Permeability Measurements Citation Morse, Virgil Franklin (1914) Permeability Measurements. Bachelor's thesis, California Institute of Technology. doi:10.7907/bs65-wp63. https://resolver.caltech.edu/CaltechTHESIS:06072024-215159715 Abstract No abstract. Item Type: Thesis (Bachelor's thesis) Subject Keywords: (Electrical Engineering) Degree Grantor: California Institute of Technology Division: Engineering and Applied Science Major Option: Electrical Engineering Thesis Availability: Public (worldwide access) Research Advisor(s): Unknown, Unknown Thesis Committee: None, None Defense Date: 1 January 1914 Record Number: CaltechTHESIS:06072024-215159715 Persistent URL: https://resolver.caltech.edu/CaltechTHESIS:06072024-215159715 DOI: 10.7907/bs65-wp63 Default Usage Policy: No commercial reproduction, distribution, display or performance rights in this work are provided. ID Code: 16511 Collection: CaltechTHESIS Deposited By: Benjamin Perez Deposited On: 17 Jun 2024 21:12 Last Modified: 17 Jun 2024 21:12 Thesis Files PDF - Final Version See Usage Policy. 6MB Repository Staff Only: item control page

Thesia Permeability Measuremants by Virgil F. Morse Class of Nineteen Hundred and Fourteen Department of Electrical Engineering THROOP COLLEGE OF TECHNOLOGY Pasadena, Cal. - 1914 - PERMF..ABIL I TY MEASUREMlii"'NT S Thesis Class of 1914 1 PERMEABILITY MEASUREMENTS. The problem treated in this paper is the measurement of the permeability of iron and steel in straight bars. After the permeability has been determined over the entire length of the bar it may reveal various kinds of mechanical or structural differences in the steel which otherwise could not be detected without destruction of the specimen. The apparatus used in making these determinations was: (l) a Permeameter constructed by Mr. R. w. Parkinson and Mr. Ray Gerhart of the class of 1913 TiffiOOP COLLEGE OF TECHNOLOGY according to the specifications of the United States Bureau of Standards as set forth in Bulletin Vol. 6 No. 1 under the heading, "'.l.'he Determination of the Magnetic Induction in Straight Bars", by Mr. Charles W. Burrows, Assistant Physicist: (2) a Leeds and Northrup Deflection Potentiometer for measuring the current accurately: (3)• a Storage cell of about 2.1 volts for the Potentiometer: (4) an Fdison five cell 6 volt alkaline battery for pro- ducing the current in the magnetizing coils and in the primary of the Mutual Inductance: (5) a Ballistic Galvanometer for obtaining balances between inductions as subsequently explained: 2 (6) a Switchboard with Mercury contact reversing rocking switches for manipulating the currents in the various circuits: (?) a Mutual Inductance of .1013 henries which I construct- ed for these experiments. In making the determinations of the permeability the first step is to get the specimen into a "cyclic" condition and thoroly demagnetized. This is done by reversing the current in the magnetizing coils about twice per second beginning with a value high enough to carry the induction well beyond the point of maximum permeability, and gradually reducing the current to a value slightly below the lowest induction to be measured. It may be necessary to go thru this operation of demagnetization several times before the steel reaches a cyclic condition such that a reversal of the magnetizing current reverses the direction of magnetization without changing its magnitude. After the specimen has been reduced to a cyclic condition a uniform magnetic flux is obtained thruout the specimen by balancing the induction of a test coil around the test specimen against the induction of a similar test coil around an auxiliary rod. This shows equal flux on the two rods and the final adjustment is made by passing sufficient current thru small compensating coils on the ends of the rods so that testcoils show equal flux thruout the test rod. 3 When this condition is obtained the test coil is connected in series with the secondary of the mutual inductance. The primary current is adjusted until the induction of the test coil equals that of the mutual inductance on reversal of their respective primary currents. Then readings are made of the current in the primary of the mutual inductance and in the magnetizing coil on the test specimen. These readings are made with the Deflection Potentiometer and are easily obtainable to one-tenth of one percent. The mutual inductance was carefully calibrated and found to have a value of .1013 henries. Then B = 790000 x .1013 x Im - .318H, where Im= the current in amperes in the mutual inductance primary; H = 39.5 x It, where It= the amperes in the magnetizing coil of the test specimen; J4,:: B/H = 2026 x Im/It - . 