eng. Co., N.U' 50 . — Chart showing Actual and Calculated Collapsing Pressures r National Tube Co.’s Bessemer-Steel Lap-welded Tubes Plotted ) Thickness of Wall. For Outside Diameters of 6f and 8f inches, in l-FOOT LENGTHS. BASED ON TESTS BY PROF. STEWART, 1902 - 4 . te that individual experiments are represented by crosses ( + ), calculated by circles (o), and group averages by combined circles and crosses (-<>)• COLLAPSING PRESSURES OF LAP-WELDED STEEL TUBES. 73 The scattering of individual results as compared with the gen- eral average appears, from these charts, to be restricted to com- paratively small bounds when it is considered that we are dealing here with a product that varies noticeably in a number of the characteristics that go to make up its strength. Since these charts represent the results of tests on the common run of commercial lap-welded Bessemer-steel tubes, taken at random from the stock, it is surprising that the scattering of individual results is not greater than that shown. Apparent Fiber Stress on Wall of Tube at Instant of Collapse. Fig. 51 shows the apparent compressive stress, in pounds per square inch, at the instant of collapse, on the walls of the tubes constituting Series Two. This chart is constructed to a horizontal scale representing thickness of wall divided by outside diameter of tube and a vertical scale representing apparent fiber stress in pounds per square inch. The crosses (+) represent the apparent fiber stress of the group averages of Figs. 43-46, the attached figures indicating the outside diameter of tube, while the curve represents the formulse E and F, plotted to the same scales. These formulae, which were deduced to represent the most probable values of the apparent fiber stress in the walls of the tubes constituting Series Two, at instant of collapse, are as follows : For values of 4 less than 0.023 : a £=500 4(l _ |/i- 1,600^ . . .(E) And for values of ^ greater than 0.023 : £= 43,335 - 693 f (F) Where 8 = apparent fiber stress in lbs. per sq. inch, d — outside diameter of tube in inches, t = thickness of wall in inches. An inspection of this chart will show that the apparent fiber stress on the wall of the tube at instant of collapse varied all the way from about 7,000 pounds per square inch for the relatively 74 COLLAPSING PRESSURES OF LAP-WELDED STEEL TUBES. Fig. 51. — Chart showing Actual and Calculated Apparent Fiber Stress on Wa l op Tube at Instant of Collapse, Plotted to Thickness Diameter, */ d for National Tube Co.'s Bessemer-Steel Lap- welded Tubes. Based on tests on 20-foot lengths by Prof. Stewart, 1902-4. Note that crosses ( + ) represent group averages of tests, the attached figures indicating outside diameters, while circles ( O ) represent calculated values. COLLAPSING PRESSURES OF LAP-WELDED STEEL TUBES. 75 thinnest to 35,000 pounds per square inch for the relatively thick- est walls. This chart shows conclusively that the ability of a commercial wrought tube to withstand a fluid-collapsing pressure is not de- pendent alone upon either the ultimate strength or elastic limit of the material constituting it. A study of this chart has led to some very interesting deductions which will be dealt with in a separate paper. Relation of Point of Collapse to Length of Tube. Theoretically a tube should begin to callapse at the middle of its length, that is, at a point half way between transverse joints, or end connections, tending to hold it to a circular form. This statement is, of course, based upon the assumption that the material of the tube is perfectly homogeneous in its physical prop- erties and that the diameter and thickness of wall are strictly con- stant throughout its entire length. The truth of the above statement becomes apparent when we consider that the strength of a tube to resist collapsing pressure depends upon, first, the transverse rigidity of its wall and, second, the tendency of the end connections to hold the tuDe to a circular form. Since the former, for the assumptions made, would be con- stant from end to end of the tube and since the latter tendency would become less as the distance from an end connection increases, it is evident that a theoretically perfect tube subjected to a fluid- collapsing pressure would be weakest at a point that is at the great- est possible distance from both of its ends, which point is, of course, located at the middle of its length. In commercial tubes, however, the material is not strictly homogeneous in its physical properties and there is also a slight variation in out-of-roundness of the different cross-sections, from end to end, as well as a perceptible variation in thickness of wall. Because of these a commercial tube is not necessarily weakest against collapsing pressure at the middle point of its length, as is the case for the theoretically perfect tube. The actual relation of the point of collapse to the length of tube, for the several hundred commercial tubes tested, is shown in Fig. 52. This chart represents a 20-foot tube divided into foot lengths and numbered consecutively, beginning at the left-hand end. Over each division is placed the Log number of the experi- mental tubes that collapsed at points nearest to that division. 76 COLLAPSING PRESSURES OF LAP-WELDED STEEL TUBES. 1 * 5 2 1 I II I I I I I I I I I I I I I I I I I I I I 1 I U-LlI I I 1 llm-lJ SURS 5555S- IIsSSkU IIIIIBSSS:* 5?3s- IIIIIIIIIISI3ili35 IIIMIIIIIMH??;?;^ IIIIIIIIIIIIIIIIM^^^a 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 - 1 1 ms- rn 1 1 n q 1 1 1 rj r s * 1 1 1 1 1 1 ii i| 1 1 1 1 1 rmyn 3 2 I ! ! * ! I * il Fig. 52. COLLAPSING PEESSUEES OF LAP-WELDED STEEL TUBES. 77 Thus, experimental tubes Nos. 100, 433 and 505 collapsed nearest the 9-foot division from the left-hand end of tube, while Nos. 422 and 452 collapsed nearest the 12-foot division. It will be observed that the greater number of the tubes collapsed at points that are at distances of 2 feet and 18 feet from the left-hand end, that is, at a distance of 2 feet from either end, while comparatively few collapsed at or near the middle of their lengths. In fact, this chart shows that more than seven times as many of the experimental tubes collapsed at two feet from either end than at a point midway between the ends. In order to have this chart show the relation of the point of collapse to the nearest end of the tube, it is obvious that we should transfer the test numbers of the right-hand half to the correspond- ing columns of the left-hand half; for example, we should trans- fer the test numbers over division 18, which is two feet from the right-hand end, to the column over division 2. This has been done for all the columns of the right-hand half of the chart, the dashes shown being made to represent the test numbers of the right-hand half of the scale transferred to the corresponding columns of the left-hand half. Since these experimental tubes were obtained by sending in orders in the usual commercial way, presumably they were taken at random from the company’s stock, and, having been handled several times before being placed in the test cylinder, it is ob- vious that, since it is not known in which direction any of the tubes were passed through the mill while being manufactured, no sig- nificance can be attached to the fact that a greater number of the tubes failed nearer the left than the right-hand end. This chart, however, shows very clearly that the bulk of tubes placed under test were least capable of resisting fluid-collapsing pressure at a point near one end. The reason why the bulk of these tubes collapsed near one end is evidently due chiefly to the following two facts, namely, (1) that a tube subjected to collapsing pressure is weakest at the point where the departure from roundness is greatest, even when this is small, see Tig. 54, and (2) that the greatest departure from roundness for the bulk of these tubes was near one end, see Tig. 53. 78 COLLAPSING PRESSURES OF LAP-WELDED STEEL TUBES. Fig. 53. COLLAPSING PRESSURES OF LAP-WELDED STEEL TUBES. 79 Chart Showing Relation of Greatest Departure from Roundness to Length of Tube. Fig. 53 shows at a glance how the place of greatest departure from roundness is related to length of tube. For an explanation of the manner of construction see the description of Fig. 52, the two having been constructed according to the same general plan, the only difference being that Fig. 52 shows the location along the length of the tube of the point of collapse, while Fig. 53 shows similarly the location of the point of greatest departure from roundness. These two charts, taken in connection with Fig. 54, show that the element of greatest weakness in a commercial lap-welded tube is its departure from roundness, even when this departure from roundness is comparatively small, as was the case with the tubes tested. Comparing these three charts with Fig. 56, it will be seen that the thinnest portion of wall, while in itself an element of weakness, is wholly subordinate to out-of-roundness in its in- fluence upon the collapsing strength of commercial lap-welded tubes. Relation of Axis of Collapse to Smallest Diameter of Tube. The autographic calipering diagrams taken from the tubes be- fore being placed in the hydraulic test apparatus show, as was to be expected, that none of the tubes tested were exactly round. This departure from roundness, while measurable by the refined methods used for its determination, was, nevertheless, small, vary- ing all the way from zero to as much as possibly 2 per cent, of the diameter. It is apparent that, for homogeneous material and uni- form thickness of wall, a tube whose cross-section is not circular will start to yield in the direction of its smallest diameter, and the axis of collapse will be coincident with the original smallest diameter at the place of collapse. That the slight out-of-roundness of the tubes tested was the chief factor in determining the place of collapse is quite apparent from an inspection of Fig. 54. This chart shows, for each tube whose test number appears, to the nearest 5 degrees, the angular distance from the axis of col- 80 COLLAPSING PRESSURES OF LAP-WELDED STEEL TUBES. Fig. 54. CHART SHOWING RELATION OF AXIS OF COLLAPSE COLLAPSING PRESSURES OF LAP-WELDED STEEL TUBES, 81 * 82 COLLAPSING PRESSURES OF LAP-WELDED STEEL TUBES. Fig. COLLAPSING PRESSURES OF LAP-WELDED STEEL TUBES. 83 lapse to the nearest end of the orginal smallest diameter of the cross-section through the place of collapse. Since no significance need be attached to the plus and minus signs on this chart, seeing that had any tube been placed in a reversed position in the test apparatus it would have also had the sign of its angular distance from the axis of collapse reversed, the test numbers having nega- tive angles have been transferred to the corresponding columns con- taining those having positive angles. In order to avoid confusion the places of the test numbers thus transferred are indicated by dashes. Relation of Axis of Collapse to the Weld. Fig. 55 is constructed on the same general plan as Fig. 54, for explanation of which see above. This Chart shows that the angular distance from the weld to the axis of collapse, for the different test numbers, is quite uni- formly distributed over about two-thirds of the possible distribu- tion and shows conclusively that the weld, in itself, is not an ele- ment of weakness for tubes that are subjected to external fluid pressure. Relation of Axis of Collapse to the Thinnest Portion of Wall. Fig. 56 is constructed on the same general plan as Fig. 54, for explanation of which see above. This chart shows a fairly uniform distribution of the test num- bers over about three-fourths of the possible distribution on either side of the axis of collapse, with a somewhat prominent increase over the remaining fourth. A study of this chart in connection with Fig. 54 will lead to the conclusion that the tendency of commercial tubes is to fail so as to have the axis of collapse at right angles to the diameter through the thinnest portion of the tube. It should be observed in this connection that the bending action on the wall of a tube while being collapsed is most pronounced at this same point, that is to say, at 90 degrees from the axis of collapse. It will also appear, from these same charts, that for commercial lap-welded tubes the usual departure from roundness has a more pronounced effect in determining the manner of collapse. In other words, 84 COLLAPSING PRESSURES OF LAP-WELDED STEEL TUBES. when these two influences are related so as to oppose each other the latter almost invariably predominates. Application to Practice of Stewart's Formulae A and B for the Collapsing Pressures of Lap-welded Steel Tubes. Table of Collapsing Pressures