US2673613A - Apparatus and method for fatigue determinations - Google Patents

Description

SEARH RSM@ 16h-wn S9 Fips-m1 0R ,2,673,613

March 30, 1954 E. M. IRWIN APPARATUS AND METHOD FOR FATIGUE DETERMINATIONS FIG-.1.

MpL/F/ER OJ A. 9 l 7 l t C O d e l .l F

[Maf/77' M. /Rw/M INVENTUR.

,4free/vsn E. M. IRWIN March 30, 1954 APPARATUS AND METHOD FOR FATIGUE DETERMINATIONS 2 Sheets-Sheet 2 Filed Oct. 1'?, 1949 EMMEN' M. /xew/N, l INVENToR.

#rra/elven Patented Mar. 30, 1954 UNITED STATES PATENT OFFICE APPARATUS AND METHOD FOR FATIGUE DETERMINATIONS Emmett M. Irwin, San Marino, Calif.

Application October 17, 1949, Serial No. 121,672V

'7 Claims. l

All known metals have been observed to have the characteristic of failing under certain types of strain, particularly strains which are cyclic in their nature. So long as these strains and stresses do not exceed a certain maximum for any particular metallic unit that unit may well stand up indefinitely. Let the maximum be exceeded, however, and the stresses cycled, failure usually occurs rapidly.

The problems presented by failures in metal through fatigue are general and widespread. They occur not only in structures which support varying loads but also in machines incorporating reciprocating moving parts subject to variations in loads. Wherever metal is subjected to stresses which vary, the problem is present. An outstanding example comprises the drill pip@musedu in oil drilling which extends downwardly into the earth, sometimes for miles, and also sucker rods used in pumping operation after the vdri'flli`i' iswihpleted. Failures frequentlyoccur in such installations at great depths below the surface and the resulting loss in time and energy is enormous.` This oil well drilling equipment is frequently removed from theA well, as in placing a new drill bit at the lowerend. At such times it would be extremely desirable to remove weakened links or to rearrange them in the setup at positions in which their remaining strength would not be overtaxed. Their blind repositioning in the well without attention to their state of fatigue and in accordance with present methods insures that failure will result and at a position governed solely by chance.

The term fatigue has been applied in recent years to cyclic failures in steel. At an earlier time the term crystallize was used due, primarily, to the appearance of the steel at the point of failure. It is now understood that if metal is cycled at stresses above certain maximums, which are known as the fatigue limit, failure will eventually occur, and that the higher the stress is above the fatigue limit the more rapidly will that failure occur.

Even today the exact nature of fatigue failure is not completely understood. A great many observations have been made but the phenomenon is one which is difficult to observe as there have heretofore been no non-destructive methods of testing to determine the changes in the material. It has been believed that the hardness of the metal increased at the area of fatigue prior to failure. This has been diflicult to determine exactly as a Rockwell or a Brinell test could not be made at the fatigue section as such a test ceding f'atigue failure there is a slippage which occurs across the grains of vtheV metal. This can be observed from photomicrographs vmade of the portion of the steel adjacent the failure. Here again the time over which the slippage occurred and its extent are difficult toy determine as the only method of determination is by the destructive testing of the specimen.

The present invention provides an accurate non-destructive method of determining the na- .grains of the metal.

ture and'extent of fatigue present in the metal being tested. 'If one of the dynamometer indicating instruments is adjusted to'read voltages which are in phase with the current it will necessarily read the watt loss occasioned by eddy currents. Any change in this floss will be occasioned by the change in the internal resistance of the steel since the voltage creating the eddy current remains substantially constant. Accordingly the dynamometer shows the slightest change or variation in the internal resistance o the steel.

A second dynamometer is adjusted to read voltages which are in quadrature to the current. These second voltages are affected mainly by changes in the stresses present in the specimen and, being in quadrature with the current, are unaffected by the changes reflected on the first It has beedobsei'ved in the use of the present invention that there are several distinct stages in the fatigue failure. At first there occurs an increase in the internal stress in the steel Without accompanying change in internal resistance. This gives an apparent change in hardness of the material and is indicated by the. out-of-phase dynamometer, no change being indicated on the in-phase dynamometer. As the fatigue prosresses further, and the hardness continues to increase, a change in the internal resistance appears, indicating strain, that is, slippage in the In the latter stages of the failure this slippage increases at an accelerated Aa new unstressed unit. This meth applied successfully to fatiguedwoi i. 1 1 ed. M

rate and little change in stress is noted. This relationship continues until failure occurs.

