US 3439405 A
Abstract available in
Claims available in
Description (OCR text may contain errors)
plll 22, 1969 BERMAN ET AL METHOD 0F VESSEL FABRICATION nmlufnllw w 3H www www m/aiJ/v/ United States Patent Office 3,439,405 Patented Apr. 22, 1969 U.S. Cl. 29-407 6 Claims This invention relates to a method of manufacturing pressure vessels and the like, and in particular, to a method of manufacturing vessels from two layers of material.
As the pressure requirements of vessels increases, it is well recognized that the thickness of the wall material must also increase. However, as the thickness of the vessel material, in general a ferrousmetal, increases in thickness, the problem of obtaining a uniform hardness of the material across its entire cross-section becomes increasingly difficult. As a result, the cost of the material becomes substantially more expensive and difficult to obtain.
In the past, the generally accepted practice in pressure vessel fabrication has been to use rolled forgings. Very worthwhile economic savings could be achieved if brake bending were used in place tof rolled foi-gings. With thinner sheets, brake bending could be utilized providing a suitable method of fabrication was available which could economically produce a multilayer vessel.
A generally accepted configuration for a pressure vessel is a cylinder with a hemispherical head at each end. Since the hemispherical heads must withstand approximately one-half the stress of t-he cylindrical portion of the vessel, the hemispherical heads can be one-half the thickness. For this reason, it is practical to construct an inner cylinder with two hemispherical heads all of which have substantially the same thickness. An outer cylinder is then formed and slid on the inner cylinder. A clearance between the inner cylinder and the outer cylinder is required to permit the inner cylinder to tit into the outer cylinder. Then the inner cylinder must be expanded into the outer one until there is virtually no freedom between the two cylinders. In order to achieve such an expansion, fabrication techniques are required by which the proper fabricating pressure can be accurately determined. Without the proper determination of such a pressure, pressure levels can be reached at which an instable fabrication occurs resulting in deformation and fractures. To further complicate the problem, the pressure for correct fabrication and the pressure for instable fabrication are very close together.
Of perhaps even greater importance is certainly that the two-layer vessel is properly fabricated. That is to say, it is essential that the two layers act as a monolayer at all pressures and that the method of fabrication include a control on whether or not fabrication is complete.
Therefore, it is an object of this invention to provide an improved method of fabricating two-layer vessels.
Another object is to provide a reliable and economical method of controllin g the fabrication of a two-layer vessel.
Another object of this invention is to provide a method of fabricating a two-layer vessel which gives a positive indication of whether or not the two layers are acting as one layer at all pressures after tentative fabrication.
In accordance with this invention a hitherto unused indication, namely, longitudinal strain gages on the outside shell is shown to be a sensitive index to determine the adequacy of internal pressurization of fabrication. Longitudinal strain gages may also be used, either with or without circumferential gages, to indicate the onset of excessive plastic deformation. Longitudinal strain gages are attached to the outside surface of the outer cylindrical shell to observe the longitudinal strain behavior while the inner vessel is being expanded by any suitable method such as hydrostatically. A suitable internal pressure is selected so that the inner cylinder will become fully plastic and contact the outer cylinder without creating an instable condition. The internal pressure is then reduced. If the longitudinal strain gages show that axial sliding exists between the inner and outer cylinders, the inner pressure is again increased to a level above that initially applied so as to slightly increase the strain gage readings. Upon depressurization, the strain readings are again observed to assure that no axial sliding exists. Should any occur, further increased pressurization is applied until no axial slipping occurs.
This invention may be better understood from the following detailed description in conjunction with the accompanying drawings in which:
FIGURE l is a side elevation in cross section of a vessel prior to expansion of the inner cylinder into the outer cylinder;
FIGURE 2 is a cross-sectional view of FIG. 1;
FIGURE 3 is a chart of the irregularities between the inner cylinder and the outer cylinder in the sections of FIG. 1 and the positions of FIG. 2;
FIGURE 4 is a graphic illustration of the average measured relationship of internal pressure and strain at section x-x of FlG. l for two pressure cycles; and
FIGURE 5 is a graphic illustration of the stress-strain relationship for tensile coupons taken from the parent and weld materials of the inner shell.