and Weights . — The probable col- lapsing pressures contained in the table, Figs. 57 and 58, were cal- culated by means of formulae A and B, see page 66. These formulas were derived from results of tests on 20 -foot lengths of Bessemer-steel lap-welded tubes. They are, however, substantially correct for any length greater than about six di- ameters of tube between transverse joints or end connections tend- ing to hold the tube to a circular form. In the columns headed “CLP.” are entered the probable collaps- ing-fluid pressures in pounds per square inch, as calculated by formulae A and B; while in columns headed “Wt.” are entered the corresponding plain-end weights in pounds per foot length. These weights were calculated on the basis of one cubic irch steel weighing 0.2833 pound. It will be noted that each weight column and the corresponding collapsing-pressure column taken together constitute a double column that is headed by the outside diameter of the tube to which this double column corresponds. Example 1 . Find the plain-end weight and the probable collaps- ing pressure of a lap-welded Bessemer-steel tube whose outside diameter is 6 inches and thickness of wall 0.180 inch. In double column headed “ 6 O.D.,” Fig. 58,. and opposite 0.18 in the extreme left-hand column read 11.19 and 1214, the first being the required plain-end weight in pounds per foot length and the second the probable collapsing fluid pressure in pounds per square inch. This collapsing pressure is for a 20-foot length between transverse joints or other supports tending to hold the tube to a circular form, but is also substantially correct for any length greater than about 6 diameters or, in this case, 3 feet. Example 2. Find the collapsing pressure of a tube 7 inches out- side diameter having a plain-end weight of 17 pounds per foot. From the double column head “ 7 O.D.,” Fig. 58, we find that a plain-end weight of 17.33 pounds per foot corresponds to a probable collapsing pressure of 1,586 pounds per square inch, and also that a weight of 16.63 pounds corresponds to a collapsing COLLAPSING PRESSURES OF LAP-WELDED STEEL TUBES. £5 q! d K' d; ej n! S d o fs Q. d K a: k’ a d Is' £ d' d fs' £ d d n: $ d d Js d d f^ d d fs' $ d' ej K £ s. - g to TO <0 CVJ O TO TO g TO <0 0> N. •* n to oo s V3 <•) s 0 ) TO CO N oo o CD TO TO — o g eg O ^ ' CD K V 3 Oj s O eg g co CD > xj' g* g* ^ O o> <0 <0 o o to vo g ffl O S (O t <3 N •o' o' Vo Vo' o' oo <*) no s eg v, n N V) CD s vj v> <0 to is ^ eg oo Vo oo to oo in . eg Is N (D CQ oo eg cd g O . V) Oi O O s * eg n eg n s- O Of oo oo CO eg’ eg’ e\i o g CO Cq IS s co <0 N Is' TO to o eg g to vo o xf co eg CO C 5 > Oo 0 > o ssnMOI is CO CD OO ffl Q vo o vj- to eg is O s' L eg' eg’ eg eg eg eg eg eg eg oo V) to o xf eg eg s. ■> TO Is ffl ^ TO oo CQ TO — eg O oo TO co N TO to to o eg s' O eg' eg' TO 0) - g to TO Vo Vo Xf eg Xfr TO 00 ^ eq eg eg' eg eg' ej og' CTO eg xf N oo eg eg s O to xj to TO O s. ffl' og' oo' g' xf eg vo N O eg TO TO is N TO 03 TO N TO s xfr' V'W »o' TO CO O CD Vo TO g g ffl evi CD TO K TO s to' to' vo' to' to' co s ej to to S o 0) TO CD TO |s TO O TO' TO' to' TO~ Is' s g go 0 ) s xf CD N TO TO TO D" eg Xf. to CD O eg |s' n' Is' is' TO' Co' o to s n eg N -s TO O TO XJ TO TO TO TO cd eg to oo eg s TO O Xf CO) ffl' oo' o' o' O* TO is q eg g ' og' oo' eg TO O g TO s s eg eg eg N oo S (D TO O)' oo' Xf Xt-' V eg TO O g CO oo CD xf xj- XJ. N ffl s <•) V) g' XJ-' v>' to' TO' eg TO °o cd is TO Vo TO vo co n OO ^ CD TO •s TO oo oo |s N Is IS 00 TO N O) s. to vq oo is o oo ffl Oo O O s |S 0 ) S Xj <0 CD n' is' ffl' co' cd' CO xj xj- TO TO TO n o> s eg g TO a XJ- CO VOCONTOO "s TO Oo XJ Vo < 0 < 00 )OS CgOoxfVoTO TO oo eo S eg <0 VJ D N W « O'NO) XJTO^OO <70 TOVoCO'sV) 03 -XJNO OO TO O) CD CD OoegVoffls. O p> V© TO TO Vooocgvoffl s XJ N O TO OO s eo (o m oono egxj N oegxjisoo s'Cs' s' eg' eg' eg' to'cd' eo' oo' cd' 0 )' xj' xj xj' xj'Vo' V 3 ' vf vo'to to n o> 6 o o VoTOCgVoCO s(*oios TO XJ Vo Vo TOVOTOCDTO o 'TO N C 3 . eo N O CD TO O eq TO TO TO TO TO s xj N o s s s' evi exj evi cd to of xj xj xj xj" vj TO TO to to to n O eo TO TO s 00 03 TO TO 00 O TO TO TO o TO TO xj *» TO eg n eg n s TO TO CD O 00 eg' eg' eg' TO egegegeg ooeoooxjxj' xj s n xj - N TO XJ S <33 XJ eg TO cd o 00 XJ S eg s o TO TO s g n oo soooTON TOxjooegs TONWogP) eg s o 00 TO eg TO co s xj nooototo ssegweg CgeDegeog xjxjxjvoTO vototototo NNNTOTO 0303000300 O C 3 OO TO to <0 s eq N N N CD ffl 00 ao OO* 0 >‘ 00 ' 0 )' o’ .O O Q s n eg n eg xj <0 N OOO cd <0 (Jo eg to to S TO S TO TO TO 00 s eg N <3 to is O s' eg' eg' eg' 03 cd to' cd N \J N- TO' TO - TO to' to'to'nn'n cq' eo' ffl' 00 ' 00 eo'oo N eg to TO TO s (D XJ TO N oo eg to ffl s TO TO TO ffl xj C 3 o O eg TO TO xj CD s g n TO O K xj O TO s TO eo 03 eg* to* to to" to xj xj xj xj TO TO o XJ <0 ed V 3 ffl o CD V 3 n o> TO Q TO TO N- OO s n TO TO TO ffl TO °0 < N xj s eo co eg n eg <0 o, xj N s g exf eg eg' eq eo g. s to xj s s <0 O TO O cq s_ Vq^ cq w TO' xj' xj' g' TO' TO TO’ <0' TO’ TO' Is' N’ K' TO ffl' O)' 0 )' OO ffl g s. eo g xj oo g ffl to TO ffl eg to ffl s co g s © ffl eg n eg to eg eo TO id TO g s ao s TO O O CD is <0 s <0 eg N TO TO CD 3 TO g 0) s O CO N TO CD s O g TO to' to' n" n' iO g ffl g O TO g eg s o eo eg TO O xj n egegegegeo DeDfflCDg ggxjxjTO TOTOTOTOTO n s g co s g ffl s g 00 ffl co > g o d o o o* TO TO N TO ffl d o d d o OOOOTO OOOOO O s eg eo g TOTONfflffl os-tgegg TOfflNaofflO eq eg eq eq eq o ci o' eo o' o o o o o d d o d o d dodo o' Fig. 57. Collapsing Pressures of Lap-welded Sterl Tubes, Calculated by Prof. Stewart’s Formulae A and B. 86 COLLAPSING PRESSURES OF LAP-WELDED STEEL TUBES, Q o' * ci d ^ N XJ- O CM C\J ^ 05 O 0 \ C\J O O) 00 N V 0 > ^ «M ^ O 0 Q N M> ^ Xfr QO'-^N ^10 0 * 0 " <*) <0 ^ CM 0 > N M) CM O <0 VO ^ CM O <*) »o 0 )n^ Mo ^ CM W C\| <0 TO <0 <51 O - s nn^qoi B N © xj © O © © B O TO d <0 B B O © TO XJ >0 TO B B O s W n C) lo N B B O s s © TO XJ d © d TO B B O O co a o' ■»" to xj’ d d to od a o' C\i e«i u vj 0 © O B B ® d XJ © s ei © TO © XJ TO © O B B <0 TO XJ © s o © TO <0 Xj © © © TO B XJ \J lej © O TO B 0) O s © TO TO ^ B (0 TO B B B O'N'i'f <0 S 0 _ 'O ^ <« 0 ) CO s' s — ' -.' s' s' s.' s' s' s' ©' cm" eg ©' eg" eg' eg eg' eg' eg' eg' K * C 0 '~ l o< 0 '- <*> © B o © c*)eo>oNS oo oo co oo co to vo } to to B B TO to to B xj xj xj x$ xj xj xj xj xj xj xj d Q 6 § 0) 0.' o to d to d 0 xj © eo to B d © ® xj 0 to to a vo © b d s to B 0 ' 0 0 o>t 0 10 xj to TO © © eg s s ^ Q 0 a a a ® b to r- 1- © ® a d to xj xj xj to a © eg © © t*> xj xj a © ^ © a 0 s eg to xj a to © n b aos©B xj d © to © a 0 s © <>> xj s -.' s' 0 0 O C O s' s' eg eg' eg' eg' eg' eg' eg' eg' eg' eg' co' 00 eo <0 e*) K' * »o ©'scoeooo 0^©g-io ©ks©© ©sseod ©ego®® ^'sO)©© o>©eg®g© eg s,s.ea©oo oos©io^ (<) eg s o 0 ®rs© | oe}- eneg^o© K©M-eoeg ooi®®*)') © d In ©' ©' N © oi Q " eg © > d © © s © © d "' eg © ^ rj irj © s © oi 6 -s' -s' eg © ^ «n ^ s s s ^ s -s ^ eg eg eg eg eg eg eg eg ©egegeo© ©©©©© ©©©©^ 3 o' $ 'oT® ® Cl d 0 ^ © ©^ xfr © © s © w 0 ^ ©, © ® O O s.' s' -s' s.' s.' s.' O O eg' eg' ©' ©' ©' ©' ©' eg' ©' ©' ©' © ©' ©' ©' ©' ©' K * © ©n©©w ors©©o> ©©e%®w s © © xj- © ©a©©o » n 0 g 0 ©so) © © X*- ©© s. 0 ©N©x}. © © O 0 > © ©©©©O ©l^©XJ.© S.©S 0 > s © © © ©^©©s ©©ejjo's.’ ©'eoxj-xj© ©K©'oio oislcgdxi ©g^©'(s© oi 6 6 s' fti fj ~~ s s s s s ^-s-s©© ©©©©© ©©©©© ©©©©(^ ©©©©© ©X^\txtXJ.X}. Q 0 § © d' d 0© ©o©xj© -s©K©xi. tg s o» mo xjeg-sojs ©x^w-s© N©x^©~ ©©x*©^ g ® D O MO K © B © O ^ ©©©xj© ©isBqjOv eo-s©eo© xt©©N© ©©0^©© N©OX}-^©© NB 9 >"© ©X}-lr>©N ©©0^© 'f'lBMS ©O"©© XJ©fs©©0 s's' s' O s' s' s' s' O eg ©' ©' ©'©'©'©'eg ©' e>> ©' ©' ©' <•)©'©'©' d Xf' K $ ©<0 ©OW ©»0 ©N 00 OB Oi 0 > ov <0 N © © © © 0 NK )©0 N XJ* O © © Ol > O K) O X) oils ©xj- eg 0 00 ©x^egO(^) eqxtwooo ©x»©ooo ©eosOis >egosd ' xj! 0^ © n is © © q ss sssss s s s s s eg eg co © eg ©©©©© © © © © © © © © © © © © © © © g. d o $ 'olco to a.' d ^^ocooxi N^*oooca jo oicon 0 ^a}'^ , ooi cvjioocor^ m- cm «o c^cn'oo^N ’ 0 n OioCViC*)^* ^rxOOO^ C^M-'ONOaOi CVJCO C 0 ^» 0 'O N 00 O *** C \1 (*) O^CN|C»>^ lijMDftO ^CMO)W)U) fs. Q)-Oj O ^ CM -s' O O -s' O -s' s ' C O cm' cm' cm c\f cm' cm'cm' cm' cm' c > crT co p> o)' ci ci ci c> > ^ ^ K £ s© B lOflOMI ©<0S(0© ©exlNslj © © © 0 X^ NBWXJ-© ©O©©^© xj-s oi © V) © 0 co©xjs© ) 0 o' s' s' eg cxi x* wi «> is © 06 ©’ o' s' s evi e>i x^ ^ vj d ©' n 06 oi o' o' s' evj ©' © ^ vj © vo n ©' SS SssSS SS.SSS S © © © © © © © © © © © © © © © © © eo © 0 ) © © © © © Q o * TO d' d so 'O’OOB® Oxi ©©<0 0)©NsO) 03 W © 0 xj, |ssioO>© rsOXJ-B© ©©©Ns© ©0 ©Bsxj-® © s © »o <0 OB*oBO ©Co so© \}.s©sxj. gj©s©© qjo©io©o © X* X|- lo K © 0 ) 0 © © x^lo N 0 )® ON ©X 3 loS N © © O S © lo®B ©0 sPixi-^io© s' s' s' s' O Css’ cow' ©'©'©'©'©' ©'©'©'eg©' ©'©'c>)©'xj-' >' x} xf x).' xj-' x^' to' $ Xf® SOOs© ©©TO©© ©©SOB Sd^^O O © © N © 00©) xj d d ® s* 06 00 ©' 0 d s' eg oi ©’ xj xi 'vi d d s 06 eo oi d 0 s eg eg © xj xj s SSSSS SSSSS sss©© © © © © © © © © © © © © © © © ©©©©©© Q O 2 •oT® d d -0(0 ©s©ao 0 s © © xj dd'° N ® ©00s© © xj xj d <0 rsB®°>o s©©©xj-d s©(o s 0 © <0 0 ©) ©' ©'©'xjxj'xj xj xj-'xj XJd'd' TO $ x>©o sd©B® ©©dss ©s©sw ssdBeo ®o>©dis e>)s©d© bbooss eo 0 IS TO0N©0 Js © O S © OOBB® ©Olds© XJolstiB d©©XJO ®©Bds-l^ ©’ 0) 0) o' s' s evj © © xj d d © n s m © B d o' s' c\i exi oi xj xj d d d s s’ © © oi 0 0 s' © ©' Sssss SSSSS SSSSS © © © © © © © © © © © © © © © ©TO©©©© ci o $ <0 o! o O -(0 sdOxjB ©S©«s dOxjB© S©®sd OXJBO)© © o' s-Niiti ci xj d v> © © s.is b b oi sss SSSSS SSSSS sssseg © © © © © © © © © © ©©©©©© d o' d d 0 © © 0 B>oB,o® (o©s®(o xj s b (o xj sa^xj© BNd©0 tsdegoaj dBOB©B Oseo© N©BdO © © B C 1 } © d — © © © xjBd~N ©®xjo d d xjTOod odo®s ©sgjs© s (o s © s n © n © n ©is©n© ntobtob to b to © to ® d B © eg Bd©®d s co xj s n xjoncoo ©tobow Bd©Bd scoxjsK xjontoo© XJ d K B 0 © XJ TO Is 0 )O©XJd ISBOWXJ TOIS® 0 © to d TO B o ©tototoB 0 © to d to © s' s' s' s' s' s' ©' ©' ©' ©' ©' ©' eo' to' to' to' to' to' xj' xj xj' xj' xj' xj' d' d' d' d' d' d' ©' ©' ©' ©' ©' ©' TO > xj to s xj K 0 © TO to a 0 © eo xj to©©©® © © *o xj to egO©©xj ©bto»o totobtosTO 0 TO S to s TO © TO © TOTOBB® V 0 _ ® B © B © eo © TO © ep © © TO © TO s (0 s © OdBxjaB d © to to © © a b* d o' s' s' © eg to co xj xj to b © d to to b b oj b’ d o' s - © eg e»j eo eo xj xj' d S SSSSS SSSSS SSSSS SSSS© © © © © © ©©©©©© 0 o' <9 d d TOXJKs xj TO s XJ TO " XJ TO s xj TOsXJBs xJBsXJ© s s qj s xj OO s XJ <0 s xjBsXJBs fgOcoxj sooBTOO BdBOTO xjcyagjxJ sB©TOO Bd©OTO xjcgacoxj ^B©BoB BOBo © TO d to 0 } 0 ©XJ©TO B s © xj vo BBsTOB toBO ©(0 dTOBO© ^.d N Bs© s' s' s's' s' s' ©' ©' ©' ©' ©' evT to' ed' to'b' to' to'xj'xj'xj' xj xj' to' to'*o' *o' d' d' ©' ©' ©' ©' ©' ©' to' to' K S XJ d © TO TO N TO TO © dxJTOTOO B©xjsa © TO B © © ©xjodo toOXjBCO TOOXJTOOTO N eg to © nwtowto © k © to © it^esi) o to b xj a tobtoto© ©sdaxj btotosbo d ©’ d to to © © a' a d d s s ©' w B to xj xj d d d d © to to b © a" b o' o o' s' s' ©' eg eo to xj sssss SSSSS SSSSS Sssss s © © © © ©©©©©eg Ht o s © to xj d © to b B o s © eo xj d © to © a osegeoxj d © to © a Os©eoxj d©TO©BO sssss sssss n eg © © © © © eg © © tototototo btotototo xjXjxjxjxj xjxjxjxj^d ooooo ooooo o p o o o o o o o o” o’dooo o’ o o ei o d d o o' o’ o' o’ o’ o o d Fig. 58. — Collapsing Pressures of Steel Tubes. — Continued. COLLAPSING PRESSURES OF LAP-WEI.DED STEEL TUBES. 87 R.Tj Stewart Williams Eng. Co., N.Y* Fig. 59. — Chart showing the Values of the Table of Collapsing Pres- sures of Lap-welded Steel Tubes, Figs. 57 and 58, Co structed , to a Vertical Scale of Collapsing Pressures and a Horizontal Scale of Thickness of Wall. 88 COLLAPSING PRESSURES OF LAP-WELDED STEEL TUBES. pressure of 1,462 pounds. Now, by the usual method for interpo- lating it will he found that for a plain-end weight of 17 pounds per foot the corresponding collapsing pressure will be 1,527 pounds per square inch. It is believed, however, that the tabular values in this table are sufficiently numerous to render it unnecessary to make any interpolations whatever while applying it to practice. Factors of Safety . — It must be remembered that these tabular values represent the probable collapsing pressures as based upon the tests. This being the case, any individual tube is as likely to fail above as below this most probable pressure. The relation of the collapsing pressure of each individual tube to the most prob- able, as tabulated, is clearly shown in Figs. 48-50, where the curves represent the tabular values, crosses the collapsing pressures of individual tubes, and combined crosses and circles the adjusted group averages. Expressed in per cent., this variation of each in- dividual collapsing pressure from the tabular is shown in column 9 of Figs. 43-46. This table shows that not one of the several hun- dred tubes tested failed at a pressure lower than 42 per cent, of the probable collapsing pressure, while \ of one per cent, of the number of tubes failed at 37 per cent, and 2 per cent, at 25 per cent, of that pressure. In other words, with an actual factor of safety of 1.75, as based upon this table, Figs. 57 and 58, not one of the tubes tested would have failed. From an inspection of the charts and table above referred to it would appear that : 1. For the most favorable practical conditions, namely, when the tube is subjected only to stress due to fluid pressure and only the most trivial loss could result from its failure, a factor of safety of three would appear sufficient. 