318 In ordinary testing the constant .318 may be neglected, tho in very careful determinations at low values of/6' it should be considered. For further definite information concerning the constants of the pe:nneameter, reference should be made to the thesis of Mr. Ray Gerhart and Mr. R. W. Parkinson of THROOP COLLEGE OF TECHNOLOGY 1913, which is on file at the Co11·ege. 4 With a variable mutual inductance of about 100 millihenries maximum, such as designed by Mr. Burrows and set forth by him in his bulletin, the work of making the measurements of penneabili ty would be greatly simplified. With such apparatus at hand, the primary of this mutual inductance would be connected in series with the magnet- of izing coil on the test specimen, and a balanceAthe induction in their secondaries obtained by varying the mutual inductance. The~ = 20000 M - . 318, where M = the value of the mutual inductance in henries. This variable mutual inductance could then be calibrated to read directly in terms of_,J,c,, - the perm.eabili ty. Improvements desirable are as follows in the order of their im~ortance: (1) a variable mutual inductance of about 100 millihenries maximum, carefully calibrated to read both in millihenries and in terms of permeability when used with the THROOP COLLEGE permeameter; (2) sufficient storage battery capacity to give 3 or 4 amperes continuously without an appreciable drop for one minute: two F.dison 5-cell, 6-volt batteries should do this; (3) a separate mounting for the Galvanometer so that it will not be jarred by reversing the switches. 5 The set-up and operation of the TF...ROOP COLLEGE Pe:nneameter. The following drawings are taken from the thesis of 1 913, to which reference is made for more comi:1lete data oh the construction and calculation of the constants of the permeameter. Fig. 10 shows the positions of the windings; Fig. 12 shows the diagram of wiring connections; Fig. 13 the internal connections of the Permeameter; Fig. 14 the connections on the mercury contact S',Vi tchboard. I Connection to the Potentiometer are made thru shunts placed in the primary circuit of the muroual inductance "M" Fig. 12, and in the circuit "T" Fig. 12, which is the magnetizing coil around the test specimen. Care should be taken to avoid measuring the current in both "T" and "A" Fig .12, which are in multiple. Im = amperes in circuit "M" Fig. 12 It = amperes in circuit "T" Fig. 12 M ll; henries of mutual inductance "M" Fig. 12 = .1013 K1 H: K3 It r : B/H ~ : 2026 = 790000. K2: .318 K3 ~ 39.5 = MizriK1/K3It - K2H/H = 2026 Im/It -.318 Im/It approximately. In Fig. 1 2 the batteries Bt, Bj and Bm may be all taken as one but the Potentiometer battery must be separate. 6 When all the wiring connections have been properly made and the test and auxiliary bars have been placed in the Permeameter with the yokes firmly clamped, the test is ready to be started. First demagnetize the test bar by applying a cyclic magnetizing force of about one period per second, which is gradually reduced from an initial intensity which carries the induction well beyond the point of maximum permeability to a final value somewhat below the lowest induction to be studied. It was found that the initial intensity of the magnetizing force was sufficient with a current of about three { 3) amperes in circuit "T". Now set the switch K3 Fig. 14 on t-m and after demagnetization the lowest force to be used is applied and repeatedly reversed, by rocking the reversing switches s.T. and S.J. Fig. 14, until the test rod is brot to a cyclic magnetic state. To ascertain when the cyclic state has been obtained, close the Galvanometer switch S.G. and note the deflection. If it is e~ual in magnitude on reversal of the current in either direction, the cyclic condition has been reached. Now move K3 to t-a and adjust the resistances R.A. and R' .A. for uniformity of flux. This may be done while bringing to the cyclic state as it involves only very small changes in the magnetizing force. 