The information gained by testing and determining the fatigue level or the build-up in the stress present may be made use of in a novel manner to insure that upon the replacement of those units in operation under proper conditions they may provide a life comparable to the life of besneeu- With an appreciation of the characteristics of metals and of the importance of determining the presence of fatigue therein in order to predetermine failure under load, it is a primary purpose of the present invention to provide a new and improved method and apparatus capable of indicating accurately the presence and the extent of fatigue in metal members.

A still further object of the invention is to provide a method, and an apparatus for carrying out that method, by which metallic units can be tested and simultaneous determinations made of the presence therein of stress and slippage or strain at each point throughout their length.

Still another object of the invention is to provide a new and improved method forgincreasing the operative effectiveness of used and fatigued parts of apparatuses which includes the steps of determining the presence and the extent of the fatigue present and the repositioning of the units relative to each other in their reassembled condition as to insure that the loads assumed will be in inverse proportion to the fatigue present.

A still further object of the of the invention is to provide a method, and an apparatus for carrying it out, by which a runningY record may be made electrically of the strain or breaks which are present, and of the stress which is present, in each unit of length of a metallic member. A still further object of the invention is to provide a method for .testing metallic units to determine fatigue therein, and an apparatus for carrying that method out, in which simultaneous determinations are made of the watt loss occasioned by variations in resistance and the change in voltage in quadrature to the current in the specimen, the specimen having been subjected to a cyclic magnetic eld.

Still another object of the invention is to provide an improved method, and an apparatus for carrying that method out, in which fatigued metallic members are simultaneously tested to obtain variations in resistance, which readings are correlated upon a continuing record to indicate the internal condition existing within the piece undergoing test.

Referring now to the drawings:

Figure 1 comprises a schematic diagram of the apparatus comprising the present invention, an element under test being shown positioned in the testing unit or coil;

Figure 2 is an enlarged showing of a length of the record showing chart A, which indicates the strain or slip between the grains of metal corresponding to a break which may be microscopic in nature, and chart B, which indicates stress in the unit undergoing test; and

Figure 3 is a flow diagram illustrating the removal of rods from an oil well while being tested to determine fatigue, their classifications according to fatigue present therein, and their relocation in the rod line in accordance with their fatigue condition.

Referring again to the drawings, and to Figure 1 in particular, the apparatus comprising the present invention is shown schematically. The exact physical embodiment and details of construction are not of the essence so long as they are such as to enable the unit to perform the functions to be described.

In the drawing a test specimen undergoing test, which may be of continuous length and sectional, is indicated generally by the reference character I0, and is supported for longitudinal movement upon suitable supporting means comprising, in the illustrated form, rollers II. An exciting unit, indicated generally by the reference character I3, is connected to a suitable source of alternating current, preferably sixty cycles, through leads I4, and is formed with two aligned similarly wound coils P through which the test specimen I0 passes, as illustrated. An operator-controlled knob I6 enables the operator to vary the exciting current in the primary windings P and so to control the strength of the magnetic eld created in the specimen undergoing test.

A secondary coil S is positioned midway between the primary coils P; the three coils together form a unit which is positioned right at the test specimen and which may be referred to as the testing coil unit, indicated generally by the reference character I1. Secondary coil -S of the testing unit is connected through leads I9 and a double throw switch I5 to an amplifier unit, indicated generally by the reference character 20. Unit 20 may be eliminated if found unnecessary but is provided for purposes of increasing the sensitivity of the unit. A pair of dynamometers, indicated generally by the reference characters 2| and 22, have their movable coils 23 and 24, respectively, connected directly to the output of amplifier unit 20. In the event the latter is eliminated coils 23 and 24 would be connected directly to the double throw switch I5 and therethrough to the leads I9 of the secondary S. The armature 26 of each dynamometer is suitably suspended, being controlled in its movement by the coils 23 and 24 in the respective units, and

extends into operative relationship to its own chart on a continuously moving paper record sheet 25. Record 25 is provided with two continuous charts indicated by the reference character A for dynamometer 2I and by the reference character B for dynamometer 22. The record roll is suitably supported upon a roll 28 at its upper end and is drawn onto a roll 29 at its lower end after passing the armature 26 which in each instance is provided with suitable marking means capable of placing on the charts a continuous running record of their own displacement. Roller 29 is connected to and is driven by a driving unit 3I connected by suitable means, such as a shaft 30, to the rollers II so that the speed of travel of the record is synchronized with the speed of travel of the elements being tested.