2. When only a moderate amount of loss could result from failure use a factor of four. 3. When considerable damage to property and loss of life might result from a failure of the tube, then use a factor of safety of at least six. 4. When the conditions of service are such as to cause the tube to become less capable of resisting collapsing pressure, such as the thinning of wall due to corrosion, the weakening of the material due to over-heating, the creating of internal stress in the wall of the tube due to unequal heating, vibration, etc., the above factors COLLAPSING PRESSURES OF LAP-WELDED STEEL TUBES. 89 of safety should be increased in proportion to the severity of these actions. Example 3. By means of the table, Big. 57, find what thick- ness of wall a 4-inch boiler tube should have in order to withstand a working pressure of 200 pounds per square inch, with a factor of safety of eight. In this case the probable collapsing pressure should equal the working pressure multiplied by the factor of safety, or 1,600 pounds. How, looking in double column headed “4 O.D.,” we find the nearest tabular collapsing pressure to be 1,647 pounds. This corresponds to a thickness of 0.14 inch or Ho. 9 B.W.G., as read opposite in the extreme left-hand column. Example Bind the plain-end weight per foot of a 6^-inch casing to withstand a maximum difference between external and internal fluid pressures corresponding to a water head of 800 feet, on the basis of a factor of safety of four. A table of hydrostatic pressures will show that this head of 800 feet will create a fluid pressure of 347 pounds per square inch, tending to collapse the tube at its lower end. Multiplying this by the factor of safety we get 1,388 pounds per square inch as the probable collapsing pressure. How, looking in double column headed “6f,” which is the outside diameter of a nominal 6^ casing, we find the nearest tabular collapsing pressure to be 1,361 pounds, which corresponds to a plain-end weight of 14.39 pounds and a thickness of wall of 0.21 inch. Chart Showing Relation of Collapsing Pressure to . — Big. 60 resulted from plotting equations A and B (see pp. 66-68) to a vertical scale of probable collapsing pressures and a horizontal scale representing the thickness of the tube divided by its outside diameter, or the ~ contained in these formulae. a By plotting in this manner, a single line may be made to repre- sent the collapsing pressures of a great variety of tubes, irrespect- ive of their individual diameters or thicknesses of wall. It will be noticed that this is the same curve as that shown in Big. 47, the difference being that it is drawn to larger and more conveniently read scales. In order to condense the size of this chart the curve is broken into the two parts XX and YY. By this means the area of the chart has been reduced to about one-fourth of that which would 90 COLLAPSING PRESSURES OF LAP-WELDED STEEL TUBES. otherwise have been required to construct the chart to the scales shown. It will be observed that YY is the upper portion of XX transferred to the left and then dropped down, the break in the curve corresponding to a collapsing pressure of 2,080 pounds and a thickness divided by diameter of 0.040. It will also be observed that the scales for the portion XX are at the lower and right-hand margins, while those for the portion YY are at the upper and left- hand margins. The smallest divisions on the vertical scale represent 10 pounds collapsing pressure, while those on the horizontal scale represent 0.0002 thickness divided by outside diameter. When reading to the nearest smallest division on these scales the error will not ex- ceed five pounds for probable collapsing pressure, nor 0.0001 for thickness divided by outside diameter. This, then, is a universal chart showing the relation of the probable collapsing pressure of a tube to the thickness of wall divided by outside diameter. It represents the adjusted values of the group averages of all the 20-foot lengths of the Bessemer steel lap-welded tubes tested, omitting the three that proved to be defective, and may therefore be used with entire confidence within the range of these experiments ; that is, for Bessemer steel lap-welded tubes from 2 to 12 inches outside diameter, and for all commercial thickness of wall in lengths greater than about six diameters of tube between joints or end connections tending to hold them to a circular form. Example 5. Find by means of Fig. 60 the probable collaps- ing pressure of a tube having an external diameter equal to 6 inches, and a thickness of wall equal to 0.203 inch. Dividing the thickness of wall by the outside diameter we get equal 0.0338. Since this value is less than 0.04 we look for it on the scale at the lower margin of the chart. Having found it on this scale, look along the vertical line through it until the line XX is reached; then look along the nearest horizontal line toward the right and read from the scale of probable collapsing pressures 1,540 pounds per square inch. This is the probable col- lapsing pressure for a length of 20 feet, but is also substantially correct for any length greater than about six diameters, or 3 feet for a 6 -inch tube, between transverse joints tending to hold the tube to a circular form. Linear Units for d and t . — It should be noted that both the out- COLLAPSING PRESSURES OF LAP-WELDED STEEL TUBES. 91 side diameter, d, and the thickness of wall, t, must be expressed in the same linear unit of measure, as, for example, in inches, centimeters, millimeters, etc. The name of the linear unit is im- material, the chart being just as applicable to obtaining probable collapsing pressures in pounds per square inch when the diameter and thickness are expressed in metric as when in English units. Collapsing Pressures in Metric Measure . — First divide the thickness of wall, t, by the outside diameter, d , both being ex- pressed in either inches or millimeters. Second, obtain from Fig. 60, as in example 5, the probable collapsing pressure in pounds per square inch. Third, reduce the resulting collapsing pressure in pounds per square inch to that expressed in kilograms per square centimeter by multiplying by the conversion factor 0.0703. Example 6. Find the probable collapsing pressure of a tube whose outside diameter and thickness of wall are respectively 15 centimeters and 4 millimeters. Fifteen centimeters being equal to 150 millimeters, g , or thick- ness divided by outside diameter, equals 0.0266. Proceeding as in example 5, we find the probable collapsing pressure to be 920 pounds per square inch. Multiplying this by the conversion factor for reducing English to metric units, given above, we get 920 multiplied by 0.0703, or 64.7 kilograms per square centimeter. Example 7. With a factor of safety of eight find what thickness a 3-inch boiler tube should have in order to resist a working exter- nal fluid pressure of 220 pounds per square inch. In accordance with these assumptions the probable collapsing pressure of the tube should equal the working pressure multiplied by the factor of safety, or 1,760 pounds per square inch. From Fig. 60 find 1,760 on the scale of probable collapsing pressures at the right-hand margin, and look along the horizontal line through this point until line XX is reached; then look down the nearest vertical line and read 0.0363 as the value of or thickness di- vided by outside diameter. We can now get the required thick- ness of wall by multiplying the value of ^ by d , which gives us CL t equal 0.0363 X 3, or 0.109 inch, or Xo. 12 B.W.G. For the same conditions of pressure, a tube 8 centimeters, or 80 millimeters, diameter should have a thickness of wall equal 0.0363 X 30, or 2.9 millimeters. 92 COLLAPSING PRESSURES OF LAP-WELDED STEEL TUBES. ' . . . W Chart Showing Relation of Collapsing Pressure to . — fig. 61 resulted from plotting equations C and D (see page 68) to a vertical scale of probable collapsing pressures and a horizontal scale representing the plain-end weight per foot divided by the square of the outside diameter, or the contained in the for- mulae. The errors of reading this chart should not exceed 5 pounds for the probable collapsing pressure, nor 0.001 for the weight divided by the square of the outside diameter. This chart is based upon precisely the same experimental data as Fig. 60, the difference being that for any given size of tube this chart shows the relation of the probable collapsing pressure to the plain-end weight, while the preceding chart shows its rela- tion to the thickness of wall. This chart should be used in cal- culations relating to collapsing pressure when the plain-end weight is either given or required, while the preceding chart should be used when the thickness of wall is given or required. Example 8. Find the probable collapsing pressure of a 6f (7 0. D.) inch casing whose plain-end weight is 17 pounds per foot. Dividing the plain-end weight in pounds per foot by the square of the outside diameter in inches we get equal 0.347. Find- ing this value on the scale at the lower margin of Fig. 61 we look vertically until the line XX is reached, then look horizontally toward the right and read 1,525 pounds per square inch as the probable collapsing pressure required. While this value is for a 20-foot length of tube, as in the pre- ceding chart, it may be used without substantial error for any , length greater than about six diameters, or in this case 3^ feet, between joints tending to hold the tube to a circular form. Example 9. Find the plain-end weight per foot of a 5f-inch casing (6-inch O. D.) to withstand a maximum difference between external and internal-fluid pressures corresponding to a water head of 1,200 feet, on the basis of a factor of safety of four. A table of hydrostatic pressures will show that this head of 1,200 feet will create a fluid pressure of 520 pounds per square inch, tending to collapse the casing at its lower end. Multiply- ing this by the factor of safety we get 2,080 pounds per square inch as the probable collapsing pressure. Finding this value on COLLAPSING PRESSURES OF LAP-WELDED STEEL TUBES. 93 the left-hand margin of Fig. 61, we look horizontally toward the right until line YY is reached, then up the nearest vertical line and read 0.41 as the value of-^- > or plain-eud weight di- vided by the square of the outside diameter. Now since equals 0.41, w will equal 0.41 multiplied by the square of the outside diameter, or 0.41 X 36 = 14.76 pounds per foot, as the required plain-end weight. DISCUSSION. Mr. Henning . — I wish to compliment the author for the char- acter of the paper that he has presented as, in my opinion, it contains valuable information which the Society was not in possession of before. Of course, a great deal of time has been spent on it, and somebody has spent a lot of money on it, but it gives us information which is valuable. I wish that manufacturers generally would give us as much information about what they produce as has been given here, so that we may know what material will do after it is finished, and not alone what it will do when it is in the shape of a test piece. Mr. William T. Donnelly . — I would like to ask whether the tube was tested while in the horizontal position or whether it was placed in a vertical position for testing? Professor Stewart . — These tubes were all tested in a horizontal position. An investigation was made as to what effect the position would have upon the collapsing strength of the tubes, and I satis- fied myself that it had no noticeable effect. Many of the tubes were of such weight as to tend to float up, and they would have floated to the top of the test cylinder had they been permitted to do so. This was due to the fact that while under test the tubes were open to the atmosphere on the inside, and were sur- rounded by water on the outside. Others, of course, tended to sink. In either case the resulting strain was quite insignificant. Mr. Rice . — As Chairman of the Committee on Papers, I desire to express my personal appreciation to the author of this paper. As far as I remember, it is the most remarkable paper presented at any meeting, and it represents an indescribable