Creator:Alexander Zeitlin and Jacob Brayman Date Created:September 18, 1962 Place Created:New York, New York Keywords:cubic workpiece configuration,ulta-high pressure calibration Context:article publicated by ASME ************************************************** Ultra-High Pressure Calibration: Influence of Cubic Workpiece Configuration ALEXANDER ZEITLIN Barogenics, Incorporated, New York, N. Y., Mem. ASME. JACOB BRAYMAN Barogenics, Incorporated, Ne* York, N. Y. This paper reports on two investigations conducted on a large size cubic unit. The first investigation confirms the advantages of large size specimens by disclosing certain phenomena which have not been observed on smaller size units. The second investigation indicates that the efficiency of pressure transmission in a cubic unit depends on the geometry of the specimen assembly. The optimum geometry varies with the pressure level. The investigation reports on general results established experimentally. Since the apparatus and instrumentation were of substantial importance, their description is included in the report. Contributed by the Research Committee on Mechanical Pressure Elements for presentation at the Winter Annual Meeting, New York, N. Y., November 25-30, 1962, of The American Society of Mechanical Engineer. Manuscript received at ASME Headquarters, September 18, 1962. Written discussion on this paper will be accepted up to January 10, 1963. Copies will be available until October 1, 1963. 1 PER COPY 5 O C TO ASME MEMBERS The Society shall not be responsible for statements or opinions advanced in papers or in discussion at meetings of the Society or of its Divisions or Sections, or printed in its publications. Discussion is printed only if the paper is published in an ASME journal. Released for general publication upon presentation THE AMERICAN SOCIETY OF MECHANICAL ENGINEERS 345 East 47th Street, New York 17, N. Y. Printed in U.S.A. Ultra-High Pressure Calibration: Influence of Cubic Workpiece Configuration ALEXANDER ZEITLIN JACOB BRAYMAN NOMENCLATURE La = length of edge of square anvil tip, in. Ity = length of edge of square face of specimen assembly, in. L = length of resistance wire, in. D = diameter of silver-chloride insert, in. P = pressure in primary system of machine, Kllobars (kb) m = magnification factor; ratio between area of primary ram and area of anvil tip for La = 1.46 m = 196 for La = 1.35 m = 227.5 Pm = p x m Pr = transition pressure INTRODUCTION Fig. 1 2700/16,200-ton cubic hinge unit and operating console conduct two investigations which are the subject of this paper: 1 The purpose of the first investigation was to determine whether the large size of the specimen offered any increase in observability of ultra high-pressure phenomena. 2 The second investigation concerned the determination of the influence of geometric relations between the workpiece assembly and the anvils of the apparatus on the efficiency of pressure transmission. The tests were conducted by Messrs. J. Bray-man2 and M. Qabey,^ under the general supervision of Dr. G. Gerard (8). The use of phase transitions for the calibration of ultra high pressure apparatus was suggested by Bridgman (1,2,3)Further contributions to this problem were made by Bundy (4,5) Vereshchagin et al (11,12) Kennedy and LaMori (13) and Boyd and England (10). All cited investigations were conducted either with the assistance of Bridgman anvils or on cylinder-plunger type apparatus of various design. Hall (6,7) has suggested the use of multiaxial arrangements. A review of various multiaxial arrangements will be found in reference (9). TCie testing and calibration of one particularly large cubic unit offered the opportunity to Numbers in parentheses designate References at the end of the paper. APPARATUS The machine used In the investigations is shown in Fig.l. It consists of six heads or platens connected with each other by means of articulating links. Each head contains a hydraulic system consisting of a cylinder (integral bore within the forged head) and a ram. The total hydraulic force developed by the ram Is concentrated by means of a tool assembly attached to each ram 2 Senior Development Engineer, Barogenics, Inc. 3 Chief of Field Services, Barogenics, Inc. Vice-President, Barogenics, Inc.; now Director of Engineering Sciences, Allied Research Associates. 