The eld piece of each dynamometer includes a field exciting winding 33 connected through leads 34 to the two-pole concentrated windings 36 of a phase-shifting unit indicated generally, in the case of that unit connected to dynamometer 2|, by the reference character 3l, and in the case of that unit connected to dynamometer 22 by the reference character 38. Each phase shifter is connected through suitable leads 39 to a suitable three-phase source of alternating current o-f the same frequency as the current impressed upon the exciting unit I3 through the leads I4. Each phase shifter 31 and 38 includes a rotatable knob 4I by which the angular position of the two-pole concentrated winding 36 of the connected phase shifter can be angularly adjusted for the purpose of controlling the phase of the secondary output into the leads 34 connected to the dynamometer exciting windings 33. Each knob 4| is provided with a dial 42 calibrated in degrees to indicate the angular relationship of the current and voltage in the secondary.

In addition to the elements which have been described and which comprise the continuously functioning parts of the apparatus, there are provided certain other units which are important for purposes of calibration and of initial setting. It is extremely desirable that a null balance be provided in the circuit of secondary S with a standard, unfatigued, unbroken unit positioned in the test coil unit I'I. With the exciting fields of the dynamometers 2| and 22 properly energized in the manner to be described, and with the marking end of the armatures 26 operatively related with respect to their charts A and B, it is desirable that no indications be given by the secondary circuit of winding S in order that the pointers upon the armatures can be properly related to their zero positions. The presence of a winding 46 in one of the leads I9 leading to secondary S positioned in inductive relationship to a similar winding 4I enables the operator, by the adjustment of knob 48 of a unit 49 to vary the ow, including the phase, of current into the winding 4'I and so oppose, to the extent necessary to render a null balance, the current flowing in the leads I9. The balance unit comprising a combined phase shifting and resistance unit 49, is provided with alternating current through leads 5| of the same frequency as that impressed upon leads |4 of exciter unit I3.

Before beginning a test it is important that the current flowing through exciting windings 33 of dynamometers 2| and 22 be so adjusted that the current in coil 33 of dynamometer 2| is in phase with the current impressed through leads I4, while the current flowing through winding 33 of dynamometer 22 is in quadrature therewith. To obtain this result and to make possible the necessary adjustments akpair of leads 56 are connected across a voltage drop 51 in the circuit of primaries P and are connected to a pair of fixed contacts of double throw switch I5. With the latter positioned as to conduct current from leads 56 to amplifier 2i) voltage which is in phase with the initial exciting current is impressed upon the movable coils 23 and 24 of dynamometers 2| and 22, respectively. As dynamometer 2| is to measure in-phase voltage changes at this time, and While its movable coil 23 is connected as just stated, the phase shifter 3'I is adjusted by its movable knob 4I until the current in exciting winding 33 is in phase with the current in movable coil 23. This relation is present when the right-hand displacement or deection of armature 26 is a maximum. Then the dial 42 at the adjusted knob 4| is adjusted to give a zero reading.

Turning now to the calibration of dynamometer 22, its phase shifter 38 is adjusted until the current in its coil 33 is in phase with the current in movable coil 24 at which time the armature 26 will have a maximum deflection to the right. If then the dial 42 on its knob 4| is adjusted to zero and thereafter the knob 4| turned through ninety degrees to a reading of 270 the phase mometer exciting winding 33 would be in quad-4 rature with the current in the movable c oil 23. In this adjustment following such a step no further adjustment would be necessary in the case of dynamometer 22, while in the case of dynamometer 2| the knob 4| would be rotated through ninety degrees, as indicated by the dial 42, so as to bring the current in exciting coil 33 from quadrature relationship to in-phase relationship with the current in movable coil 23.