amount of work, and I think we should take special notice of it on that account. Then it is notable in respect of the contribution it 94 COLLAPSING PRESSURES OF LAP-WELDED STEEL TUBES. makes to the knowledge that we have on the subject. I consider that one of the special functions of the Society, to contribute to human knowledge. In this connection I want to bring out that many of our members may have knowledge of valuable data or information which, if the request be properly made to the gentlemen who control it, will be permitted to be published for the benefit of the pro- fession. A Guest . — Some time ago I had occasion to go before the Board of Supervising Inspectors on a subject which is covered very largely by this paper. The occasion for it was that the Board of Supervising Inspectors, at one of their meetings, in their wisdom, had made a new rule designating the thickness of flues in steam boilers coming under the regulations of the marine service, upon a formula which took no cognizance of the length of flue. It was not Clark’s formula, hut I think some formula that had been adopted by the British Lloyds before the Fairbairn experiments had been measured up. For a year, under the hasty action of the Board of Supervising Inspectors, the condition of the work done for marine practice was quite chaotic, but was finally relieved to some extent by the Secretary of the Treasury suspending the rule. How I hope that this paper will come to the knowledge of the U. S. Board of Supervising Inspectors, because they need it. They need it now almost as badly as they did at the time I speak of. Through the efforts of some person, whose interests were more largely concerned perhaps than the members of the Board of Supervising Inspectors, a formula was adopted which did take cognizance of the length of flue. How, I would like to ask for my own information from Professor Stewart, whether he knows from his experiments how they com- pare with the existing formula of the Board of Supervising In- spectors. How as to the matter of the failure of the tubes. The experi- ments would seem to indicate that the tubes which were tested have perhaps in rolling been laid on a stand, which tended to flatten the tubes at the particular points where they were laid down. Professor Stewart . — I have not made as yet any such compari- son as has been spoken of ; but I am satisfied, however, that plain commercial tubes in lengths of about six diameters are no stronger, to anysg>rpj^feftb¥x^t^peast, than similar tubes having lengths up to say zO feet or more. This is clearly brought out in the m o 3 1004 . .72 .. 10 V .fwaami • i •• ■ • - I £3\ft3fe -t£± art'git V- iN* Ufcfei \\t.\>: >!V> V.V'*' , Vi\ W 0*4 >\ |»*1 , v '* vva *\ vm.qi v.\ ;' v : "VOi r ■# *■ t > • B. t>S > o ! i*\( r - 1 . * j \t> ^kStjVA v. ,| \vto*** I n -' : l " •“ /•••• 4 V\;b fVt’= ^ , •tv/* , ! . w\,b J J*A,V • <511 ' ' - "THU' 1 * • ,. mM ' : ivu ... 01 1; t \\*£ U*t •vwittvjVil 2^.1 ■Ol ■* Transactions American Society op Mechanical Engineers, Vol. 27. Reid T. Stewart. I SHOWING THE INFLUENCE OF LENGTH OF TUBE ON THE COLLAPSING PRESSURE, SERIES 1 < for lengths of zi to 20 feet, between end connections tending to hold the tube too circular ( form. For on outside diameter of Ox inches and thicknesses from 0.130 to 0.322 inches. PRINCIPAL RESULTS OF COLLAPSING TESTS ON NATIONAL TUBE CO’S LAP-WELDED BESSEMER STEEL TUBES CONDUCTED BY PROF. RT STEWART, 1902-4 F PK ,1905 Test Number Outside Diameter Thickness o f Wall Length of Tube Weight at Tube Collapsing Pressure Collapsed Portion Physical Properties Chemical Analysis % Material Remarks Commercial Designation of Tube as Reported Nominal Average At Place of Collapse nominal Average As , Pounds Sg Inch Used Length End Distance tb tren s ,h / Sgllru 'Inch Etanfohen %>n^S of Are* S.hcon Pho 3 Mang Carbon Or..:,! L.o„ Lease Nominal Actual In Feet m a, a; , 2 157 cm 0.190 o no 20 120 19 9/8 15.92 4 SO _ *' o - 9 3 ,, O' - 12“ 5g 410 35 670 26/3 57 tO 005 .069 109 35 08 2 9 63 7 o./ts 0./79 19 992 1? 696 17 24 625 6' 3 m 2 7 IS 6' - 30“ 60 490 39 030 21.04 57.20 006 /It .3 2 .075 3 t 625 t 64/ 0 180 0 / 84 0.162 20' 0* /9 997 19 695 16.07 16.77 535 B 5' 6‘ 4 3 * 60 020 36 420 24 00 58 73 .006 II 5 .3 1 .0 75 Bessemer fi Casmg 10 07/0* 4 9 440 — — 0 IS3 0/9/ 0 171 20.004 19 702 16.54 450 — S’ 6' 7.7 15 3" - 30“ 59 640 35 S 10 2/92 59 43 006 .079 3 2 02 — Steel — 5 9 631 — — 0.191 0 197 0 173 20.0/1 19 709 17.23 620 — 5 ' O' 7.0 4 ?' + 15“ 5 ? 160 36 350 23.00 5 7 50 .010 .067 .106 3 / 0 75 — — firerogt t 643 0 125 0.171 o nz 2a.au 19.724 U.74 536 S’ t 7.7 5? 344 3* 396 23.22 5 2. S3 007 074 3 2 .07 7 4 t 65/ — — 0 196 0 .192 o 184 / 5.0/0 14.708 / 6.92 S75 — 6‘ 0 " 2.3 4 O' - t' 52 100 37 060 If 54 52 60 .015 .072 .108 3 5 075 7 t 65/ — — 0 174 0./7S 0.145 15.025 14.723 15.76 425 — *' 0" 8.3 9 O' - 27“ 56 410 35 610 20 75 5 6 90 .008 .06 5 1 C 5 3 3 ■ 08 — — f 9.525 f 6 55 0/80 0.193 0./94 0.162 /s'o“ 15.004 14.702 16.07 16.545 55 0 B 8.3 - 25“ SS500 36 600 2/46 57.40 .015 073 .103 .3 3 .075 Bessemer ti Casing 16 07 lbs 9 g *55 — — 0.191 0.2/5 0.1 82 15.010 14.709 17.21 6/0 — 6’ O' 2.3 4 o' -173“ 52 520 37 890 2/ .79 5 6.70 .008 ■ 073 .112 3 3 .07 — Steel — 1 0 g 652 1 — 0.199 0.191 0.192 15.002 14.700 16.95 5 90 — 6' 6" 9.0 7 6' 4 14“ 59 950 36 S 50 23 21 55 30 .006 .071 34 .0 75 — — Ruermg* 2.453 0 194 0 195 0.172 15.0/0 !4 708 16.66 5 49 *' r 2 4 57 496 36 742 21 15 56.98 .010 .071 .106 34 075 , / g 662 — — 0 194 0.222 0 192 10.017 9 7/5 16 65 575 — s’ o " 7.0 3 f 4 27“ 57 590 35 340 22.52 60.50 .008 .068 .in 3 5 .02 / 2 g 654 — — 0 180 0.190 0./74 9.993 9 69/ 16 26 570 — 5' O' 7.0 3 /' 4135“ 59 410 38 5 20 23.75 60.30 .010 .076 .110 3 / .075 — — t 62 5 0 180 0 177 0 197 0.17 3 /o' o' 9 997 9 675 16.07 / 6.03 59 0 B 5' 6" 7.7 3 55 990 34 200 24.59 60.47 .065 .2 3 .02 Bessemer / 4 9 649 — — 0 17/ 0 .190 0./55 9.797 9.695 15.5/ 455 — * 9 ' 0" */ 2.5 5 9' 4 67“ 60 220 37400 21 67 51.20 — .07? 110 3 8 .02 5 — Steel R • 5 8 65/ — | 0 178 0 197 0.173 !0. 007 9.7 OS I6.H 550 — S’ 6 • 77 5 3' 4 90“ 59 490 38 430 21 98 57.10 .006 .069 .113 .3 5 .07 5 — — 4tere,e 8 656 0 178 O./ff 0.171 1 0.002 7 700 16.1 1 542 S’ J" 7.4 St SI9 36 77 8 2/ 8? 57.91 009 .071 .112 32 .079 / 6 2.650 — — 0 182 0 196 0/52 4 990 4 688 16 43 6 / 5 — 5' O' 7.0 2 *- 4 32“ — — — — — .070 106 3 2 07 t 7 9 65? — — 0 lit 0 .197 o no 5.0/0 4 708 16 42 575 — 5 ' 0' 7.0 2 7“ -173“ — — — — — 062 J g .075 — — K 12 5 0.180 0 180 0 ISS 5 O' 4.702 / * 07 1 6.34 5 40 B 5' 0" 2 8" 4/07 * 076 II 7 2 g 07 Bessemer tiCnmn, It. 07 16. / 9 8.645 — — 0.173 — — 5.020 4.7/9 15.64 5 25 — S’ 0 • 7.0 2 6' - 1 2“ — — — — — 0 9 f 113 3 7 02 — Steel — 20 8 45 4 — — 0 122 0./90 0 179 5.005 4.703 16 43 70S — S' 0" 7 0 2 3' - 23“ — — — — — 074 .112 34 .07 — — R.crage 9 65 g 0 180 0 190 0/6$ seat 4.704 16 25 see 5- O' 7.0 072 in 073 2 1 g 65 1 — — 0 182 0/92 0 169 2 500 2 / 99 / 6.45 9/5 C — 2' 6’ 3.5 l 4- r-iii- — — — — — .025 121 .2 6 076 — — 2 2 8 657 0 162 0.189 0 177 2.5/5 1523 2/5 B 2' 6 - 3 5 5* 4/35 * ,02 5 .35 075 2 3 9 62 5 9 647 0 180 0.185 0 16 5 2’ *' 2 522 2.220 /* 07 16 4/ / 0 9 5 C 2 ' 6" 3.5 <$* -/ 12“ 075 32 .075 Bessemer 84 "Casing lb 07 tbs 24 8.655 — — o no 0 175 0 140 2.5/4 2.2/2 154/ 1095 c — 2' 6" 3.5 / 3' 0“ — — — — 100 118 ■ 3 6 .08 — 5 reel — 25 g 64/ — — 0 191 0 /89 0./63 2.507 2.205 1643 99S C — 2’ 6 " 3.5 • 3“ 1 4iii“ — — — — — 07 5 112 3 2 07 — — 9 656 0 176 0 18 5 0 163 2 5 12 2 2/0 15 99 977 2' *' 3.5 024 114 32 075 2 C 8 604 g St 0 219 0 230 0.2/0 /?' 2" 19/69 18 866 /9 57 8 70 e s‘ *• 7 7 !4 /O' _ 5 ge 5 6 700 34 060 22.7? 57 40 006 068 105 3 8 .07 R 2 7 9 622 g 44 0 233 0 239 0.186 / 3' 8' 13 675 13.373 20.75 II /S c — 5' *" 7 7 to / - - 7“ 57 770 34 730 22 33 53 70 — 070 102 3 2 075 — R 2 g g *25 8 659 8 6(- 2 63 0 22? 0 2/3 0 222 0 195 /2'/r 12 932 12 630 20 !0 19 23 9 S 0 B / 4 5' 6' 7 7 2 2“ - 67“ 60 530 37 700 /6 67 57 20 — 023 117 3 5 02 — Bessemer — ti' Casing. 20 tO tbs 29 9 650 t 69 g 62 0 213 0 22/ 0 180 / 2' 8' 72 674 12.372 19 16 7 50 B / 4 5’ 6" 7 7 f / - -145“ 6/ no 37 3 00 23 67 59 70 — .00? 1 1 1 4 5 .01 — Steel — 30 . 2 670 8 49 9 46 0 197 0 2/1 0 175 / 2' 3' 12.252 1/ 950 / 7 79 6S0 a / 4 5’ 6 m 7 7 4 2" - 58“ 60 250 37 560 20 71 55.70 .006 062 3 9 075 — — Rrfragr 2 442 ... 2 62 0.2/5 0 22 S 0 ,,7 IT 34 247 1.4 5- 7.7 ST 404 36 270 2/ 2? 57 14 072 10 9 3 8 .076 Fig. 11.— Tabular Statement op Principal Results op Tests, Series 1. .Y£ .joV , easts vriorcil jaoiwahosM so ythiooB k/.imiulA t-\ wu.?:. . ;-.T . d i. . f* Y \>S C.S ') • >’ VVV& ) .' ." jr.AV'.^Y \Cr *tAv> < D '•'A sU kb *-.*tfV* s*s MkbV***’' 1 SStt.rt'SA ‘ A ', - .'.' v A *.'?.■>» , 41' it. ? ' \.S ‘ - *4 'T‘ 4 ■ ' I'ek *»_ • T ' '» , i&.O. ■it* s 4*, a S 4 v 4 S Vit *,i 44 i < -S " M)V0 0 *VS 1 £ > -i '-44 S Ul SJ • i \ . .A 0 si'- 0 " 4 ’ 1 Jim : Aav:_ p t . : VVh . v Vovn ' -) ; \ . i ts 4 \ A 4 ■, •. \ \«» 0 'III VlSte| r 7 1 ' *w « •. • SM . : F iUjiiiiifS- t. 44 \ <10 .4*. 441 -i too O' t »»* % -■ v • ' . f. m.4 .'1U t.ti v \ k ! V «t> •»n & Civlo ‘ > S 0 0 "A V 014. S i HIT K vevo ' "il 1 t*\ . SOS.SS 1st Kii ; & k *** V Mi.5 • tio SiS.’vj j •A 4.1 V4 4A n./' 'iSfl VS Oti * -i 4 U.|9 0?. 4 s 4 V ■>nu *:. vi vis o VS', A OS 4.« s 1 £ i ■. . 4 ? 'S.'O OH.O Ot i i sK» « *vv>, 1 -.41 4 4,.'. s 44 4 % t k S5>.S '"+'i J ..itS.'O S' '. r, 014 ? i y* ? m.o ‘ 1 ? t ?irt - 1 ! 0 4 i V Jit 0* *45 0 ■ ■ . fH/' uv; » 4'.;tk ' v S \ ^ 1 *♦* V ' ; ■ ■itt V f.is J ■ S.V S.O iS>..s ■ i'S - | i » t*t v. M.O oosio ViS 4 .: >4 _ — — - - — - Z1Y OSS. 4 \*vo *0' A. . ** M o , VV v ', Vv.:. ! i ?. SiO iS ■ ut o 1 4\f.;a 5AS.0 1 ft ! S\0 £' •mo ; .*U US4 it, s . « » N- Sis 0 SIS. •- <■>■ * < v : u 'AV L I '.n.: I Si' t ■’-TT-J ;;M‘* - u ■-Vi" j svso ‘ a 1 t , K'F J . ^ - % ~ -• r ~ '* : ‘ 7 Transactions American Society of Mechanical Engineers, Vol. 27. Reid T. Stewart. ,T£ ,joY ,BHaavTi 8 V[|jI JtAom/.HoaM 10 ytsiooS m av w i.\ao av-r wo -aavjT 10 v, owsa -aawiurwvx ■am ovvnovxz. kA. V SUVA VaVwA o'* \>rV\\M>rs\ t«\uVk-v*A^r\\ i& \t> ■vs»\«>mt..'*» &^Vi» «» \o\ ,m\ •, It** ■ •*4viT i vw» ?.?T.rv A^vs\T s^bve\ :.' '': t » .virv' p 'A 1 1 ^ w •* !;, IV *1 J » l A", A »«t*.»\i«** i*S.O m.i i'VS.a : fcV^.t ,m.v. ‘V * >-M t . OVCV.ftt ( *iS,*V IU.I , a*s i . asi.i as*.» ' ’ uti \ *• It* t mt *ft 'fiK VAX.* ; aH»’ as*.t \t*.v . ti'* t ; ‘ ~ « t*V * aw.t }♦•*.% an A V.S.I, |0fvjt J Wi.< , as*.'# f . ' *4* t ft**.*! 1 ftCVA et£v« <»*S.9 iiv^t av* % ia*s a* ’**, t a ix.ft SIS. A .‘iO-C iH.) ■ «!,A fc J 'ft it» k> a >*.* ! 'XS.'4 Ik* *■ |/wiv ! 5*S.ft ■ ' ***A. an.| ta*.« * A t»V t. | atr> j *Vz - aas* ! 'n£.6 l 'AS.ft J av*,» S.4 S . }*•« as* % V A U-, V }■&**.»> • r n. : a ass 1 K.»S,ft <*s •> 8*i» a * v. i AV-.S , t <* vt- .V 1 ttt.V . S*M ! tn.ft V4S.0 ati.s «v*.x t»^': sr^wvijft T»'..S S t 4V.S ‘ >.ss a ’ i;-s.v> . tas.a ; i c .i s ’fttV’i . :i * *V V AW i ! O^S : vav& j : as>.s. an -v. [35ft.i V V t *-S ■ ASft S ; w*.s %'. 's ’ j Vkt.Vk tvs > ii-H s»*4 as*.« *1- V ' AtV.S | %\t .< t , tiS.tt K j Mi-1 tv AW.S f ftfta.S. ’ tliS.tt ' tss.,4 ' as* s fts».v j VU » V V f VtV.S ’ s*s a v'si’.t a^i.s n*.? a^^^xtv. AVA.tV J tftft.OS ■1 v» .•Sl.V , iw i ,' ' a -. :. . v Vi.. a V ■ >s£4 TV sv* n A '.ft* :.,u f tavn ut.» , AV ft. . i :»fc t v ' ■V.' f , s*s » t Vt/<» . tli.ft : eifti . an.s ft.*M j ■ '' is> -? rv ► At^sv | I >;v A ** tl wi.%, ViS ^ <■>«.* »<# *v^s.t l*SA ftU.H ftVi.S . . y**.s vs ' \ IV* M IffftVtt ti-S.V* t M., 1 S4S •. | ata.s | at* v ;*n «'; > t tW >1 | Itt M VCSY ' isa. 1 } : *ss.« i*a • 1 ats.i v*4 I V s «• 4tW\ i Vt+M "0 IV . 44S.1 iVSwtl : ttsA I HS.6- f esM J aw •» ■ii* f, asvi » ***.». ; att.Vk .•»s.« ! *n..o . V>a ftaft.s atv.% a**.« is) •vw!.*t at^.Vl tf*v ■* TSS.O t+s.a ; i asa.H , at* ,s >n.« asv C^.M j : iiil avvo a f A.s va* .« y f\ taa.t Svft.Vi tVs.» aic.'i QVS(V aia.s asv.^ tn.'i ' Wft.t .SiS.ftl ^vs.t> i \IS.kj a.v.s.^ tv*.? ftVV.S, >:»•*.» 1 l^su V .■ ftftVftl M * 0\ • j cts.v. a ft <■*•:. a a)*/d an.* t-U.S as*.! K - » i M.4.V. vtv t *»s 'a iVS.ft" a*a.^ tJft.i V 1 *n r Mft It ■A> & 'V, »ss.c ; sxs** , ftSft.S 1 • ' *. : : tftft.ft' *44.| f sn.vs : 1 V TS ft XtV V VV4.» | •. .'ia ik ,\“ "*, ll ’♦.■ i * « ■»* e *•■ \ ~ \ Dl I Transactions American Society of Mechanical Engineers, Vol. 27. Reid T. Stewart. r SHOWING THE INFLUENCE OF LENGTH OF TUBE ON THE COLLAPSING PRESSURE. J<# „ mMrkt kn akkkrk , s/ ,.