Referring now to the drawings and particularly to FIG. l, a vessel 11 is shown formed from an inner cylinder 13 with a hemispherical head -15 at each end. Surrounding the inner cylinder 13 is an outer cylinder 17 open at both ends. As was previously pointed out, a double thickness in the hemispherical heads 15 is not required since the hemispherical heads can withstand identical internal pressure with only approximately one-half the thickness of the cylindrical portion. An annular space 19 exists between the inner cylinder 13 and the outer cylinder 17. Since these two cylinders 13, 17 are formed by brake bending, irregularities exist in their shape. However, a nominal clearance of V16 inch did exist between the two cylinders with variation in clearance ranging from lf2/2 of an inch to 1A@ of an inch. The variations in clearance are caused by local changes of curvature as well as overall out-of-roundness. These noncircularities also vary axially. In addition, material property and plate thickness variations create stress nonuniformity. The end conditions are not considered because they are local and the heads would reinforce in operation as well as fabrication. In FIG. 3 is a chart of the actual measurements taken at sections 1A through 5A for positions 1B through 8B of FIG. 2.
It should be understood that the scope of this invention is not limited to either the specific configuration or the dimensions or both as herein described and as illustrated in the drawings. Similarly, the specic material and its properties are not limitations on the scope of this invention. The particular embodiment described is being used purely by way of example, as the explanations of the concepts involved and the advantages of this invention are best understood in the light of a speciiic numerical example.
The various dimensions of the particular vessel being utilized are shown in FIG. l and need not be repeated. The particular material selected is USS T-l steel.
3 4 The chemical composition of this steel is as follows: Using these values, pi is equal to 21,400 p.s.i. This CMn sis P Cr Ni CuVMo B outer cviinder .i7 .90 .2s .oi7 .oii .53 .79 .25 .07 .4G .003 Inner cyliiider .16 .84 .22 .021 .012 .53 .80 .25 .07 .46 .0025
In this specification the following symbols will be used: pressure should be on the low side of the true pressure azinner radins Oi the inner cylinder (inches) `because the plates are generally received thicker than ao=Inner radius of the outer cylinder (inches) nominal' ,:Outer radins of the inner Cyiinder (inches) A comparison of the tentative pressure calculation wth bozouter radius of the Outer Cyiinder (inches) 10 the'instability pressure pu shows that the useof longic=Radial distance to the elastic-plastic boundary of a iudiiiai Strain gages as en ald m the determination of the cylinder (inches) fabrication pressure is important. E=Y0nngss Modinns (p si) Strain gages can be located at numerous places about nizinternai pressure (psi) s the vessel 11. However, for purposes of simplicity of p0=Interface pressure between the two cylinders (p.s.i.) i explanation. this explanation will only Consider ioiigi' rZRadiai distance of a cylinder (inches) tudinal strain gages located about the vessel 11 en line izinner radins of a cyiinder (inches) x-t of FIG. 1.-Water was used as the pressurization rozonter radins of s cylinder (inches) medium and was introduced through the opening 21. L=Radial displacement of a cylinder (inches) e0 For the purifose of pressure fabrlcauontile limsi ves' Y=Yield strength of the material (p.s.i.) sel was pressurized from zere to 22,400 p.s.i. in nine'increergRadiai strain (inches/inch) ments and the depressurization was accomplished in five Elizcircnniferentini strain (inches/inch) steps back to 400 p.s.i. The vessel was subsequently preseZ=LOngitndinai (axial) strain (inches/inch) .surrzed to 23,100 p.s.i. in ve steps and unloaded to zero zpoissons ratio in six steps. The pressurization to 22,400 p.s.i. and subseirrzRadiai stress (p si) quent unloading is designated as Cycle'l. The reloading to as :circnrnferentiai stress (p si) 23,100 p.s.i. and its subsequent