In the operation and use of the present invention let it be assumed that the apparatus is positioned atjjne oil wellM d sucr rods are to be withdrawnantested. The es clfunit II is positioned at the top of the well so that the rods will travel upwardly threthrough as they are pulled from the well. The remainder of the apparatus may be positioned at a distance and connected suitably through cables. In one preferred form the apparatus is mounted in an automotive truck positioned as close as convenient to the well.

Before the pulling operation begins a standard test specimen is placed in the testing unit coil I'I, a null balance is obtained by the adjustment of the resistance phase-shifting unit 49, and the phase Shifters 3l and 38 are adjusted so that current flowing in exciting coil 33 of dynamometer 2| is in phase with the exciting current, while the current flowing in exciting coil 33 of dynamometer 22 is in quadrature with it, as previously described. The driving unit 3| of the recording unit is suitably connected to the rollers II of the test unit so that the travel of the charts will be synchronized with the movement of the rods. All is then in readiness for the test.

In the test itself the operator need only check to see that the various parts are performing their intended functions. Electrical adjustments having been made as stated the unit is self-operating, and as the rods are pulled from the well and through the test coil unit II a continuous record is made in charts A and B. As the apposed portions of the two parts indicate the characteristics of identical parts of the rods being tested it is possible by reading the charts together to determine the exact condition existing in each rod.

Referring to Figure 2 in particular, a portion of the running record comprising charts A and B is shown. It is to be remembered that chart A indicates the in-phase voltage, while chart B indicates the quadrature voltage as determined by dynamometers 2| and 22, respectively. Defiections in the two graphs which are opposite each other comprise indications taken at the `same position in the elements being tested. In

curve A the normal straight-line portion of each curve comprises the null balance point, the deflections upward, corresponding to movement in Figure l, indicate points of higher electrical resistance which may be produced, and in the cases of points X and Y, are produced, by flaws or fractures. In the case of chart B it is noticed that directly opposite points X and Y of chart A are points L and M, respectively, which indicate stress and fatigue in the rod being tested at, the exact point at which a fracture has occurred as indicated in chart A. The maximum deflections indicated by the points O on both the charts are without significance as they represent variations resulting from the rod joints passing through the testing unit. Attention is particularly directed :to point R in chart B which is a point of maximum deection in a long, gradual curve. This represents a rod which is fatigued throughout e 1ateasegestiegene@., e maximumugle be'ifivgA at point R. Nowa'ture has occurred in the .nod as yet as indicated by the absence of any deflection to the left on chart A opposite point R. Aged of this type has been greatly weakened by fatigue and will fracture if unduly loaded. However, if no fracture is present and the rod in its future use is subjected to much lighter loads it can continue to function just as satisfactorily as a new rod.

The rods are pulled continuously from the well during the operation and each rod is charted as to its characteristics in the running record, as described. As the rods are pulled from the well they are disconnected one by one and removed to storage racks, as illustrated diagrammatically in Figure 3, bearing the reference characters 5Fl, 60, 10, and Si). Rack 50 is to receive rods substantially unstressed, unstrained, and in almcst perfect condition. Rack Gt will receive the next classification of rods and so on until the rods which have entirely served their usefulness and are so defective as to be entitled to no further condence will be placed in rack 99. Those rods placed in rack will not be used again, but the rods classified in racks 5B, 6d, iii, and 8S will be used again and will be so positioned in the well that the loads to which they are subjected will not be in excess of their strength. As the rods at the upper end of the well are subjected to the greatest loads it follows that the poorest reuseable rods, those in bin 8B, will be taken iirst and will be repositioned lowest in the well to be followed by the rods in bin it, then the rods in bin 60, and nally the rods in bin 50 which will be connected to new rods replacjng those discarded in bin Sil.

This classication of rods as they are withdrawn from the well and their repositioning, a movement indicated by the arrows in Figure 4, makes possible the intelligent and safe reuse of fatigued rods. Were the rods to be reused without being classied and repositioned according to strength unnecessary failures would occur in many instances, for the weakest rods would sometimes be positioned in the line at points in which they would be subjected t0 stress in excess of a safe maximum in View of their weakened condition. Under this repeated stress they would fail whereas if positioned where they would notI be overstressed they might have lasted indeiinitely.