„ PRINCIPAL RESULTS OF COLLAPSING TESTS ON SERIES 1 l for lengths of 2t to 20 feet, between end connections tending to hold the tube to a circular N- N.t eoZred”" NATIONAL TUBE CO’S. LAP-WELDED BESSEMER STEEL TUBES ( form. For an outside diameter of Oi inches and thicknesses from O.IBO to 0 322 inches. CONDUCTED BY PROP. R.T. STEWART, 1902-4. F.P.K.J905. Test Number Outside Diameter Thickness »f Wall Length of Tube * Feet Weight of Tube Collapsing Pressure Collapsed Portion Physical Properties Chemico / lno/ysis % Material Remarhi Commercial Designation of Tube os Reported. Nominal Average M cZ , napxt f Nominal Average A* Pla Coll cm of pse /fa Sf. Inch Gege Amta of Length End Angular Distance from Weld Strength llt.parSf In Yield Point Peunde per^ Hah fatten fteieethm of Area 7- 5. >w PAS Hang darken Oxide Graatasf toast Croatia* t least Reported Nominal //dual lUpeaSt*. In Feet In Oie's. 6 0 9 673 8.685 8.640 0.276 0.290 0.265 7 750 7.576 24.75 1850 44 y 0 " 7 0 y y t- 5* 58 510 36 730 23.42 53.20 .075 /( .41 .08 6 / 8.668 t 685 9.62 5 0.265 0.290 0.256 10.0/0 7.656 23.75 1150 1.6 2‘ 10 “ 3.7 9' - ,4t * S8 980 37 980 23.34 54.60 .075 .36 .07 62 9 625 8 677 8.685 8.655 0271 0.281 0.261 10' 0" 9.970 7 636 24.38 24.05 1628 c 2.7 4‘ 6 " 63 2' 52 800 35 220 ! 3.9/ 34 50 35 08 Bessemer Oi' Casing 2430 Lbs 6 3 8.662 8 675 8.660 0.26 7 0.29 5 0.259 7.795 7 64/ 23.74 158 5 3.7 T 0 77 S' 0 " 0 • 58 720 37 7 50 22.30 SI 10 — .080 .1) J7 .075 — Steal — 64 8.657 8.675 2.6/5 0.260 0.283 0.234 10.020 7.666 23.28 1450 1 7 5 ' 6 0 7.7 S' 6" -no 9 56 67 0 34 130 26.96 61 50 — .065 .li .42 .0 75 — — fiveragp t.661 8.681 8.637 0.267 0.288 0.255 9.773 7 639 23.75 1533 2.7 5' 0 " 6 9 57 136 36 362 2/ 99 5 0.98 082 n .38 .076 65 8.656 8.665 8 635 0.277 0.325 0.260 4 980 4 626 24.80 1750 3.7 S' 0 - 7.0 2' 7" 4 10 * .064 19 .37 .075 — — 6 6 8.653 8.675 8.645 0.262 0.280 0.260 4.995 4.641 23.97 1520 A 7 s' 7.0 3' 4 so * .089 .41 .08 67 8.625 8.6 SI 8.665 2.615 0.271 0.269 0.300 0.250 S' 0 " 4.995 4 641 24.38 24.07 1640 c 5.2 5‘ 7.0 2 ' 7" 4100 * .083 .35 .075 Bessemer 8*' Casing 24 38 Lbs. 68 8.648 8.655 8.615 0.257 0.277 0.233 4.992 4.638 22.77 1345 6 5' 0" 7.0 2' 7“ - 35* — — — — — .077 ./2 .37 .075 — Steel — 6 7 8.670 8 675 2.665 0.26 8 0.274 0.255 5.005 4.651 24.02 1725 4 2 5' o' 7.0 2' 6 * -135* — — — — — .122 13 .30 .07 — — 8.656 8.671 8.635 0.268 0.277 0.252 4 793 4.637 23.77 1636 3.7 5' o - 7 0 087 /» .36 .075 7 0 8 673 8 670 8.625 0.253 0.266 0 221 2.501 2.147 22.72 1475 3.0 2 • y 3 5 O' ,y -120- .077 .10 .35 07 R 7 / 8.650 8.655 8.625 0 27/ 0.307 0.267 2.500 2.146 24.20 1730 4 3 2 ‘ 3.5 - ,5* .082 .41 .08 7 2 8.625 8.662 8.665 8.645 0.271 0.273 0.317 0.258 2' 6" 2.480 2.026 24.39 24.47 1880 c 2.7 2‘ 3.5 2 • 079 .27 .075 Bessemer si'c.t.no east 10 3. 73 8.680 8.675 9 630 0.267 0.312 0.252 2.470 2.136 24.15 1850 4.8 2‘ y 3.5 , • 2" 4 / 3 * — — — — — .077 ■ 1 1 .36 .075 — Steel A 7 4 8 644 8.620 9.620 0.274 0.273 0.250 2.500 2.146 24.50 1785 2.6 2' 6” 35 1 ' 3" - 15* — — — — — .094 .10 .31 .07 — R Average 8.662 8.665 8.629 0.268 0.27S 0.250 2.474 2.120 24.01 1784 3.5 2 " 6" 3.5 082 JO .34 .074 7 5 8.640 8.675 8.605 0.274 0.282 0.260 20‘ 0" 20.003 17.647 24.44 1375 4 1 y i 8.7 14' 2 ' 4 6 8* 55 680 32 880 24.54 6/00 010 .061 ID .35 .08 — Bess Steel R 7 6 8.663 8.685 8.655 0.272 0.272 0.271 20' 0“ 17.788 17. 634 24.38 1410 3.4 — 14' II" 4 40* 59 290 35 5 30 1 8.63 58.00 — .065 id .40 .09 — — 77 8.625 8.660 8.66S 8.605 0.281 0.274 0.3/1 0.262 I9'I0“ 19.855 19.501 25.00 24.47 1275 c 7 J 7' 7“ 10.9 7 5’ - 120 * 43 7 40 25 240 15.21 28.30 — .017 .14/ Tract Traca 1.68 West Iren A 8' Line Pipe 25.00 Us 7 8 8.660 8.700 2.575 0.280 0.294 0.274 18' 7’ /8 750 18.376 25.08 1250 7.5 7' 0" 7.7 7' 3' 4 92* 45 5 10 29 520 13.21 25.30 0 54 .038 .1 10 .1 1 .06 5 2.58 A 77 8.664 8.670 8.650 0.266 0.280 0.246 19' 4" IS. 340 17 78 6 23.84 1425 8.5 6' 6" 7.0 IS' 3' -135* 46 430 26 710 15.92 26 80 062 .040 .119 .11 .065 1.72 A,N to 8.656 8.675 8.585 0.266 0.301 0.258 15.000 14.646 23.85 1385 2.7 f 6* 70 10' 4 - -100* 56 310 34 860 23.67 59.50 006 .064 .106 .33 075 — — t l 8.666 8.735 8.575 0.276 0.286 0.270 14.775 14.641 24.73 1305 3.8 2". 0* 58 500 34 5 90 23.50 61.20 .061 .no 41 .08 82 8 625 8 662 8.705 8.625 0.281 0.271 0.276 0.256 /S' 0 " 14.770 14.636 25.00 24.27 1470 c 4.5 y 3" 9 7 4 ' 8" 0* 58 6/0 38 720 21.42 57 80 — .057 .105 .35 .08 — Bessemer R,N 8 line Pipe 26-oe Lbs 123 8.665 8.675 8.650 0.277 0.305 0.287 14.775 14.64/ 26.47 1675 y 6" 90 4' 2“ 4 35* 58 130 34 560 20.84 58.90 — .090 .105 31 .07 — Steel — 124 8.644 8.675 8.5 75 0.263 0.277 0.253 14.775 14.641 23.5 6 1250 3.5 6‘ J' 87 3' 9 ' 4 ,5* 59 710 35 300 23.76 56 10 .005 007 .121 34 07 — Average. 8.657 8.705 8.606 I.Z75 0.287 0.265 14.775 14.641 24.58 ,42 1 4.0 b ' S" 87 5 8 252 35 646 22 68 58 70 .072 109 35 075 8 3 8.667 8.725 8.595 0.270 0.305 0.263 10.0/2 7.659 24.20 1440 3 5 y 0" 8.4 7 2“ 4 8* 58 510 j 34 710 24 79 58 20 058 no .38 .08 — — 8 4 8.675 8 710 2.670 0.275 0.27 1 0.247 10.002 7.648 24.72 not y b" 7.0 5 4- 4/52 ' 57 250 36 270 19 6 7 5 8 90 .065 .103 .30 .065 — A.N 8 5 8.625 8.664 8.68 5 8.615 0.28 1 0.260 0.273 0 240 /O' 0" 10.000 7.646 25.00 23 28 1575 c 4.7 6' 3" 8.7 4- S * 58 240 36 400 23.00 59 60 — .070 .112 .31 .08 — Bessemer — 8 "Line Pipe 2500 Lbs. 8 6 8 663 8.675 8 565 0.278 0.286 0.260 7.790 7.636 24.85 1 545 4.8 b ■ 3“ 8.7 b 2‘ —132* 57 340 35 720 21 00 58.40 __ .070 109 .34 .08 - Sre*l 8 7 8 664 8.670 8.615 0.272 0.284 0.268 10.010 7.656 24.40 1645 4.1 6' y 8.7 4 10" 4/65* 58 770 33 910 24 58 58.00 — .071 123 .36 075 — — Antra fa 8.671 8.677 8.6 10 0.271 0.272 0.256 10.003 9.647 24.27 1541 4.7 S' 3 * 8.7 58 422 35 402 22.61 58 62 067 in .34 .076 • 2 J 4 5 6 7 8 9 10 II 12 13 U ,7 !$ If 20 21 22 23 24 2 5 26 27 28 27 30 ■*' 32 33 J4 Fig. 13.— Tabular Statement of Principal Results of Tests, Series 1 . ■V.YZ^.xVV 1V\ r.v. - vv>-? .V\ <-'V\ & , \n 1 W Mmji .an«4> Tii\» c\ ■..■hu'j<. .Vv&YV; .'• <\ >•-. v .V^. 1 T iV ' vt3 oraAHDaM **ak>o 8 .M* f ^oi'toa- ^hT o | s z:\mt. O - y.’x Wtfv *-a ~;->wa -aavi™* *& ‘ : \-.«» Wav ■ . . J -vw/: aW J ' A ^- A - A -' l ' i - •J' \tAS»VU»' * V* ft »W W*# .r.-,-. ■ - \\> ^v; ^ ; ■ •- ' ‘\ j \fcWHV s A> iv j .* ' ■ - ; i . Ax rivH> • \\,: 'X ><► e*a."\A>\*.T \o >*Vft \fc V.>-**V •},. V v AtVj^ -I U*Y; ._ • - **%* .V.a.- i -■ ■> »-S*'.\'"‘ '"' v . ; \,v,^vw v . ... . .■-<*<■' '■■■• •■ •'■ ' " \ifoW ’A/v< . ■»»■>!' ftV*V S\ vv '\kS'X\ •t\ b/- • '"♦r b. * i ; is tv fk« *»» "• , ' n •ft!! | .:••'] a , ' ftftVft Mri.V i 7ZFT " ftl\ ft * ! ■A. ft i > Vi | ’X ;• . t '•• t “V, -. , ft ‘a; * a * su f4 ,.'4 Kt' 14 . • t r.v :fts.i %»\ Jift . ft 4 »>f i. * ' \ fttrt .w— ~-4 *-* • ? - - • rjr J Mft * • c*** VMlVMlft -.i'8 Transactions American Society of Mechanical Engineers, Vol. 27. SHOWING THE INFLUENCE OF OUTSIDE DIAMETER AND THICKNESS OF WALL .SERIES 2 j ON COLLAPSING PRESSURE, for lengths of 20 feet, between end connect, one \dmg to hold the tube to a circu/ar form. Reid T. Stewart. PRINCIPAL RESULTS OF COLLAPSING TESTS ON NATIONAL TUBE CO’S. LAP-WELDEO BESSEMER STEEL TUBES CONDUCTED BY PROF. P T S7£WflRT, 1 902-4 F.PK.,UOS. Test Humber Outside ter Thickness of Wall ches Length of Tube y Feet Weight of Tube Lbs. per Foot Collapsing Pressure Collapsed Portion Physical Properties Chemical Analysis % Material He: varts Commercial Designation of Tube as Reported Nominal At Place of Vsmsial Average At Place of Collapse Reported 7 / / Pounds per Sq. Inch Cage Used Pate of Length Oistance End Strength Lbs ptr Sq In Pounds ptr Square Inch Flongatitn T*"h 3 fioduetton at Area % 5, l.con Sulphur Phos. Mang Carbon Oeida Greatest Leost Greatest Least Nominal b\ ptr ^ In Feet In Oia’s. 200 L ,„ 4 030 C 123 0135 0 104 10 11 m- 7.77 450 2 3 ' o - 6 2' 5i’ / $o* *5# 35 0 *37 470 9 11 21 *54.10 047 114 .31 .00 b 4024 4 OteO 5.110 0/27 0 175 0.094 20 1 7 'of 1.0 0 5 00 2 2' 9" 5.5 S' 4" +150* * 54 220 * 37 430 9 17.25 *52.40 — .040 .1 10 .50 .00 ft b 4 000 4.015 4 030 5 130 0 134 0.130 0 135 0.103 20' 0“ 20 11 <0l 8 24 8 11 575 B 2 3 * 4" 7 2' 4 ' -150° *50 330 * 37 430 *17.71 *5 7.80 — .044 .105 .44 .07 — Oositmer ' b 4' Converse Joint 8.24 Los 20 3 4 on 4 020 5.140 0. 130 0,1 37 0.1/0 20 11 iar 8. 1 1 540 2 3 * 0" 4 2' O' -ISO 9 *58 5 30 *44 140 9 1 5.38 *41.40 — .059 .107 .40 .075 ... ■ Srtti b 204 4 010 4.020 5 140 0.131 0.147 0 1/4 20 If tay 8.24 530 i 3 ' 4“ 7 7' 10 " 0° *58 110 *37 500 * 18.21 * 4/43 — .052 1 10 .41 .085 — A.b 4.0/7 4 029 5 144 0.121 c.ut 0.104 20 inal- S.OI 511 i.s 3 ' Z" 4.3 5 8 0 84 31 47 8 17.58 55.2/ .041 101 .44 .0 79 105 4023 4.020 5 140 0.131 out 0.101 20 11 /oy 8 22 530 2 3' 2“ 4 3 t 1 - 80 * *51 220 *44 3 SO *11.00 *50.40 .048 112 37 C 7 6 201 4.021 4 00 0 S 140 0.121 0.133 0.104 20 11 / oi * 8.12 4-80 / 3‘ 4" 7 3' 0 180 * *41 480 *41 7 80 9 14.17 * 50.20 — .057 .074 45 .085 — a 207 t, 000 5110 5 140 0.134 0 134 0.178 20' 0 * 8.24 J 45 440 6 2 0" 8 3' 0 - 10° *41 130 *43 420 *1314 *54 10 .040 ll 1 .41 .075 Bessemer 4' Converse Joint 8 2b ids. 209 6 013 4 010 5.120 0.135 0.134 0.1 10 20 If toy 9.45 510 2 2‘ 1 “ 5.5 2' / 1- 10° *51 710 9 40 710 *11.58 * SI 13 — .041 .10 7 .45 .08 — Site/ b 207 4 on 5.710 5.910 0.128 0. 142 0.1 !0 20 if •oy 8.07 485 1 3' 3" 4 5 7' 0 180 9 *54 350 *39 250 9 I4.04 *55.00 — .041 .103 .47 .09 — b Average ~4 017 4.002 S 13 4 0.131 0.147 a. io? 20 irni ■ 8 24 521 it ,■ 4" 47 51 5 78 41 942 ia.ts SZ-33 043 .10 5 AS .171 210 4.007 4 020 5.170 0.147 0.178 0/47 20 n •*r 10.44 1025 a 3 4' /- 8.2 2' 4" - ISO 9 *58 410 *44 280 * 15.75 *53.70 074 105 09 211 4.024 4.070 5 .170 0 147 0.218 0.141 20 n •oy 10.51 130 8 5 J ' f 7 18' 0 " -H35 9 *42 580 *45 010 9 1 i 54 *53.40 — 071 .49 .08 — t 212 b 000 i 4.0 1 1 4.030 5.190 0 154 0.144 0.174 o./so 150 B / 4 ' 2 * 8.3 5' 0 " *41 000 * 4b 270 9 18.77 *55.80 — 080 .112 .34 .09 — 6 SfC.„n, IbAtlU 213 4.037 4.050 5.770 0 .144 0 178 0.115 20 n 10 30 850 B 2 4' 0" 9 2' 4 " -I40 9 *44 5 30 *47 430 *14.54 *48 20 — 071 .III .075 — Srtti b 214 4 0)1 4.030 5.100 0.171 0.177 0.144 20 n nr 1C. 4 1 1010 c 2 5 ‘ 0 " 10 2’ * ' - 10 9 *57 140 *42 320 *15.25 *54.80 — .077 .109 .33 .07 — 6 Hu *ra 9 * 4.022 4.040 5 .179 0.147 o m 0.144 20 IT‘ IIS' 10.4 7 141 2.4 4- 2" 8.3 41 112 as in 15.57 S3 42 .07? .109 .Stt Cl, 21 5 4.033 4 0/0 5.140 0 14$ on 4 0.124 20 11 n * 10 55 1010 2 3' i ' 7 2' t " 4- 45 9 *42 4/0 9 44 370 *14.04 *55.20 — 081 .111 39 .07 — 210 4.02$ 4.050 5.110 0.155 0.222 0.127 20 If toy 1 75 740 2 4 ' 0“ 8 2' 4" - 40* *58 440 *43 b40 *13.72 *5 2.20 — .072 39 07 — £ 2 1 ? 4.000 4.03$ 4.070 5.110 0 154 0./7S 0.187 0143 20' 0" 20 11 Hi" 10.44 10.72 10/0 B 3 4 ' 0 " 8 2' 4 " -t IS 9 *41 310 *44 0/0 9 IS. 38 *53 80 — .0 70 .39 .07 — Bessemer 6 5f tol.ng If At to* 218 4.00$ 4.030 5 .150 0 148 0 177 0.130 20 ll / 12" 10.50 840 2 4‘ 0" $ 3 ' 4" 4- 10 9 *42 840 *44 400 *15.12 *5 0 . 70 — ■ 084 .112 .00 — Steel t, 217 4.021 4.040 5.190 0 /44 \ 0.191 0.131 20 11 • oi" 1031 180 2 4 ' 0 " 9 2' 0 " 4- I0 9 *44 050 *48 200 *15.44 *41.10 — .071 .107 ■ Jb .07 — b Flvrrag* 4 024 4 040 5.174 0 144 \ 0.174 0.131 20 11 'll " 10.42 124 2.2 3' ir 7.8 41 914 AS 328 15 34 52 34 077 .101 .374 0 72 221 4 034 4 080 5 110 0 It t 0.141 0.114 n toy 10 40 740 E 2 3 . 4“ y 2‘ 0 * - ISO 9 *44 320 *47 310 *1425 *54 20 _ .084 III 31 a? _ h 222 4 000 4 037 4 010 5 100 0 154 0.173 0.130 20' 0 - 20 /oy 10 44 f.7t 980 B 2 4' 0 * 1 2 2“ o 9 *57 470 *42 540 * 12.25 *44.40 .077 .107 40 S) ’ Co., ny / 0 At Lhl- 224 4.024 4 040 5 190 l.ltt 0 141 0.141 20 lf toy IC.40 I//0 C 2 5 * °" It 4 3" 0 9 *43 000 *44 440 *18.25 *52 50 — .074 .,01 35 aj — Srtti 6 Average 4.032 4 070 5.757 0 143 0.170 0.131 20' Oi 11 /oi" 1020 117 2 4' 2” 83 41 443 45 537 14.12 5703 .078 .101 .30 .07 223 4.024 4.040 4.000 0.141 0 171 0.152 20 11 r • 10.54 1070 c 5 y o - i 4' 0" 0 • *57 710 *43 450 *14.50 *52.80 085 104 .35 .005 — a 225 4 039 4.050 5.170 0 144 0.1 10 0.1 17 20 11 il- 10.31 7/5 B 3 3 ' 0 " 4 17' 8 ‘ - ISO 9 *57 170 • 3$ no *11.71 *54.70 — 044 018 .42 .08 — a 224 4.000 4 029 4 100 5.170 0 154 0.14 $ 0.137 20' 0" 20 1 9 sf 10.44 ; II .14 1425 C 4 w 5 2 ' 2 ' - 20 • *43 440 *41 450 *13.84 *53.40 .075 .104 .40 075 Btsotmor 5 J * Caving / a At LA,. 239 4.037 5 110 0.1 47 0 190 0.1 to 20 n if 10.45 7 75 B 3 3' 4 ’ 7 2 ' I " 180 9 *57 040 *42 230 *17.38 *58.20 — .042 .084 47 .00 — Srtti a 211 4 034 4.070 5 170 0149 0.180 0.133 20 • * •f 10.50 1050 c 5 2' 4 " 5 IS' 3" 0° *42 150 *45 270 *15.43 *50.40 — 085 .HO .39 .075 — • * w 4.044 5 1$4 ana a.nf ana 20 it- . r 10.40 1007 4.2 2' it’ 58 40 144 AS 7 At n, at 54.34 .075 ,iai Al ■ 071 Fig. 34.— Tabular Statement of Principal Results of Tests, Series 2. .72 .jqV .gHaaviio^a jadiv^' vaM *o rraiDoa -Aoma?^ av? 6 noAWi/.JiT S ?, 3 W\ 3 k£ ?maonw W WTlw .'.a va sgn^.sv* \»i« k A . ! ' ■' ■ * » uo\ < V\^*' tyv* > V.v^ Va yt\ v ^w.n l4V A i VVft **\ u Utv-' <-,,. ' tvva , V • • , VSV-* ‘e-s : wM tv£v-?. : i V^£. - UJM. r«S vif... 1 V !> ; i i ' - i 's U ) SVo.. t A* ft* ft A S' ft -V . ftftiVA VIS ■ vV-. ..^ . *>* x , ; ...