7 If the connections have been properly made the test coils "t" and "a" will be in series-opposition such that on rever-sal of the current in the magnetizing coils "T" and "A" the current induced in "t" will be exactly neutralized by that in "a" when the flux is the same in both bars and will be observed as a zero deflection of the Galvanometer "G". Now move K3 to t-j and adjust R.J. until uniform- ity of flux along the test rod is obtained as above. This operation consists in passing sufficient current thru the mompensating coils Jt, J't, Ja and J'a Fig. 13, to compensate for the magnetic flux leakage at the magnetic joints and for the effects of the yoke. Having made the adjustments necessary to secure uniformity of flux and brot the steel to a cyclic state, the induction is measured by balancing the electromotive force induced in "t" on reversal of the magnetizing current against that induced in the secondary "m" of the mutual inductance on reversal of the current in its primary "M", This current is adjusted to give a zero deflection on the Galvanometer "G" when K3 is on t-m and the switches S.T., S.J. and S.M. are reversed simultaneously. In order to exclude any influence due to mechanical vibrations the entire magnetic circuit is mounted on a layer of felt. These lower values of the induction are produced by small magnetizing forces and therefore are easily 8 altered by incompleteness of demagnetization or failure to reduce to a cyclic state. As a check, therefore, after determining this first point on the induction curve, it is well to carry the steel again thru the demagnetizing process, readj ust the compensation, reduce to a cyclic state and redetermine the point. If this second determination does not yield the same result as the first induction, the full process should be repeated more carefully until concordant results are obtained. Having thus obtained s a tisfactory data for the lowest magnetizing current it is not necessary to demagnetize again. The remaining points on the induction curve may be secured by passing to the next higher magnetizing force, adjusting the main and compensating currents, repeating the reversals till a cyclic state is reached and so on till the required data are secured. The curve determined by these points, the Band H, is called the Normal Induction Curve. It was found to be much more satisfactory to get the final balance between the test coil "t" and th :," mutual inductance "m" by Jhhe deflection than by the zero method. That is, to read the deflection of the galvanometer on reversal of S. T. and S.J. with K3 on t-m and then get an equal deflection by reversing S.M. The reason that the direct zero method was found to be not so accurate as the 9 deflection method was that the discharge of the mutual inductance was practically instantaneous, whereas there was a noticeable time-lag of the discharge of the test coil, due to the steel or iron in the circuit. This lagging E.M.F. would cause some deflection of the galvanometer after the influence of the mutual inductance was past. Another difficulty was that it was practi~everse the theee switches S.T., S.J. and S.M. absolutely simultaneously, so that the galvanometer would be deflected in one direction by the circuit first reversed and then in the opposite direction when the other WI\S reversed. With a mutual inductance of fixed value, readings thruout the tests are made with the Deflection Potentiometer of the current "Im" in the primary of the mutual inductance and of the current "It" in the magnetizing coil of the test specimen only, not the current in the auxiliary circuit. With the constants of the machine and the currents "Im" and "It" known, data can be compiled and curves plotted showing any desired relation between B, H, and;c. A very careful test was made of a piece of Stubbs' Drill Rod, a high grade tool-steel, and a curve plotted at all densities, the B?' curve, of the rod as it came from the shop and then the same specimen was heated to redness at the center in a ·Bunsen flame. After cooling slowly in air this rod was again tested and marked differences in the 10 permeability noted, the maximum bei~ 100.,% greater. curves and data sheet are submitted as a part of this thesis. A long rod of Bessemer steel was tested and then stressed in tension at one end beyond the elastic limit. A test now shows marked diminution of the penneability in the stressed portion al.tho the unstressed end was unchanged. It was hoped at the beginning of this work that the Permeameter might be a good detective and the results have ·justified the assumption. In the case of the Stubbs' drill rod the Permeameter revealed very positively the fact that there was a marked difference in the magnetic properties of this 15" specimen of which only about 2" at the center had been heated to redness, and the Bessemer rod showed