unloading to zero is desigorzzLongitndinai (axial) stress (Psi) nated as Cycle 2. In all cases, at pressures above 20,000 p.s.i., each pressure was held for at least fteen minutes. The first step in fabrication is the selection of a tenta- The pressure Was not changed until none of the strain tive pressure by means of calculation. In order to detergages showed any further increase of strain mme a tentative pressure a yield Point for tht? Vessel ma Initially as the pressure inside the inner cylinder 13 is terial must .be obtained, The use 0f the equations of per' increased, since there is rio contact between the cylinders 'fffct Plasticity requires that the siifess'siram Curve Suhl 13, 17 there is no strain on the outer cylinder 17. It was Ceniiy approximates Such a material. only i0 the Strains ascertained from analysis that because of the large gap 19 reached- For the T i Steel used consider the Point on the the inner cylinder 13 becomes lfully plastic prior to its corinVerage tensile yield cnrVe at Which the total strain is tact with the outer cylinder 17. Since change in the 1onequal t0 that necessary t0 close' the nominal geP- ln this gitudinal strain of the outer cylinder 17 is zero, the slope Casefhc ratio 0f the gap (si/16 inch) t0 the inner radius on the pressure versus longitudinal strain line from 0 to (18 inches) is 0.0104. The use of the gap to inner radius pi is innite as seen in FIG 4 ratio as the total strain of the inner cylinder is arbitrary. 40 At internal pressure p1 the inner cylinder which is fniiy The Strain 0n the inner cylinder is a hit higher than this plastic contacts the outer shell 17. The outer shell 17 is smce 1t eXPsndS Into the 0uter cyl1nderradially expanded, but the pressure is not sufficient to The tensile test stress'straln curves may he seen ln prevent axial sliding between the two cylinders 13, 17. FIG: 5 fOr the Specimens from the Parent metai and Thus, from internal pressure p1 to p2 the outer shell 17 specimens transverse to the weid. If the tOal Straln 'Of indicates the behavior of an elastic open-ended cylinder 0-0104 is Used in conjunction With the mean 0f the scat- 45 subjected to a pressure po where p 0 is the contact pressure. ter band 0f the inner cyllnrier, the result is a yield stress At internal pressure p2 the pressure contact between 0f 112,300 P-s-lf0rthe 1nner cylinder- The tentative the two cylinders 13, 17 is sufficient to prevent adidtional Pressure fOr fabrication 1s Only approximate and for axial sliding. The sliding may not be totally prevented; simplicity, s yield Strength 0f 112,300y P-si- Will he Used but, for the purpose of the analysis it is assumed that the fOr hOtl'l cyllnrlers- 50 subsequent pressure force is resisted jointly by the two The inner cylinder becomes fully 'Plastic and contacts cylinders. 13, 17 in proportion to their cross-sectional the Outer cylinder at Which tirne hcth an inner and Outer area. The pressure increase is continued until the pressure pressure acts on the inner cylinder. A relationship bepi reaches p3, the tentative pressure tween the pressures and the getimetric and materiel Prop' Upon reduction of the internal pressure the inner shell eftieS 0f the Cylinder may he Written as 55 13 reverts to elastic behavior. The two cylinders 13, 17
an noticeably act as a four-inch thick walled monobloc cylpi=po+ Yo ln [m] inder with closed ends. As FIG. 4 readily shows, the slope The value of po may be determined from consideration ifahfctlilivielsnc Cie 1 from p2 to psf 1s less steep tn of the outer shell. Since the outer shell feels only the dirreren in sloeg gg iiireeisur Cltlivef Tg; [z3 isili* e pressure on its outer surface and the criterion for fabricanot be totali preniied orilleiodil asn is l (iisg nary tion pressure is assumed to be initial yielding ofthe outer tendenc to Sii'idi is reduce so theregs'houid 10a mign. cylinder, the result may be obtained from Trescas yield at th hilf he d i d. U A eiio s1 me criterion at the inner surface and written as f e g. r presslires urmg un Oa mb' not er fesso 2 G5 or the difference in slopes may be the local plasticity [1 9 0) :l cfects on loading.