«V f>, KT Transactions American Society op Mechanical Engineers, Vol. 27, SERIES 2 SHOWING THE INFLUENCE OF OUTSIDE DIAMETER AND THICKNESS OF WALL ON COLLAPSING PRESSURE, for lengths of 20 feet, between end connections tending to hold the tube to a circular form. r»r £ tonparion Red , Reid T. Stewart. PRINCIPAL RESULTS OF COLLAPSING TESTS ON | NATIONAL TUBE CO'S! LAP-WELDED BESSEMER STEEL TUBES it-Stewart, 1902-4 FPK,i9oy CONDUCTED L Outside Diameter Thicknes of* Wall L ength_ of * Tube. Weight of Tube Lbs. per Foot Collapsing Pressure Collapsed Portion Physical Properties Chemical Analysis Material Remarks Commercial Designation of Tube as Reported Number Nominal Average At Place of Collapse Nominal Roerage At PU Colt opsV As Pounds Sg./nch Rate of Length 'Zo Angular Distance^ Lbs per Sq In. Yield Point Pounds per Square Inch Inches Reduction 5,7,t#n Sulphur Hang. Outdo Greatest Least Greatest Laa.t Nominal Actual lbs per Sec In Feet In Dio’s. 4 035 4.050 5.780 0 177 0.174 a.ui 20' 0i“ lti ■■ 11.17 1075 C 3 4 ‘ g z' r - 70 ' *40 830 *42 570 ' 14.57 *53.70 .070 .105 .33 075 A 6 4 021 4.075 5.770 0 217 0.240 0.17 8 20' Oi" toy 13 . 57 1400 c 1 4" 2' 4" *4/ 430 * 47 420 *18.25 * 55.00 .048 .070 ■36 .07 5 * t 242 4.035 4.040 5.770 0.220 0 I8S 0.177 0158 20' 0" Zf eh" ioi * 14.20 / 1.57 1272 C 3 0" 10 3' 0“ *57 170 *44 200 *10.76 *57.70 .067 107 .32 .065 Bessemer Si' Conn, to 20 Lb 3 24 3 4.034 4.080 5.770 0 170 0.170 0.144 20' Oi" 17 ioi' 10.42 875 0 2 3‘ 4’ 7 2' 0“ + 70 9 *41 240 *44 340 *14.46 *56.80 — .076 .106 .39 .075 — Steel 6 244 5.788 4.020 5.740 0.173 0.253 0.173 20‘ Oi" 17 ioi ' 11.77 1750 c 2 5' 0" 10 17' 4" o' *41 770 *41 7 10 *14.72 -48.80 — .077 .101 .37 <>■ ‘ 6 ft ¥9 rage 4 023 4.053 5 774 0187 0.215 0147 20' oi" ir' ioi" 11.78 1318 2 2 4' 2" 8.4 40 732 44 854 15.04 S4.44o .069 .079 .36 .072 245 4.040 5 .780 0.173 0.252 0.174 20' OK" 17 7 • 11.77 1375 c 5 4' 0 " 8 17' 0" - 20 * *41 240 *40 870 *15.58 *55.70 — .080 .103 40 .075 — a 244 4-010 5.930 0.230 0.252 0,197 2 0’ Oh." 17 tT 14.17 1700 c 5 3' 4’ 7 2' I" + 20' *45 570 *49 750 *16.48 *51.40 — .081 .103 .40 .07 a 247 0.000 4.05 5 5.770 0.220 0.1 77 0.200 0.153 20' 0" 20' Ok" ti- 14.20 1010 B 5 3' O' *42 170 *44 470 *13.34 *54.10 .075 Bessemer Si C..„, ,4 20 Lb. 248 b.004 4.040 5.870 0.232 0.243 0.2/0 20' 0 ' 17 ll' 14.30 1800 C 5 4' 4" 7 2' 3" ~ 20' *47 440 *47 ISO *13.13 * 50.60 — 086 '-107 .40 .08 — a 249 b.034 4.050 5.780 0.228 0.237 0.172 20' 0 ' 17 > 1 ■ 14.15 1200 c 5 4' 4" 7 17' 0“ -f IS' *45 850 *48 140 *16.67 *5180 — »74 112 .35 075 — A. a Average b 021 4.050 5.750 0.212 0.237 0.184 zo'oi' If >3" 13.16 1457 5 3 ' II" 7.8 44 510 46 724 15.04 ' 52 76 085 .106 39 0 75 250 4.014 4.030 5.970 0.172 0.201 0.174 20' 0 “ If si- 1 1.74 1450 c 2 S' 0" 10 3' 6" - 5* *42 370 * 46 000 *16.34 *5 5.27 — 096 102 .41 n — 4p 251 4.014 b.OSO 5.780 0.2/4 0.233 0.173 20' Oi" si- 13.25 1750 c 5 S' 10 - io' *44 070 *50 000 *14.25 *52.20 077 .104 .075 252 4 000 4 020 4.050 5.940 0.203 0.222 0.235 0.204 20' 0" 20' 0 " Si' 12.04 13.73 2075 c 5 4 0" 12 3 ' 8" * 45 330 *46 810 *13.58 * 47.70 .072 too .25 .07 A, b P 5l-C..tn, 12 .04 Lb, 25 3 4 020 b.040 5.750 0.184 0.248 0.144 20' oi" 17 si' 1 1.45 7 SO B 2 4' 0" 8 7' 7“ O' ' 61 030 ' 47 240 ' 13.50 *56.00 — .054 .085 .40 .08 — bp 254 6.005 5.750 0.220 0.251 0.201 20' oi" 19 si' 13.57 1550 C 2 S' o “ 10 15' 4" o' * 41 870 * 47550 *14.27 •50.40 — .083 .113 .40 .07 — bp Average 4 015 6.052 5.744 0.204 0.234 0.187 zo'o!.' 17 5n 12.80 155 5 3.2 S' 0 * 10 42 744 47 520 14.37 52.75 .083 .101 37 .075 255 4.028 4 050 5.770 0.172 0.208 0.170 zo'o “ ti- II 78 770 0 2 3' C’ 7 2' O' - 45 • 9 41 410 *42 720 *14.72 *56.10 — .075 .107 .38 .07 — a 25b 4.021 b.OSO 5 .770 0.185 0.175 0.147 20' oi" ll' II 55 1450 c 5 4' 0" 8 2' 0“ - 20' *45 S30 *48 270 *15.83 *56.70 — .086 .107 .45 .075 — a 257 4.025 4.040 5 .740 0.203 0.187 0.210 0.145 20' 0" 20 ' Oi" 17 ti- 12.04 1 177 1250 c 3 0" 4 IS' 7 " 180' *43 740 *46 440 *14.21 *55.20 .051 .090 .36 .07 Bessemer St Count It 04 Lb. 258 4.013 4.020 5.950 0182 0.20 7 0.150 ZO'O " 19 ll' 1350 c S 4' 0 " 8 14' 10“ + 10' *57 730 *44 070 *13.88 * 50.40 — .085 .III .35 .075 — Steel ■3 257 4.024 6.060 S.760 0.182 0.217 0.140 20' 0 • 17 11- 1 1.35 1100 c 3 ' o' 4 3' 0" 4- 30' *45 540 9 45 470 *15.57 *52.50 — .072 .107 .40 .08 5 — a Average 4.022 4.043 5 770 0.184 0.208 0.157 zo' oi" if «j- 11.57 1188 3 -7 3' 4" 7 43 274 45 442 14.87 54.22 .074 .103 .39 .075 210 4.054 4.180 5.720 0.240 0.284 0.242 20' Oi" 17 4i' 14.06 1755 2 4 6’ 7 17' 6' - 10' * 40 840 *41 650 *17.13 *55.70 — .078 .112 35 .075 — A. bp 2b 1 4.027 5.770 0.251 0.274 0.218 20' Oi" 17 Si- 15.47 1750 2 4' 12.5 10' 3' - 10' 57 3 40 41 400 13.38 57.7 0 .088 .33 .07 A , bp 2b2 4.033 — 0271 0.243 — — 20' 0" ZO'O * 19 Si’ 14.70 14.20 — c — \ — — 57 150 37 480 22.72 57.50 .071 .34 r ot Bessemer A .bp 5j - Cos.n, lb.70Lb, 2b 3 b.025 4.040 5 740 0.280 0.300 0.245 20' oi" 17 54’ 17.17 2400 s S' o' to 16’ 7 m - 5* 58 830 37 140 17.05 S4.70 — .074 .103 .47 .075 — Steel 214 b.022 4.080 4.000 0.240 0.272 0.244 20' oi" 17 si* 15.77 2450 5 4' 0 " 12 17' 0 * O' 58 050 36 810 13.82 60.50 — 085 .104 37 07 — bp Average b 032 4.070 5.748 0,243 0.289 0.242 20 oi" If si" 2137 3.5 5' 5" 10 7 .097 .104 .38 .074 273 4 021 4 050 5.740 0.270 0 287 0.240 20’ oi" 1 ti- 14.57 2270 2 3' 8" 7.3 18’ O' 0- *43 810 *47*510 '18.33 *5 8.40 — .090 .101 .38 .07 — 27b 6 033 0.270 0.224 20' oi" ll- 24 75 5 4 * 0’ 8 2' 4' o' '45 830 9 47 570 '14.29 *5 4.00 .07 7 .10 7 42 277 6.072 4.100 4.000 0.271 0.254 0.304 0.217 20' 0" 20' oi’ ii- 14. 70 15.77 2340 c 5 3' 7.5 . 18' 0 " o' 57 NO 37 120 17.29 58.30 .072 .103 .37 .07 Bessemer Sl C..;, lb 70 lb. 282 b 024 4 040 5.770 0.247 0.288 0.230 20' oi’ ti- 14.44 2250 5 / ' 10“ 3.7 o’ ir o' *57 080 *34 030 '22.17 *58.70 — .092 .106 .52 .085 — Steel A. a 285 4 022 4.050 5.780 0.270 0.322 0 222 ZO'O " 17 >r 14.55 2550 5 S' O'" to 3' 3" 4 20' *43 440 *42 ISO * 16.33 *54.80 — ■ 075 HO .34 075 — * A v era fa 4 034 5.782 0.244 0.277 0.227 20 0i~ I* li.Zi 2381 4.4 3' 8" 7.3 .099 .109 .*> .076 Fig. 35. — Tabular Statement or Principal Results op Tests, Series 2. .?£ .joV ,8aaawo>ia ,wv»\ i*W»~w ‘o'V ••«•, Transactions American Society of Mechanical Engineers, Vol. 27. SHOWING THE. INFLUENCE OF OUTSIDE DIAMETER ANO THICKNESS OF WALL SERIES 2 l ON COLLAPSING PRESSURE, for lengths of to foot, between end connection* ending to ho/d the tube to a circu/ar form Reid T. Stewart. PRINCIPAL RESULTS OF COLLAPSING TESTS ON NATIONAL TUBE CO’S. LAP-WELDED BESSEMER STEEL TUBES COHOUCTCD or FOOF tt.T. STCWART. 1*020 FFK.ltbi. Test Number Outside Diameter Thick ne of Wall has Length of Tube y Feet Weight of Tube Lbs. per Feet. Collapsing Pressure Collapsed Portion Physical Properties Chemical Analysis y. Material Pemerhs Commercial Designation of Tube as A e ported. Nominal Average At P! a coin Greatest pee Least Nominal Average At Place of Collapse As Reported Actual Sq. Inch Gage Used n.i..t L ength End Tan si /a Strength Lbs. per S g fn Squafe f?ch elc %h!r Inches vl Silicon Sulphur Hong Carbon <5 Greatest Least AcXvai Lbs par Sac In Feet India's. J 00 4.457 4.475 4 SOS 0 157 0.125 01 52 20 ok' lV7f 10.87 7 10 3 2' 0" 3.4 2" 0 * * to 9 55 540 34 3 10 21.4 7 57 20 047 10 7 .38 .07 SOI 4.457 4.420 4.595 0 145 0 182 0.145 20 08 !9‘ 9s' 1 / .47 7 20 3 2’ 4.8 57 730 34 530 21.43 5 6 20 .065 .104 .39 075 a so? 4.025 4.4S3 4.705 4.570 0.172 0/42 0.129 0 153 20' 0 " 20 / 9 ’ 9t: 1 1.42 a 3 2' 3 4 58 340 37 890 21.82 60 20 .059 HO .38 .08 Bessemer 3 03 4 453 4.425 4.575 0 145 0.205 0 145 20 17- IK- 1 1.35 5 3 ' 5.4 —! /O’ 57 no 34 330 23.04 5 9.70 .060 .090 .39 .07 Stoat 304 4.452 4.425 4.400 0.144 0.214 0.123 20 Or*’ II- *4- 1 1.42 400 3' 0" 5.4 2 ‘ 180 * 58 310 37 520 22.59 56.60 — .068 .108 .38 .075 — 4 Or.ro,. 4 454 4.424 l.Sl* •■it* 0.194 0.144 20 ei’ 19- rr 11.35 Lit 3.4 2' 4" *.L 57 4/f 34 5 14 tz.lt 58.02 .014 104 SI .074 305 4.420 , 4.725 4.405 0.194 0.220 0.120 20 Os' nil' 13.57 i too J 2’ 0 * 3.4 2' # o * 59 790 34 200 23.21 57. !0 — .048 112 .37 .09 m 3 01 4.493 4.700 4.590 0.200 0.207 0 194 20 0 " I9‘ 9f 13.85 1205 4 3' 0“ 5.4 2' 4’ O' 57 770 34 030 If. 84 42.30 — 051 ■ 108 •Jt .065 — a 307 4.125 4.705 4.545 0.203 0.202 0 227 0.124 20' 0 ' 20 or If 8s’ 13.32 13.99 1275 c 5 2' 4 " 4.5 4 10* 57 070 33 720 22.42 60. SO .054 .075 Bessemer 6k' Casing 13. 32 Lbs 3 ot 4 427 4.47 0 4.575 0.205 0.225 0.200 20 or I9‘ 14.17 1245 S 3 ' 0 - 5.4 17' 4" 0 * 57 100 34 090 24.42 61.70 — 064 .102 .37 075 — Steal 309 4 474 4.475 4.570 0.194 — — 20 0 " /9‘7hr 1 3.57 1075 * * 0” 5.4 7 ‘ 57 490 35 970 23.42 57.90 — .06/ 078 . Jf .07 — m Mu a rage 4.424 4.495 4.521 0.200 0.220 0 .190 20 ’ Or: /9' 7 ; J : 13.83 ' 184 4.2 2' 8’ 4 9 57 248 35 402 22.46 5 5.94 ■ 060 105 38 .075 310 4 44/ 4.440 4.410 0.240 0.272 0.245 20 Or: Hi «- 17.74 2275 5 3' 0 " 5.4 2' 4” 0* 57 590 35 930 24/3 59.90 .055 too .39 .07 Bess Steal P e ill 4.474 4.455 4.40 5 0.254 0.277 0.225 20 or 17.39 2140 5 4 ' 4* 7 3 2 ' O’ 4 to* 57 49 0 34 4/0 25.33 6 0.80 — .06 7 .1 16- .34 .07 — P.A 4.425 4.449 4.445 4.520 0.232 0.252 0.244 0.224 20' 0" 17.02 17.25 1975 5 3 ' 5 4 4 15* 54 340 32 740 20.42 62.60 .071 106 .37 CIS Li" Cas.no 17 02 Us 313 4.457 4.415 4 555 0.250 — — 20 or /9‘8i” 17.09 1820 5 2' *' 4.5 18' 0 “ 180* 47 /JO 33 220 13.09 23.10 — .0/5 1 9 6 Trace Trace — Wra-t han n a 314 4 471 4 450 4.595 0.251 0.22 3 0.228 20 0 •9 '/or 17.20 2/75 5 3' 8“ 5.4 13' 4" 0* 57 no 34 480 2179 60.40 — .05 4 .100 .37 .07 — Bets. Steel a 3IS 7.044 7.070 7.000 0 158 0.177 0.122 20 or if u- 11.41 570 4 2‘ 4 * 4.3 9‘ 4" 4 15* 57 430 38 510 17.92 54.50 — 053 Of j .34 .07 314 7.040 7.075 4.995 0.158 0.177 0.134 20 or 11‘lf 1 1.58 550 4 2' 0 " 3.4 %’ 4 " 0 * 54 440 35 280 21.1 7 5 9.70 — .057 / 06 .35 07 — a 317 7 000 7.043 7.090 4.990 0180 0.1 54 0.170 0.122 20' 0" 20 II- ti- 12.34 1 1.31 575 e 3 0 " 3.4 IS' 8 * 0* 54 850 36 7 80 16.7 1 61 60 — 056 .095 .075 — Bessemer a L>- Casing 12 34 Lb s. 3 1/ 7 0 33 7 .040 4.920 0.214 0.13 5 ok" ll' If 12.1 3 5 80 3 2 ' 4.0 0* 59 520 41 140 22.42 52.90 .069 .37 .07 Steal 319 7.059 7.075 4.995 0/43 0.217 0.130 20 or a- w 11.94 S 40 5 2 ' 0 " 3.4 14' *" - 120 * 54 070 38 920 19.25 58.30 — .059 .106 .36 .08 — e. 1 044 7.070 L.no 0.140 0.191 0.130 20 or ii- H- 11.72 5*3 3.8 2' 2* 3.7 57 242 38 246 11.41 57.40 .059 .101 ■ St .0 73 320 7.050 7 .025 4 990 0.245 0.257 0.225 20 or ’f ti- 17.74 <775 * 2' 4* 4.0 9' 9 * 0* 59 840 37 220 21.94 51.30 — .072 .110 .35 075 *.* 321 7.050 7.070 4.920 0.241 0.273 0.220 20 0 " ll- H’ 17.50 1 575 4 2' 4 • 4.3 3' 0 * 4 15* 59 440 32 250 213) 56.30 — .065 ■ 101 .39 .07 — - a 322 7.000 *1.057 7.020 4.920 0.242 0.243 0.202 20' 0" 0 " 1’’ it" 17.51 14.9 7 1525 c 5 2' 4.6 S' 0* 57 730 35 500 23.34 57.90 .06 1 .094 .38 07 Basse mar Li ' Casing i7.St Lb. 323 7 045 7.090 4.920 0.245 0.270 0.220 20 or a T 17.77 1475 5 3' o~ 5.1 2' 4’ 180* 40 210 33 910 23.50 56. Lf — .06 1 .102 .37 .0 75 — Steel a 324 7.047 7.020 4.990 0 242 0 241 0.204 20 or 17- • i " 18.02 1850 5 2 ' 4" 4.3 r 4 m 0* 57 720 35 730 24.7/ 57.30 — .060 .100 .37 .07 — *.<* Buarag'e 7 050 7 081 4.984 0.242 0.241 0.2/4 20' ok" ii- tf \480 44 2' T 4.5 59 036 36 134 22.97 55-7 8 .064 .101 .37 ci; 400 4.443 4.750 4.540 0.1 54 0.1 59 0.1 to 20 /t ii' fi- 10.48 *400 *10 *20’ 0’ * 342 /S' i ■ tew* 58 260 44 83 5 13.21 51.10 .061 089 .37 .07 P.e 401 4.452 4.730 0.520 0.157 0.190 0.130 20 or ll' >r 10.87 585 3 4" 8.2 5' 3" 4-150 m 52 78 5 37 565 15.50 53.90 — .058 .090 .33 .07 — c 4.445 4.7 30 4.520 0 ISO 0/52 0.142 0125 20' 0 * li- 10.39 10.57 585 0 3 4' 9“ 8.4 7' 0" 4140* 574/5 37 025 21.00 57.10 — .070 .105 .35 .07 — Bessemer c Specs! Li' OP 10 31 Lbs. 4-441 4.71 0 . 4.570 0154 0.122 0.140 3 0 ck' ii- ti- 1 0.47 540 3 S' /0.0 4‘ 0 " 58 385 38 035 21.72 60.10 .0 75 .IIS .07 Steel 4.457 4.720 4.590 0.153 0.173 0.125 zo Ob’ ll' U- 10.59 520 2 4 ' 4" 8.2 15' 0“ 0* 54 445 35 495 20.46 54.20 — .070 .107 32 0 7 — « ‘ 4.722 4 574 0.1 54 0.173 e.iLi 20 o}~ n-i " IC.tt 543 2.8 a- if 8.8 54 458 39 031 18.42 5 5.28 .on .101 ■ S4 .07 Fig. 36. — Tabular Statement of Principal Results of Tests, Series 2. • ■ ' ' ■ ^ Y> - ■- V'. ■••> ™ ■-■'•■ . ; ■': ■ .. ) ■ .. . ■• •- •** > L-. • .; ■ * j V.1JV ' -•- - — v — — —t* i ;•« ! «'• • , ••• ; *“ i at* * ■ vu * ' ; Transactions American Society of Mechanical Engineers, Vol. 27. SHOWING THE INFLUENCE OF OUTSIDE DIAMETER AND THICKNESS OF WAIL SERIES 2 { ON COLLAPSING PRESSURE, for lengths of TO feet, between end connection s id mg to ho/d the to bo to a circular form. ere/ hetTfoot. She Reid T. Stewart. PRINCIPAL RESULTS OE COLLAPSING TESTS ON NATIONAL TUBE COS. LAP WELOEO BESSEMER STEEL TUBES conouctco or pnof. at stcmpt, 1902- Test Outside O/omeler Thickness of Wall Lengt^of* Tube Weight of Tube Collapsing Pressure Collapsed Portion Physical Properties Chemical Analysis JS mini Average At P/a r , m . Average fit Place of Collapse fis Lbs />« rt... Pounds Gage Rate of Length O.xta me. Tenmile |] Rad yet, an Material Commercial Designation Greatest Greatest Least Reported No mi no / s/toch Used In Feet ere 18‘P~% *• Sqvere inch "ET *fAraa Sulphur H.ny 7' 0 ‘ + 13* 5# 465 38 195 25.33 58 30 — 089 3* .075 — c 6.455 6.48? 