distinctly in which portion it had been strained. It is a well known fact that stressing steel beyond its elastic limit makes it harder and that heating to redness with subsequent slow cooling softens it, and now the Permeameter shows that hardening stebl decreases its magnetic permeability and that softening steel increases the permeability. Therefore the Permeameter can be used to detect either hard or soft spots in a rod which might appear unifonn. Mechanical or structural flaws ,vould probably also be detected but as yet I have not done a sufficient number of experiments to make this certain. BESSEMER STF,EI., DATA SHEET Original Im It Stressed H Im It H B ;V .0360 .090 3.56 2880 812 . 0195 .12 4.74 1560 330 .0425 .10 3.95 3400 863 .023 .13 5.14 1840 360 • 0495 .105 4.15 3960 957 .026 .14 5. 54 2080 378 .0545 .115 4.54 4360 963 .03 .15 5.92 2400 406 .0640 .123 4.86 5120 1055 .0325 .16 6.33 2600 413 .0690 .132 5.21 5520 1060 .038 .18 7.04 3040 434 .0895 .163 6.44 7160 1116 .0465 .21 8.30 3720 450 .130 .225 8.89 10400 1170 . 0666 . 29 11. 46 5328 466 .155 .30 11.9 12400 1048 . 40 15. 8 7200 458 .174 .40 15.8 13920 883 .124 . 60 23. 7 9920 418 .198 . 76 30. 15840 528 .147 .81 32.0 11760 468 .224 3.25 128. 18920 140 .195 1.88 74.3 15600 211 .205 2.35 93.0 16400 177 .212 3.25 128.4 16960 132 The data under "Original" gives the magnetic properties of the Bessemer steel rod as it came from the shop, and that under "Stressed" shows the magnetic properties of that portion of the same rod which had been stressed beyond the elastic limit in tension. STUBBS DRILL ROD DATA SIIBET Original Annealed Im It H B / .004 .11 4.35 320 .005 .13 5.14 .010 .24 . 0135 Im It H B /' 73.2 .0056 .11 4.35 448 103 400 77.6 .0072 .13 5.14 576 112 9.48 800 84.0 .0085 .15 5.92 680 115 . 305 12.05 1080 90.0 . 017 .235 9.3 1360 146 .0238 . 402 15. 9 1900 119.5 .036 . 335 13. 2 2880 220 .0305 . 481 19.0 2440 128. 5 .0845 . 452 17. 9 6760 378 .0343 .500 19.75 2740 139 .088 . 470 18. 6 7040 379 .042 .569 22.5 3360 149 .0905 . 49 7240 374 .048 . 612 2.. 4,2 3840 159 .0935 . 515 20. 3 ' 7480 368 .064 .72 28.4 5120 180 .101 . 58 22.9 8080 35 2 . 070 .77 30.4 5600 184 .113 .7 27.6 9050 327 .079 .825 32.6 6320 194 .117 .75 29.8 9350 313 .0865 .898 35.5 6920 195 .122 .80 31. 7 9760 308 .092 .95 37.5 7360 196 .129 •9 35.5 10300 290 .1065 1.1 43.5 8520 196 .134 1.0 39. 5 1.0700 272 .1213 1. 265 50.0 9700 195 .149 1. 29 51. 11900 235 .136 1.48 58.5 10880 187 .158 1.53 60. 5 12640 210 .146 1.73 68.3 11680 172 .179 2.5 98. 8 14320 145 .158 2.01 79.5 12650 160 .182 3. 118. 5 14550 123 .164 2. 21 87.4 13120 150 .186 3.5 138. 2 14880 108 .169 2.43 96.0 13520 141 The data given here show the .178 2.98 117.8 14240 121 magnetic properties of the rod .182 3.5 138. 2 14550 105 before and after annealing. 19. 3 Steel Core of the PACIFIC LIGHT and PO'lTER Corp. BIG CREEK Transmission Line Cable. Im .0020 It .155 H 6.12 Bi B2 160. . 0030 .23 9.10 . 0040 . 30 . 0060 DATA S:F!llli"'T. 143. 2 B3 140. 4 fa 240 . 215.0 210. 9 265 • 11.85 320. 287.4 282.1 271 . . 33 13.00 480. 444.3 438.4 369 . . 0075 .39 15.40 600. 557.7 550.7 390. . 0080 • 46 18.15 640. 590.0 582.0 353 • . 0115 . 54 21. 30 920. 861.4 851.8 433 . . 0160 .63 24.85 1280. 1211.? 1200. 5 515. .0205 .66 26. 05 1640. 1568. 4 1556. 6 631. . 0400 . 82 32.40 3200. 3110.8 3096.3 990. .0470 1.05 41.5 3760. 3646.0 3627.3 910. . 0620 2.4? 97.6 4960 . 4691. 0 4648.0 510. .0665 3.33 131.5 5328. 4966.0 4907.0 405. 259. This test was made on a "seven strand plow-steel cable" taken from the core of the Aluminum cable used on the BIG CREEK Transmission Line. Area - 78500. cir. mils - .3978 mn. B:i_: Mim1{1: 80000.Im B2 ;:. Bi_ - K~H H: 39.5It Is_~ 790000. K A1 : A2: .3978 a=: 1.6698 .518 K 11 2 :. (a - A2 )/A ~ 3.2 2 \ B3 : BJ. - K 2H 2 =(a - A1)/A1: 2.75 A 2 was the actual area in sq. cm. was the area calculated taking the radius over all ;.406 The curves plotted were B1 H, B3H, ~ • and B~. . -·- ' ---. -- iI • I I .-- ·· -·-i - - ·· : --- ~ i ~ ! ... (;, - I I _ ; ·_.:. __ ) _ _ i ···-··i---- It : --~ -, i • i -· -•- . ·--' ·. -- --- --K'- "' --ep r - i !_~_ -i· ' I ' · ·-1·----: -- r : r , ··,· --·-- --- ·r --- -·-:i p I ! . ; ; i. ---: •• •• r-· r· ··1 T ! I I ~- c.. 0 ~ B A1/o/ine's i ~- '~ . I o 11 I I . i___b,__ I,_ P -- 1' ! -1 --: ;··· ' I - -r--T-1 ··'·· -· · ' . --- r-·· -- --· :.. i - r-· 'I ·--- f -- - I ·· -- · - I ·' i • i • I V. I I .. ···!-- __ ;___--- ;--- - ➔- . . - -,' -- - --- '' ' l . ! ·i -- - -- j-······· l.. Ii .. I' 'I -! . -l :..\ .. .i