While the particular apparatus herein shown and described in detail is fully capable of attaining the objects and providing the advantages hereinbefore stated, it is to be understood that it e. is merely illustrative of the presently preferredl embodiment of the invention and that no limitations are intended to the details of construction or design herein shown other than as deiined in the appended claims.

I claim:

1. In an apparatus for determining fatigue damage in a magnetizable element comprising a rst coil to impress an alternating current field upon an element being tested and means to indicate the presence and extent of fatigue damage, said means including a second coil positioned in inductive relationship to said element as to have induced therein a current flow by the magnetic eld of said element, a pair of dynamometers each including an indicating armature coil connected to said second coil and an exciting field coil, current sources each including an operator adjustable phase shifter connected to said exciting eld coils to supply current in phase with the current in said armature coils to one of said exciting coils and in quadrature to' the current in said armature coils to the other of said exciting coils, and an adjustable combined phaseshifting-and-resistance unit connected in inductive relationship to the circuit of said second coil and said armature coils to obtain a null balance for said armature coils with a standard unfatigued specimen element operatively positioned relative to said iirst and second coils.

2. A method of servicing mgltiplelwengmmsumggr;

M. wf

ngLsydod from the well, imprgainsnmternating current magneticiieldupon said rod progressively`alongrit'slength as it is withdrawn, simultaneously inducing into a fatigue indicator unit a current flow by the magnetic iield of said rod, rrecording on a continuous record synchronized to the withdrawal of said rod the voltages which are in-phase with and in quadrature with said indicator current to indicate the fatigue in the individual length of rod, ls'l-ggr;gjgin,gathe individual lengths of said rod having generally similar fatigue indications into groups, and replacing said rod in the well by groups with the d rwnirrmsenavnrgtnianmmt fatig u e a t thejitomend-the fgroup having the arraigadas@ agire-mau- 3. i'ethod of servicinghrngllliplelllgth.suc er rods in a well comprisingthe steps `of withd aww-l nating Ycurrent magnetic field upon said rod progressively along its length as it is withdrawn from the well, inducing a flow of current in a testing circuit by the magnetic eld present in said rod at a particular point past which said rod is moved in being withdrawn, recording on a first continuous record the voltage of the induced current in said testing circuit which is in phase with the current, simultaneously recording on a second continuousirecord the voltage of the induced current in quadrature with said current, said rst and second continuous records being synchronized with each other and with the rate of withdrawal of said rod from the well, .coniparr` ingdsaid voltage records to determine the fatigue in the individual lengths of said rod, classifying andsegregating the lengths of said d having substa'tiall'y` the same fatigue records into groups, angLreturning s aidlengthsrof rod to said well, the lengths of rod of the group having least fatigue being positioned in the well at the height therein at which maximum loads are encountered while the lengths of rod of the groups having greater fatigue are positioned in the well at heights such that the loads thereon are proportioned in accordance with the likelihood of fatigue failure.

4. The method recited in claim 3 characterized in that the lengths of rods in the groups having the smallest in-phase voltage are discarded.

5. The method recited in claim 3 characterized in that the lengths of rods for which the inphase voltage records disclose minimum voltage at one or more points along/each length are discarded.

6. The method recited in claim 3 characterized in that the lengths of rods for which the inphase and quadrature voltage records indicate substantially constant voltages for the entire length of rod are repositioned in the well as to assume greater than minimum loads.

7. The method recited in claim 3 characterized in that the lengths of rods for which the quadrature voltage record discloses substantial variations from an established norm are reassigned to heights in the Well at which less than maximum loads are present.

EMMETT M. IRWIN.

References Cited in the file of this patent UNITED STATES PATENTS Number 5 1,686,815 1,813,746 1,897,634 1,906,551 1,933,547 10 2,098,991 2,124,579 2,290,330 2,405,859 2,435,985 15 2,507,863

Name Date Imes Oct. 9, 1928 Kinsley July 7, 1931 De Forest Feb. 14, 1933 De Forest May 2, 1933 Drake et a1. Nov. 7, 1933 Zuschlag Nov. 16, 1937 Knerr et al. July 26, 1938 Irwin July 21, 1942 Somes Aug. 13, 1946 Stewart et a1 Feb. 17, 1948 Mesh May 16, 1950

Images