0.251 0.217 0.242 io‘ li • ifti" II.3S. 22/4 i > 4- 7" >* 57 <74 35 801 2/72 52.12 05/ m .34 0 7 o' 410 4 484 4 73 0 6 570 0.241 0-257 0.220 20' li' 11’ If 18.51 1880 5 4' 0 * 7.3 10' 0" + 50* 55 180 38 710 2/ 25 S3. 10 — .077 105 32 .07 c 41 1 4 684 6 700 6.630 0.252 0.217 0.217 20’ if 11' if 17.24 1940 2 6 ’ 0" 10.1 8' o'" - 43* 57 J#0 35 465 21 84 58 80 — .074 102 36 .0 75 c 4/2 6 675 0.23# 0.258 0.301 0.224 20' 0" 11' If 17.02 17.5/ 1780 C 3 S' 0 " - 180 * 5# 395 37 375 24.48 51.50 .077 100 .07 Bessemer 8*' Casing 1 7 02 Lbs 413 6.47 1 0.248 0.28 3 0.223 20‘ li’ 11 If 18.71 / 7/0 2 S' 8" 10 0 7* 6" + 45* 58 715 34 #30 20.38 57.30 — .072 .017 .31 .07 Sveel c 414 6 582 6 7 00 6 600 0.248 0 282 0.217 20' if 11' If 17.01 1815 8' o m 10.1 18' 8’ 4 70* 55 775 3# 180 20.25 81 00 — .080 .103 32 .07 — c Average 6 4 81 5 710 4.604 0.241 0.282 0.220 20' li" 11' If 17.07 1745 3.2 5' 4" 9.8 58 841 37 128 21.84 57.14 078 .101 3 3 07 1 4/5 6 072 6 030 0.250 0.277 0.203 20' li" if ti- IS. 53 2220 3 4 6* 1.0 18' o • _ 20 » 55 90S 38 885 11.13 80 10 092 Oil 34 075 c 4/5 6 .042 5 050 5 000 0.282 0.277 0.208 20' If 18.18 2555 2 4' 8" 1.0 13' 0“ 8 75° 57 235 38 255 21 SO 81.70 — 078 108 35 .07 C 4/7 4 .000 5 080 5.000 0 27/ 0 271 0.307 0 222 20' 0" 20 • li" if si" 18.70 2400 c / 12.0 4 7i* 5# 245 39 185 22.93 51 80 .083 104 32 .07 si" Caving IS TO Lbs. 41 1 5 080 6 020 0.2 77 0.307 0.244 20- If • r ni- 17.08 2270 5 * O' 12.0 3 ' 0“ -170* 54 720 38 005 22 3# 58.10 — .095 108 .35 .07 — c 41 f 6 044 6 060 0.27/ 0.313 0.113 20' li" if ti- 18.73 2575 * 5' 0" 10.0 7* 0" ~ 85* 5# 445 37 475 17 08 57 80 — .010 105 .27 — C *»•'«». 6 047 6 076 6 0/2 0.288 0.217 0.2/4 20' If If '/Of 1 0.44 25 22 3 S' 2" !0.4 57 330 37 773 20 ft 51.28 083 .104 J3 .071 1 8 857 — — 0.178 0.17 0 0/70 20.120 11.818 15.12 450 — 8 ' 0" 8.3 II' 0" - 18* 58 4/0 35 470 28 13 57 SO 005 081 101 .35 ot — l f 637 — — 0 111 0 115 0.171 11.918 11 818 17.24 825 — 6 ' 3" 8.7 IS' 8" - 30* 80 410 3# 030 2 1 04 57.20 008 077 .118 .32 075 ■ These Rttt 3 2 425 7 64 / — — 0.1 80 0.188 0 184 0 189 20’ 0' 11.117 11.815 1807 1877 535 5 — S' 6' 7.7 4' 3* ■8120* 80 020 38 420 24 00 58.73 008 077 .31 0 75 — Bass tmar reeled eam li" Casing 10.07 Lbs 4 8 540 — — 0 183 0.11 1 0.171 20.004 11 .702 18.54 450 — S' 8" 7.7 1 5* 3" - 30* 58 840 35 510 2 / 72 514J 008 071 .32 \ 09 — Sraal Senes 1. s 8 538 — — o in 0.117 01 73 20.01 1 17.23 820 — S' 0" 7.0 4 ‘ 7" ■6 /5® 51 180 38 350 23 00 51.50 .010 .087 3 1 .0 75 — Average 8 643 0 185 cm 0.172 20.028 If 74 538 5' 8'' 7.1 SI 344 Jo j to 23 22 58.53 .007 074 no 32 .077 24 2 604 8.510 8-580 0.211 0 230 0.210 11' 2 " 11/88 18 888 11.57 270 e S’ 6* 7 7 14' 10” - 5#* 58 700 34 080 22 77 51 40 008 .04# 105 3# 07 — 27 2 529 8 540 8 510 0.2 33 0 231 0 188 13' 8" 13.87$ 13.373 20 IS 1 l 15 c — s’ 6* 7.7 10' 1 * - f • 57 770 34 730 22.33 53 70 .070 .102 .32 .075 — re*,. n*n 28 8 525 t 457 2.550 8.530 0 221 0 213 0 222 It'll | 12 132 12.830 20 10 <723 250 8 S' 7.7 8' 2* - 67* 80 5 30 37 7 00 / 6.67 57 20 .1 17 .35 08 Bessemer ttpledbem 8* Cosing 20 10 Lbs 21 8 510 9 520 0 213 0 22 1 0.180 1 2 8' 12.874 12.372 11 18 750 B 1.4 5' f 7.7 r -145* si no 37 300 23 47 57.70 — .089 .45 08 — Steel Sene s 1 30 8 570 2.510 2 440 0.117 0.21 I 0 175 12 3’ It 252 II ISO 17 71 650 B 1 4 S' 8" 7.7 4' 2* - 58* 80 850 37 580 20.71 55.70 004 088 .1 10 37 075 — keeroge 2.5S9 2 520 0.21 5 0.22 5 0/21 If. 34 847 / .4 s' 6“ 7.7 51404 38 270 21.23 57.14 .072 .109 38 078 SO 8.550 #•605 2 585 0 27/ 0 227 0 265 20.000 11.848 24 28 1435 4 1 S’ *• 7 7 18' 10“ + 22* 58 220 34 120 23 75 57 10 004 .070 112 .35 .02 — 51 8 529 #7/5 8 80S 0 274 0 220 0 288 11.188 If 834 24 5 4 1430 6 2 0" 2 4 9' 2" -ISO* 59 810 37 300 20 12 54.70 084 105 .075 rh.se r.,r t 52 ti * Cosing 24 3# Lbs 53 8 550 8.3 75 2 825 0.272 0 282 0 255 / 7.772 >1 838 23. / 3 24 32 1520 4 \ S’ 7 7 "• 3" - 6#® 58 85 0 34 300 24 13 5 7 t0 1 085 '07 .27 04 — Steel S4 t 550 f 645 0.282 0 220 0 25 5 / 1.113 li 831 23.52 1485 5 2 8 ’ 0" 8 4 7* 5" -ISO* 57 5 00 35 #00 20 71 51 90 — .073 101 .38 — Rue nag* 2 555 #4#7 0 287 0 272 0.258 nets itici 23.18 1438 4 7 S' 7“ ,0 58 378 35 700 22.48 57.88 .077 lot 33 .077 Fig. 37. — Tabular Statement of Principal Results of Tests, Series 2. , n* \* *Mw*V ^ ; iv> . «sa«> M»Wv\o «> ‘i 1 ' V.O* "' V ' -•'■•*'' jtoV &msmm3 oaoim oaM *o rraraoB vur>m^A tooitoabkajiT ) aciyjt \ £\y.v ■?. . ^ 1 j*syM ha^j Vtsv ’" '* jv t . f V f "js r *tr.r» | Itt 4 V 1^4 t’ Art. in Vi* »k *«♦ V ' [ • U^*' ■ . r-J- ■ | 1 . > - T ■ I*»* ' |*fk ‘tiS \\* ' r* .’'«* PV { " »; ‘ *s lV ' r i L :, Vs rr, -osT "its T< \ U w V4 St • »* • « * ,*»v rv V r is ' s* "$!* | 1 ’ os »- ■» " ’■> '' rii iz%m / : ‘j .wio’t^wva v, w> ®Gr**N * 4 V V,\sA V ' .iV;^ A'i ff | r.V, •«V»^ HW« \v '•?••»•.’ \ •'•'•' No ■sjjoV*** '.<>■ "voN^rcw -X ■-'. ■ v.0 >/ ' ' ( O ^ •- • • - v ' -i » .•. .• •'. c,vj> ;VV?v 'it -V, • j' j . 'i . i\.x\ | iv •*' 'is ’** .■ \ % u *■ j v» '->■.; fh ruv*s : «s 'si ss t*» JJ . )iS > LI ", ! Vii., SV «1 i t-f '■! Sftv -. j -> i 1 . A j * v. ^ h > : 'T ** £\r'& 7 CliV.M. . i\ ‘U, -j \ ■■: > ' '\ ■•" iv j* ns <»S iv < >V 1 "it - US v* It ■U •* • S SS ■» . is '^ , '- IV : w i y ;»i n>: is '*sl i is '** t>, -r s.v> » 1 • 5 , •> I \Ji S S ,: H\ s v t j r * \ j W '„ . : • . tat .> val..i | i‘ : ,t'. -i j ill stf. vajft (, K>i 0 4 y*\ s ,Mn t &ij 3 V',V! i o i» .Vi ; . ? • s '. s» | , :' | 'vV'j.'. tS 1.1 i. * i SV V ’ ■ s I I h A t "V * • ▼ • Kf* ?.i'i i SUV..^ t.VS-Q j >S\ ."» *>v« j . i'.H liH U j . ?b . ,V.t j ,SSQ>- 1 W i *• J t 1 X.l j %vs •• iyv 4 liV oi'T Transactions American Society of Mechanical Engineers, Vol. 27. Rbid T. Sit wart. ( SHOWING THE INFLUENCE OF OUTSIDE DIAMETER AND THICKNESS OF WALL 5 „ .. SA „, M!J) . ' PRINCIPAL RESULTS OF COLLAPSING TESTS ON ON COLLAPSING PRESSURE, for lengths of 20 feet, between end connections cid See et test u..d. e.n.n., n.m..M NATIONAL TUBE CO’S. LAP-WELDED BESSEMER STEEL TUBES lending to hold the tube to a circular form. am, noted, net in eeerege CONDUCTED BY PttOF. R.T STEWART, 1902- d F.P.K.ItOS. Test Number Outside s D'. a ™ e * er Thickness of Wall Length of Tube Weight Of Tube Lbs per Feat. Collapsing Pressure Collapsed Portion Physical Properties Chemical Analysis % Material Remarks Commercial Designation of Tube as Reported. Nominal Average "'tZ'fff!.!' V ' J average At Place of As Repartee/ Actual . ^ sffnch • 7a ge ifsed Rate at nerru w’ IbsptrSH Length f £nd Distance Tensile Strength Lbx per Sq In Yield Point Squore Inch C %Ze“ Inches Reduction of Area 7. Silicon Sulphur Phos Nang. i .// A . ' / . Op, do Greatest Least Greatest Least Nomina / Actual In Feet InOia’s LU 5 424 i 040 0.273 0.246 0.244 to Li’ / 4 ‘/oJ“ I6168 2780 3 5 * k - ,, / 7 • 3" - 25 * 57 007 35 360 22.75 54 00 054 0 77 35 .0 75 P c ■INI 6 000 5 430 0 264 0 274 0 244 . . 20 2f 1 4' lOi’ 16 45 2/50 4 5' 6' 1 1 17' 0’ - 145 ' 54 580 33 745 24.98 57 80 — 061 .047 35 .075 — ’ c 4 000 5 940 0.020 5 460 0 220 0 264 0.29/ 0.252 i t0 20 2a" n/ot 17 12 16 45 2590 C 2 S’ 0" 10 18' 0" -145* 65 695 35 745 2 1.42 63.50 — .053 .097 35 .075 — Be center C Special 6' OD 17 12 lbs 4 000 5 .450 0.271 0.225 0.250 20 2 * 20 Li’ 19' lOt 16 55 2460 4 6' o" 12 17' 0 " 0 0 58 420 36 525 24.04 59.40 ■■ ■ 060 .103 33 07 — ■ Steel c 444 6.030 5 450 0.274 0.24/ 0253 20 26" 19'lOt 16 73 2455 -< 4' 0 * 2 19’ o" - 20 ' 57 I0S 33 57 0 25.96 59.60 — .055 .048 .33 .07 — R,C Hy.ro, e 5 993 6 042 S.tSC 0.27 1 0.22 5 0.250 20' 2i“ "‘Hi’ U.S7 2487 3 2 S’ Z" 10.4 56 5 61 ss out 23.83 60.26 .058 ott ,3d .073 44 5 10 037 10 / 50 9 4/0 0 167 0.250 0 147 20 2t 14’ /Of 17.59 210 5 8' 6" 10 2 4' o * - 60* 57 425 35 090 18 46 S 2.20 — .054 1 04 .42 07 P,C 44* 10.031 10 120 9 900 0 /6 7 0.204 0 153 20 25' /9'IOt 17 59 225 3 1 1 ' o' 13 . 2 14' 0" 4 22? 55 845 38 130 20.92 59 60 — — .057 .1 10 03 075 — R.c ' 0-/66 20' O' 26" 19 I0i" B 3 7' 55 405 37 775 20.58 62 60 .32 449 10.035 10.150 4.400 0.170 0.194 0. 150 20 2h" 17.94 240 5 r o - 8.4 IS' 6“ 0° 56 245 37 845 14 71 .046 .040 .07 _ s're'eT" *.c earene 449 10.190 9.950 0 157 0 122 0.150 20 it /f'/oi" 16 55 210 5 0 “ 10 8 S' 0" 4- 35* 57 415 36 475 19.54 59.10 — .06 3 .no J4 07 — R.c Average 10.041 10.149 9 920 0.16 5 0.205 0/49 20 >?: 14' /Oh’ 17.43 225 4.2 8' 4" 10 0 S6 477 37 213 14.84 57.42 056 .102 35 .071 450 10.027 10 140 4.950 0.206 0.232 0 127 20 iX- /f'/oi’ 21 57 425 5 7' O' 8 4 3' O' 4 45' 56 645 37 105 21.46 57 70 — .052 .103 .32 .07 R,c 45/ 10 024 10 120 4.420 0.144 0.2/0 0 134 20 i i- 19' ft 20.35 390 5 8 ' 0" 4.6 4' O' - 45* 56 635 38 7 40 19.75 56.80 — J)9 / .113 3 3 07 — R.c 452 10.000 /0 00 5 10 120 9 970 0 203 0 Its 0 292 0 143 20' 0 " 20 »r ,9' 9 r 21.00 305 e 4 13 2 12 0" —165“ 58 880 40 485 18-38 53.10 080 .118 .27 .07 5 Bessemer 10" Boiler Tuhmg 2/ 00 Lbs 453 10.033 10 160 4 420 0 170 0.221 0.194 20 >/■ If 4l" 19 44 345 5 6' 6" 7 8 5' o' ~ 55* 53 675 36 455 15. 58 5 8.50 — .087 .104 31 .075 — R c 454 10 037 10.100 9 4/0 0.19 5 0 272 0 152 20 li’ 19' 4t 20 54 4 00 5 O' 4.6 •7' O’ - 60° S3 86 5 32 300 22.2/ 55 .40 — 091 .100 .28 .07 — R. c f/yereg* 10 024 10 129 4 944 0 174 0.243 0.160 20 >&■ II ■ fS- 20. 37 383 at »■ r 4.7 55 95 0 32 7/7 17.51 56.30 .078 30 .072 455 10.000 10 100 4 400 0 314 0 356 0 301 20 2i" /4/0t 3 3.01 1180 4 u ■ 6' 7 8 17' o' 4 60* 54 145 35 505 22.32 54 80 — .054 047 33 c / c 45* 9 440 10 090 4.400 0.3/2 0.350 0 303 20 25" /4'/0\" 32 27 1350 3 8' 0 0 9 6 16' 6" - 20* 57 065 36 135 23 75 55 40 — — .063 .102 .38 .07 — , 457 10.000 4 494 10.130 4.950 0 300 0 363 0 304 20' O' 20 it I4l0t 31.0 7 3 2.7 1 1275 c 4 r 9 0 IS' 4 25* 60 360 39 595 22 42 56 70 .05 7 105 35 .07 Special 10’ OO 3/ 07 L6t 459 10 003 10 100 4.420 0 317 0 356 0 310 20 2i" 14’ 10 1 32 91 1305 4 7' 0" 3.4 7' 6' - 25* 57 440 33 455 23.50 5 9.80 — 065 .108 .32 .07 — Steel c 4 57 10 023 10.040 4 240 0 314 0.391 0 247 20 2f if'ioV 32 61 1385 4 8 ' 0" 4.6 5' O' -120* 58735 35 725 21.67 58 SO — 053 .34 07 5 — £ Agrafe 10 001 10 100 4.242 0 316 0 361 0 303 20' 2i" If lei- 32 68 1319 J 8 7' 5 m 94 58 S 49 35 735 22.74 56.44 .058 .104 .34 .071 440 3 440 4.020 3 470 0 119 0 132 0 099 20 2 ' 14' Ji- 4.91 425 B IS 3 ' 0 - 4 0 S' 0- We/d not 5 7445 37 240 19 08 55 10 — 041 •J* .07 R,d 4 u 3.490 4.030 3.450 0 122 0 134 0 096 20 2 ' lt' 4 • 5 06 975 6 20 3’ 0" 4.0 4 ' 6" found 61 005 44 770 17.74 50.10 — .044 .102 33 07 — R.d 442 4 000 3.990 4.010 3 970 0 120 0 122 0 147 0. 100 20' O' 20 zt 14' 4 ’ 4 89 5 03 1030 C 5 4' 3" 12 7 19' 0 " -145* 58 380 34 150 23.46 5 7.60 — 043 078 .32 .075 — Bessemer c 4 'Converse Joint 4 84 Lbs 44 3 4.001 4.0/0 3 490 0 120 0.140 0 095 20 2 " /9' 4 " 4 98 / 030 c 5 3’ 3“ 9 8 16’ 10" We/d not 54 240 37 245 5 6.00 — .06 2 .100 .32 .07 — Steel d 444 3.442 4 040 3.950 0 124 0 049 20 25" If 3t 4 7/ 960 e J 4' 0" 12 0 2' c found 61 490 45 405 17.75 53 7 0 — 063 III .35 .07 — C fiyr.,4 3 493 4.022 3.964 0 114 0 135 0 095 20 2,i" If Si- 4.44 464 4.6 3 6' 10.5 58 5/0 4/ 172 19 04 54.50 .052 .09? 34 .071 44 5 4 010 4 020 3 990 0 173 0 203 0 140 20 24" 14' 4 ’ 7 08 2050 5 2' 8" 9 0 18' 7 ■ 4 25* *43 70S *34 105 *10.25 *37.30 .092 .144 40 .07 C 444 4 014 4.050 3 490 0 172 0.277 0 159 20 24 19 ' 4 '' 7.28 2225 3 S' 9" n o 17' 0 " 4 50* 58 675 32 075 23.83 5700 — 064 .100 .32 .07 — c 447 4 000 4 012 4 050 3 460 0 190 0 173 0 200 o no 20 ' O' 20 24" 735 7.1 1 2425 c 3 S' iso 0" 180 * 56 005 40 100 19 00 44.10 .103 32 .07 Bessemer Si'fnghsh 7.35 Lbs 449 4 019 4.050 4.0/0 0 /94 0 /92 0 165 20 2t 19 3t 7 5 3 2540 3 6 0 " ii.o 16 0 " - 60* 6/ 965 34 465 21.21 53 .30 — .086 .172 29 .0 75 — Steel c 444 4.0/7 4.040 4.000 0 /64 — — 20 it 19' 3f 6.44 2/60 3 5' 4 * 17.2 I7‘ 0" -125* — — — — — — — — — — Ay.f.,9 4.014 4. Cdl 3 499 0 /7S 0.2/9 0 159 20 24 ’ If Si- 7 17 2280 3.4 S' 0- 15.0 52 882 37 3 80 21.35 53/3 .085 no 33 07/ / 2 1 * 5 * 7 $ T 10 II IL 13 I* IS It 17 1$ IT 20 2 / 22 2 3 2* 23 2 * 2 7 21 2 7 30 3! 32 33 3* Fig. 39.— Tabular Statement of Principal Results of Tests, Series 2. Transactions American Society of Mechanical Engineers, Vol. 27. Reid T. Stewart. ( SHOWING THE INFLUENCE OF OUTSIDE DLRME TEH UNO THICKNESS OF WALL *. „„ si,.., Nf JJ PRINCIPAL RESULTS OF COLLAPSING TESTS ON ON COLLAPSING PRESSURE, ter l.n,H,s of 20 feel, bofwoon end connections d-S.. r.„ n..o ss.„ Nf 33 NATIONAL TUBE CO'S LAP-WELOEO BESSEMER STEEL TUBES Test Numb or Outside Diameter Thickness of Wall Length of Tube y Pear Weight of Tube Collapsing Pressure Collapsed Portion Physical Properties Chemical Analysis % Material Remarks Commercial Designation of rube as Reported Nominal Average At Place of Nominal At Place of Col/apse as Reported sf.fnch Used tbiperSec In F< Length End Tens Ha Strength Lbs par Sg In. Yield Point Pounds per Elongaticn S olpfll.r — Mang. Carbon — Least Greatest Least Nommol ? et AM 4 027 .... „ 0.227 0 /81 to 2 i' <*• j }■ ITT 3 12 5 3 3' o * 1 IS’ 3" 4/40- 58 445 35 7 10 23 25 54 70 052 109 3f 07 d 4.030 0 213 0 24 5 0 171 20 py 3 j- 8 47 3125 2 2' 83 18' 5" - IkO * 57 145 3k 835 24 54 55 20 050 103 33 ■ 07 4 000 0 22k 0.7/0 0 243 0.200 20’ 0 " 20 2i " If it 7 00 8 57 3150 D 3' 7 1 1 ' o’ ■*■110’ 58 245 34 SkS 23.75 42.50 052 .054 .43 07 }f Full He.fO, TOt Lis 473 a 020 0.2/7 0 252 20 2i- If jf 8 82 3375 2 2 ' 7 ’ 8 3 -130° 55 80 5 35 475 24.13 55. 