2 bn When depressurization reached an internal pressure of From these two equations, the value of p1 can be calp4, axial sliding reoccurs. The outer cylinder again acts as Culated llSng data 21S fOllOWSI an open end elastic cylinder with pressure p0 at its inner E=30X106 p s i 70 surface. The inner cylinder acts as a closed end elastic =0'.3 cylinder with a pressure p1 on the inside and po on the a=18 inches outside. b 20 inches The axial sliding between the internal pressures p4 and a0=20.188 inches p5 of the first cycle is an important feature of this invenb0=22.188 inches 75 tion. The pressurization for fabrication should be carried out in at least two steps. The first step should be a pressurization to the tentative fabrication pressure such that reverse sliding to relieve the axial force between the cylinders occurs on depressurization. The second pressurization should not cause large overall strains of the outer cylinder.
The graphic effect of reverse sliding is a feature of longitudinal strain gages, This axial sliding cannot be noted if circumferential strain gages are used. This is because with circumferential strain results, the slopes for the monobloc cylinder behavior and for the reverse sliding are identical.
The large axial force between the two cylinders 13, 17 caused by diiferent elongations in loading and the friction force was largely dissipated in the reverse sliding. Therefore, on reloading for the second cycle the cylinder contact is sutlicient to cause elastic monobloc action to the maximum pressure of previous pressuriaztion. This is shown in FIG. 4 between p5 and p6. At p6 the strain is the same as at p3, since the scale has shifted due to axial sliding.
Additional pressure is added to cause a slight increase in longitudinal strain as shown between p6 and p7. The longitudinal strain gages at this point in the fabrication serve to give an indication of adequate fabrication pressure as well as warn of plastic instability. With longitudinal gages, a severe change in strain indicates that deformation is taking place. For this reason, while the additional pressure increment is being applied in the second cycle, the longitudinal gages should be closely observed and the pressure released at any showing of a run away condition. It should be pointed out that with circumferential strain gages, a small amount of deformation may be noted at the elevated pressure of the second cycle on individual gages as well as on averages of the gage readings.
The depressurization shown as p7 to p8 takes place as a monobloc cylinder and since no reverse sliding is evident, fabrication can be considered complete. Where reverse sliding occurs, the general procedure of the second cycle would be repeated as a third cycle or until reverse sliding stops.
It is to be understood that the above-described methods and arrangements are simply illustrations of the application of the principles of the invention. Numerous other methods and arrangements may be readily devised by those skilled in the art which will embody the principles of this invention and follow within the spirit and scope thereof.
What is claimed:
1. A method of fabricating a two-layer pressure vessel:
forming an inner shell, said inner shell being a complete enclosure with an opening therein;
forming an outer shell about said inner shell, the inside of said outer shell being slightly larger than the outside of said inner shell;
determining a tentative internal pressure level for application to the inside of the inner shell to expand both the inner shell and the outer shell;
applying a tluid medium to the inside of the inner shell through the opening therein to increase gradually the pressure up to the selected tentative inner pressure, thereby expanding both the inner vessel and the outer vessel;
removing the fluid medium to decrease the pressure within the inner vessel thereby contracting both the inner vessel and the outer vessel;
observing the longitudinal strain on the ouside of the outer shell in relation to the decreasing pressure to determine if their ratio remains constant thereby determining presence of reverse sliding;
reapplying the lluid medium to the inside of the inner shell through the opening therein to increase gradually the pressure up to the selected tentative inner pressure;
further increasing the pressure a slight amount over the tentative inner pressure;
removing again the fluid medium to decrease the pressure within the inner vessel thereby contracting both the inner vessel and the outer vessel; and
observing again the longitudinal strain on the outside of the outer shell in relation to the decreasing pressure to determine if their ratio remains constant thereby determining the presence of reverse sliding.
2. A method according to claim 1 wherein said inner vessel is formed from a cylinder with a hemisphere lat both ends.
3. A method according to claim 2 wherein said outer vessel is a cylinder open at both ends.
4. A method according to claim 3 wherein the cylinder of the inner vessel and the outer vessel are fabricated by brake bending.
5. A method of fabricating a two-layer pressure vessel:
forming by brake bending an inner cylinder;
forming two hemispherical heads;
welding said two hemispherical heads to said inner cylinder to form an inner vessel, said inner vessel having an opening therein;
forming by brake bending an outer cylinder, said outer cylinder being at least as long as said inner cylinder and having a nominal inner diameter slightly -larger than the nominal outer diameter of said inner cylinder, said outer cylinder and said inner vessel having substantially the same thickness;
sliding said outer cylinder over said inner vessel;
placing longitudinal strain gages about the ouside of said outer cylinder at a location substantially remote from either end of said outer cylinder;
determining a tentative internal pressure for application to the inside of the inner shell to expand both the inner shell and the outer shell;
applying hydrostatic pressure to the inside of said inner vessel through the opening therein to increase gradually the internal pressure within the inner vessel up to the tentative internal pressure; observing the relationship between the longitudinal strain on the outside of the outer cylinder and the increasing internal pressure to avoid deformation;
removing the hydrostatic pressure to decrease the pressure within the inner vessel thereby contracting both the inner vessel and the outer cylinder;
observing the relationship between the longitudinal strain on the outside of the outer shell and the decreasing internal pressure to determine if the ratio remains constant thereby determining the presence of reverse bending;
reapplying when reverse bending occurs a hydrostatic pressure to the inside of the inner vessel through the opening therein to increase gradually the internal pressure within the inner vessel up to the tentative internal pressure;
observing again the relationship between the longitudinal strain on the outside of the outer cylinder and the increasing internal pressure to avoid deformation; further increasing the hydrostatic pressure a slight amount above the tentative internal pressure; observing the effect of the increase in internal pressure on the longitudinal strain to avoid any substantial increase in longitudinal strain; removing again the hydrostatic pressure to decrease the internal pressure within the inner vessel thereby contracting both the inner vessel and the outer cylinder;
observing again the relationship between the longitudinal strain on the outside of the outer cylinder and the decreasing internal pressure to determine if the ratio remains constant thereby determining the presence of reverse bending.
6. A method according to claim 5 wherein the tentative internal pressure is determined by the relationship Y a., 2 ao foi-albe) l+Y mlm-bl in which p, is the internal pressure within the inner vessel, Y is the yield strength of the material, a0 is the initial inner radius of the outer cylinder, bo is the initial outer radius of the outer cylinder, a is the initial inner radius of the inner cylinder, and b is the initial outer radius of the inner cylinder.
References Cited UNITED STATES PATENTS 10 CHARLIE T. MOON, Primary Examiner.
U.S. Cl. X.R.
U.S. DEPARTMENT 0F COMMERCE PATENT oFFlcE washington, D c. 20231 UNITED STATES PATENT OFFICE CERTIFICATE 0F CORRECTION Patent No. 3,439,405 April 22, 196
Irwin Berman et al.
It is certified that error appears in the above identified patent and that said Letters Patent are hereby corrected as shown below:
"tof" should read of Column 3, lines 56 to 5E Column l, line 25,
line 5lI "cylinders. should read cylinders Column 6, lines 52, 53 and 75 "bending", each occurrence, should read sliding Signed and sealed this 14th day of April 1970.
WILLIAM E. SCHUYLER, JR
Edward M. Fletcher, Jr.
Commissioner of Patents Attesting Officer