40 .044 055 .31 .075 474 4.017 4.040 3 770 0.205 0.224 0 175 20 17’ 4 * 8 32 3075 3 3' O' 7 4 8 - - 35° kO 280 34 125 22.88 54.70 — .045 103 .41 025 — d Average 4 02k 4 048 4 000 0.212 0 238 0.174 20 Pi’ a' 3,r 8 43 3170 3 2’ n" 8.7 58 372 35 782 23.71 55/8 053 102 37 082 475 4 020 4 030 4 010 0 324 0 3 80 0.27k 20 2 ' 17' 4 " 12.77 5525 2’ k" 7 5 / ' 4" -120* 57 75 0 38 640 22 .17 55.30 — 055 .117 .30 0 7 — d 47k 4 on 4 040 3 780 0 332 0 373 0.30k 20 oi ' 17' 2 ' 13 05 5425 2' k " 7.5 18' 4* 4 25* 575/5 34 8 75 23.52 40.80 — .052 074 28 .08 5 — d 4 000 4 01k 4.030 3.770 0 32 1 0 32k — — 20’ 0" 2 0 12 47 12.85 5425 D 10 2' 7 5 1 ‘ k" 0 * 40 17 S 40 875 21.1 3 54.20 .043 .070 .28 .08 Bessemer 3t' C...4 5,,.ng IT 47 Lbs. 479 4 011 4.020 3 770 0 3 2k 0 3k7 0 3 00 20 ii’ 12.85 SkOO 10 2' 3" i.r I S’ -*1 75° 58 740 37 240 22 89 58.20 — - .044 08b .31 07 — Steel d 477 4.020 3 770 0.328 0.370 0.300 20 ,2 " 17' 3i * 12.87 5425 10 2' 3" k.8 /' 7” -120’ 54 455 35 255 24.83 42.50 — .053 .104 .33 •08 — d Average 4014 4.029 3.772 0 327 0 378 0 301 20 fi“ ,fil- 12.87 5 5k0 10 2' 5* 7.2 58 447 37 501 22-55 57.40 055 .077 30 on 2 777 3 020 2 780 0 no 0.1 10 0 070 20 0 “ if ,t 340 1550 5 3 ' 0 " 12 18' 0 ’ 44 400 41 470 17 .43 51.50 054 IIP 34 075 d 3 001 3 010 2 770 0.103 0.130 0 088 20 0 ’ if ir 3. 18 * 4k5 5 3 ' O’ 12 18' 0 ’ +150’ — — — — — — — — — R, d 3.000 3 00k 3 030 2 740 0 107 0.1 10 0.1 18 0 07k 20' 0 " 17 nl’ if ,f 333 3.40 1 k 30 C 5 2 ' 3' 18' 5’ + 170* 57 540 4/ 535 18.00 54.40 052 34 07 Bessemer 3" Slander a Barter Tobmg J 33 LAs 493 3 001 3 010 2 7k0 0.1 10 0117 0 077 20 o " if if 3.40 1725 5 / ‘ 4* 4 1 ' 4* 0* 4/ IkO 37 575 20.77 54.40 — 04k .102 .33 .0 7 — Steel a 484 2 773 3 000 2.7k0 01 II 0.078 20 0 ’ if if 3.425 2025 5 3' 7' 15 nr + 35’ kO 305 42 555 20.50 52.10 — - 043 l l l 34 075 — d flvaroga J 000 3.014 2 7kk 0.107 on 0.014 20’ 0“ if if 3.34 1733 5 2' 8 - 10.8 kO 70k 4/ 537 17.73 54.15 C5J 104 35 073 495 i 78k 3 010 2 770 0 112 0.078 20 o’" if if 3.4 5 1800 5 4' 0“ lk II' 0 ’ + 30’ 58 285 40 445 21.13 52.30 .046 088 .25 0 7 — d 49k 2 777 3 020 2 770 0 l 12 o no 0 072 17 Ill- if if 3.45 1850 5 3 ' 0“ 12 /' 7‘ 57 075 41 700 17.13 52.30 — 045 103 .30 .08 — a 497 3 000 2 77 8 3 020 2 780 0 120 0 130 0 07/ 20 ’ 0 " Ill’ n if 3 43 3.45 1 7k0 c 5 3' 3*' 13 2’ 0 * + 45° 45 170 51.70 .040 31 075 Bessemer 3" Locomotive Bader Tubmg 3 k3lks 439 2 778 3 020 2 77 0 0 / 14 0.1/S 0 100 17 if ,f 3.53 2025 5 3 ' 4” /4 2 ' o’ -no’ 42 455 43 35 5 17.72 50.60 — 053 112 27 08 — Steel d 497 2 772 3 000 2.760 0 113 0 12 3 0 104 20 0 " if if 3 48 2/75 10 3’ 7’ IS I7'I0" 55 195 42 0/5 18.54 57.20 — 0k5 .104 33 075 — d average 2 774 3.014 2 770 0113 0.122 0 077 n III - „■ ,f 347 I7k2 It 3' k m 14 kO 4/5 42 4/7 17.53 52.72 .050 .100 30 .074 470 3 000 2 777 3 0/0 2 790 0.150 0.147 0 151 0 138 20’ 0" 20 ik" 17’ 2l ’ 4.57 448 3350 o j 4* 7“ 15' 0 " 0* 55 70S 40 785 20.42 50.00 074 no 27 075 Bess Steel d Special J’OO. 4 57 lbs 2 78 7 3 010 2 770 0 ISO 0137 0/37 0./25 20' 0 " 20 17' It 4 57 2575 20 2 ' 10 13' 0 * 45 705 02 5 07 Average 2 772 3 0/0 2.775 0 143 0.145 0. 132 2f 0 f If 2j- 43k 27k3 3 * 8 ' 14.5 52 705 35 90S 21. 75 53.45 4 75 2.770 3 010 2 770 0.170 0.2/8 20 ol If 2f 5 48 4200 3 2 ' II 5 18' 3" w fVt» n : r 70 315 45 350 2/25 51.60 071 077 .32 02 — d 47k 2 77k 3 020 P 780 0 17/ 0.2/5 20 If 2f 5. 73 4200 3 3’ k" 14 10’ 3“ + 35* 57 455 37 755 20.42 57.50 — .060 103 3 1 ■ 02 — d 477 3 000 2 777 3.020 2 7k0 0 180 0 170 0.215 0.16/ 20’ 0 ’ 20 !f 2f 5.42 5.42 4175 0 3 3' 3“ 13 13’ O’ 55 2/5 37 330 22.25 5 k. 80 100 .30 .075 Bessemer Speed! 3’ OD 5 42 Lbs 478 3 000 3 020 2.7k0 0 182 0 172 o its 20 1 ’ If 2i- 5.48 3740 5 3' 3” 13 Ik' 8" + 15’ 58 375 34 255 17.42 5 7.50 — ,044 .102 .30 07 — Steel d 477 4 174 3 000 2 780 0 187 0.227 0.170 20 li If 3 " 5.45 4200 3 3 ' 4" 14 10 * 3- W ’J‘Sd l 42 800 38 415 22.4k 57.60 — .045 .075 .27 07 — d Average 2 77$ 1 3 014 2 770 0 0.1,0 0 U8 20 - -V 2 f 5.44 4075 3.4 3' 3" 13.1 kl 432 35 02 1 ~zi7iV 54 .28 0,4 .100 3C 0 75 Fig. 40.— Tabular Statement of Principal Results of Tests, Series 2. „T£ .joV x JAory.ABO^W 10 tt&ooS - .wiaawJ .ihoitoasa :?T hwsiwx qw *vta>ma *a\rv j*> >0 *v.v*$v?. '«v\*!> 4$. *fc.N\#vva\ vA ,V.W2.^S 'AVA?,*^, •■/>■>. •a*Jv%v\'o « »\ iA\ V\«r *>*( Transactions American Society of Mechanical Engineers, Vol. 27. Reid T. Stewart. ( SHOWING THE INFLUENCE OF OUTSIDE DIAMETER AND THICKNESS OF WALL o.n.r.t sn.ar Ntji. PRINCIPAL RESULTS OF COLLAPSING TESTS ON SERIES 2 < ON COLLAPSING PRESSURE, for lengths of 20 feet, between end connections c s.'/JZ'.r n.r .n e./.r.i n.m.m s*..r Nt ly. NATIONAL TUBE CO'S. LAP-WELDED BESSEMER STEEL TUBES \ tending to held the tube to e eircutar form. CONDUCTED BY PROF R.T STCWART, 1 002-4. FPK.I9C5. Ttit Number Outside Diameter Thickness of Wall Length of Tube " Feet Weight of Tube Lbs.por Feet. Collapsing Pressure Collapse , d Par Hon Physical Properties Chemical Analysis % Material Remarks Commercial Designation of Tube as Reported Nominal Average At Ptec^ of Nomina / Average At Place of Cell epee As Reported Actual Unsupported Round* Gago Ra,.., Length from Distance from Weld Tensile Strength Lbs.perSg.ln. Yield Point Rounds per Square Inch Hengetie* %in8 Stlicon Sulphur Phos. Many Carbon Oxide Great**/ Leant ' Greatest laaat Nomina/ Actual Ux>C*rS« In Feet InDia's. 10 771 10 930 10.740 0.508 0.570 0.477 If III’ „■ ,■ 55199 2450 s 9' 9‘ 7.5 If 7 ' ■919 5* 54 405 30 90S 27.83 57.7 0 .050 088 .37 07 e 5 01 10.779 10.910 0.51 1 0.593 0.490 15' 0" If si- it' 2i m 5 9.099 2575 4 8.7 ■9140' 55 250 37 210 19.99 90.40 .054 .100 .39 .07 c 602 10 750 1 0.707 — 0 500 0 509 0.470 20 O’ n'nk m If 7j" 54 25 5 5.524 2920 c 3 ' — — 59 090 31 245 25.42 57.40 — .057 .087 .34 .07 — c IV'fstre Strong S4.25.L9x 10.799 10 950 10.710 0.509 0.524 0 499 Iflli' 55 575 2470 5 7' 3" 8.1 3' I" O' 57 575 33 780 22.72 42.10 .051 J2 500 10.797 10.910 10.740 0.529 0.579 0.51 1 17' Si’ If It 57 .999 2770 4 8' 0“ 8 7 15' O' o' 65 200 31 47 S 23.42 52 50 — .059 .071 .34 .07 — c Reorogo 10.777 10.939 10 713 0.512 0.59 5 0.497 If S|- If 3i’ 5 9.194 2595 4.2 rir 8.7 55 702 32 707 23.97 5 4.82 .054 .094 .35 .071 505 12.772 12.830 12.730 0.505 0 570 0.490 If II V IT' IT 99.297 2395 5 10' 0 ' 7.4 7' O' ■9135' 52 495 31 795 24.27 92.40 .057 .097 .33 07 R,C SOD 12.772 12.920 12.710 0.519 0.577 0.497 IS' O' ifni- If ti- 97.535 2295 5 H' O' 10.4 7' 9’ ISO ' 55 405 31 095 22.75 59.70 — .057 .075 30 .07 R.c 507 12.750 12.910 12.990 12.730 0.500 0.507 0 529 0.499 20' 0* i fu- ll' 7|- 95 00 99.549 2070 c 3 10 ' 9" 7 7 55 985 38 035 25.75 94.9 0 .055 .092 .27 .07 Bessemer / 2“ Extra Strong. 85.00 Lbs. 509 12.774 12.950 12.750 0.515 0.552 0 502 lfill'. If 7 i‘ 97.529 2000 5 7' O' 8 5 4' 9 * -20' 53 7 SO 30 80 5 21 St 5 1.90 — .058 09 7 .32 .07 5 __ Steel p c 507 12.792 12.930 12.730 0.513 0. 673 0.501 ifni’ If 7 1' 97.157 2220 5 7' 9" 8.7 4' 7' ISO' 54 395 31 075 24 1 3 90.90 — .055 .092 .32 .0 75 — R.C Reorego 12 770 12.938 12.730 o.5i ; 0.598 0.498 ifni- 17' 9 ' 97.007 2179 4.5 10' 0 " 7.4 54 394 32 577 23 70 57.98 .057 093 .31 OIL 510 13. 042 13.070 12.770 0.243 — — 20‘ 0" 17' 9i‘ 33.300 440 5 13' O' 12.0 f 5' O' 58 005 35 935 28.34 57.40 .08 7 III .39 .08 R.c 13.024 13.070 12.790 0.244 if ' la o* 20' 0" 17' 9i' 33 250 43 0 4 II' 0 ' 10.2 O' 57 5 95 37 245 23.29 57.30 .40 .07 512 13.000 13.03 8 13.050 12.790 Site 4 0.244 2 0*0" 20' O' 17' 9i ' — ■ ■ — 33.350 SIS B 5 12' O' 11.1 ■9 25' 59 880 34 390 2 5 43 90.80 .055 .37 .075 Bessemer Hi’ Casing if 00 5/5 13.039 13.090 12.770 O.W.6. 0.249 — — — 20' 0 ' 17' Mi- 33.900 490 5 H' O' 10.2 13' 9" ■9 10* 91 790 41 395 20.17 59.10 — .048 .102 .34 .08 — Steel R.c 5/4 13.030 13.090 13.020 0.245 — — 20' 0 " n' 9i" 33.500 450 4 13' 0" 12.0 9 ' 9 ' •f 30' 59 225 35 975 29.21 57.90 — .052 .087 .37 .075 — R.c Rrerage 13.039 13.099 12.784 0.244 20' 0‘ 17' 9i m 33.400 493 4.9 12' 0 * II. 1 58 987 39 979 24.5 3 59.94 .051 .100 .38 .0 79 7 9 21 28 3 1 Fig. 41.— Tabular Statement of Principal Results of Tests, Series 2. - 2 . lIO V ,sns«mo*ia jAoiKAHoaM to rwiooB kaoiictmA wohdawahT ] S ai«vaa »ftS Waft aH vWw**' ) .. ., , * ny tv «\-v; VMk?. ^ -AttlVTOft ^ ^ Jvtl ^ U ^ ***** - - » ,\*W '* m\tt*vvifc’»3'J ' X t*'»»»'*' ” *"}* 'k-'fV V-V". V.-VU'M^l “‘• V MW>■' *» tvfc'Av M*S tivA \v*t t 0»V»5: «#**»-***' r >»- •, U *>t *»■*■ rtWv».- %*'•-! »- v-fc «. '<■.«-> IW.'^VV ' .it** » >•’ .tftft W* A 1- s Sx»*» *?'»* ' *'•** " ' ,MS * V v.>V * *v *«*<*.» VV»>w c . *• ■,.<»£ ft**l %4v«*>»» A* V.*W*r'»vV: «"M\V» «Vt *♦>»' ■><* in» «Wft" » ' -'*'*’* v i ‘" 1 V.»x* Au t 'l A^i >-?-& *Av*A \Q V,A*> *»<> "1 * ■ ' '* *>'■£ .}** A'.t.Oki *A 'i' l \\.A l ifc»’>U<'A«' O'*' v.*A* "i* £ttemsas» A|P T**xN»*.- -'a fr™** ■' '•'•ft t f ' r* -wvaA. - . »>{ >Vi,» * M>cva*A< » * fc ■ ^ -V>1T> »**t l ; ~ «i* a**! Bit*. ftftt »«v. V'“ i'AW» -''■>*< "w*"' ^W-SSWOra* ' tf£? ■•■th-ir^j?;'-. K.sa^ t 4\ «.' v*v 'VC ' '• •= ' •• ^ -^JUax- • ' «*•'»*» iAd \ANftdiA( ua %wl*Vr ..s** **«•' l «✓* - ^ N * A«\Mt -.•«V'»' S- 1 . - - •'•’'* •ft** »'*» *' k \**-^ P* ,1**1 »>•» t vJ*. *"»?. Vvt4tV« SH»A'V» '• ■* ' ? . ..-■ »*-•* ' •"'vW • WA >.' '*>•» \,->tV tvt X,\^ r>s *Mt PftV >U v »> . v *>„<*»* *" ■*'•'' v -' '” ,„«.w , .' j' «« x-Wit'o-. t'O »w** »'•< •*••••». /W^VWJtC- t»w*'. n •'A «'»-'*• V a- "■••■> »A\ Ov\* VVV.-V* ,»•» S**CT.O,\» ■■• "» V*X. 1V»«-A • ‘'V- ' V l'*' 00 V Iti'u <»*«sv XA'w. ' ». •, V. vVv.Xi V«\ XO.O+ «vo»* V-WV .-1M' v*" 't»-V**'- V »*"* ,, V Vv»W''4'»M« ■ <*' -• '-«■«** Vox'trto **'•• S'"* a ,.* n*.i-*w*a -ia*^ ' ,., \.»« ■»*%» «•%*»»«*'•" ,U»\ So ... v K\t« A\«^^ 4*X%t) **" .’"M* “* ■. ,, ,, •, .... ■■ .-■ *»**. *< ■• " ■•• ,i„'^ «*••''•”• - »'*”-•'»**' .' • .AS- • ■»>'* • v -(*v.v." s.AW"' ■'■■•' ■ • '■ '“•• ; ' \ / Vol. 27. Reid T. Stewart. Fia. 42 .— Tabular Statement op Principal Results op Tests, Series 2. Transaction's American Socnerr or Mechanical Engineers, Vol. 27. r / COLLAPSING PRESSURES OF LAP-WELDED STEEL TUBES. 95 photographs that I have here. I have photographs of all the tubes tested, and if you will look at them you will see that a long tube is only distorted over a small portion of its length. Referring to the photographs of tube No. 50, Fig. 16, p. 30 for example, it will be seen that the left-hand 14 feet of its length has not in any way been distorted by the test. We had a very precise way of determining the distortion which is fully explained in the. paper. This photograph of tube No. 50 shows clearly that had we cut off a length of about six diameters from the right-hand end of this tube and put it in the testing apparatus, that this portion of it would have collapsed, just as it did when attached to the 14 feet that showed no distortion whatever. These commercial tubes, taken at random from the company’s stock, were, generally speaking, slightly more out of round near one end than elsewhere along their length. This is clearly shown in the body of the paper (see Fig. 53). The tubes are evidently weakest near one end on the average, but the results of this weak- ening influence are of no practical importance, not exceeding 13 per cent, and averaging 4 per cent, for a series of determinations made for it. It is the practice of the National Tube Company, so far as I know, to keep the tubes continually rolling while cooling down, so there is no chance in the regular operation of the mill for a tube to be distorted in the manner suggested by the last speaker. President Taylor . — Is there any further discussion? If not, I would like to add to what Mr. Rice has said in appreciation of Professor Stewart’s paper. It seems to me that the scientific man- ner in which the subject has been treated is most worthy of com- mendation. The fact that the tubes experimented with were taken at random from the stock, adds very greatly in my opinion to the value of the tests. It seems to me that we should be very thankful to Professor Stewart and to the National Tube Company not only for making tests of that sort, but for going to the trouble of presenting them to our Society. It is just such papers as this which are of the greatest permanent interest not only to the members of the Society, but to all Engineers the world over, and which gives our Society the international standing which all of us who are ambitious for the Society are anxious to have it attain. [HE UBKABT Of [HE JUL 2 3 1924 UNIVERSITY OF ILLINOIS M