1 . 2 2700/16, 200-ton cubic hinge unit Fig. 3 High pressure hydraulic schematic on a comparatively small face of a tungsten-car-bide anvil. The six carbide anvils press upon the six faces of a cubic specimen assembly. Pig.2 shows the machine with the front platen removed. The hydraulic circuit is shown in Pig.3. Pump 1 provides the prefilling of the system up to a pressure of 2500 psi, pumps 2 and 2a are used to raise the pressure up to the maximum of 13,000 psi. Each ram is rated at 2700 tons at a primary pressure or 13,000 psi. The special design of the hydraulic seal, Pig.reduces the friction to an amount which can be considered negligible within the accuracy of measurements reported here. Anvils Tungsten-carbide anvils used in the investi- 2 Table 1 Parameters of Test A-ssemblies a) Design La Lw Lw (in.) (in.) La Used for Transitions Fig. 6 (straight wire) 1 4G 1 61 1 10 Bi I-H Bi ii-m - 1 75 1 20 It " Th II-111 1 U2 1 25 II it " 1 90 1 30 II " " Ba n-IIl - 1 97 1 35 It tt ii it 1 98 1 36 - n tt 2 27 1 54 - 2 50 1 71 - it 1 35 2 30 1 71 11 " - - Bi V-VI 2 so 1 85 tl II _ _ ii Fig. 7 1 35 2 50 1 85 tl ii •» (coiled wire) b) La Lw L La D La (in.) La (In.) L (in.) D 1.46 .615 2.4 .25 6 1.35 1.71 .675 2 .675 2 1.35 1.85 •i 2 .56 2.4 1.35 1.85 1.95 .75 La - Ed:-c of Anvil Tip Lw - Ed^e of Specimen Assembly L - Length of resistance wire D - AgCl Diameter /VP £OAD LOAD£~£> Fig. 4 Deflection compensating seal Fig. 5 Anvil assembly gation are shown In Pig.5- The high-pressure tip Is a square. Two sizes were used In this investigation: a square tip with 1.46 in. edge length (2.14 in.2) and a square with 1.35 in. edge length (1.82 in.2). Specimen Assemblies Kiese are shown in Pigs.6 and 7• They consist of a cubic outer layer of pyrophyllite, a cylindrical insert of silver chloride and a sensl-tive-element assembly. She sensitive-element assembly, Pig.6, consists mainly of a straight piece of wire while Fig.7 shows a silver-chloride cylinder with a wire wound on it in the form of a spiral. Ihe contacts between the sensitive element and the outer portion of the electric circuit are provided in a conventional manner by means of tabs abutting against the tips. Fig. 6 Conventional sample assembly r np.- ^ ' LfJlgrH of \£AL/B/?AT'OM IV/A'EJ fl/' OF TUK'/Ci' = 0.0002 ohm = 0.001 ohm = 0.005 ohm 0 .024- ohm 0.12 ohm 0.60 ohm The repetitive accuracy of readings was better than i 1A percent. (c) In general, the pressure and resistance Pressure Range Table 2 Residual Lw, in. T1 II-III and lower........ 1.75 Ba II-III and B1 V-VI........... 1.71 Residual Volume, cu in. 5A 5.0 Table 3 Volumetric Transitions of Bismuth Fig. 7 Coiled sample assembly Table 1 indicates the construction and the dimensions of specimen assemblies used in various test runs covered in this paper. METHOD AND INSTRUMENTATION Method The method of investigations consisted of measuring the variations of electrical resistance of metals as function of primary pressure and determination of discontinuities (phase transitions). The total initial resistances were as follows: Bismuth (straight wire) in order of 0.05 -0.1 ohm Bismuth (coiled wire) in order of 0.250 ohm Thallium (straight wire) in order of 0.002 ohm Barium (straight wire) in order of 0.007-.04 ohm Instrumentation (a) The primary pressure was observed by means of a l6-in. pressure gage. This precision Bourdon gage had a range of 0-15,000 psi. One division of the scale was equivalent to 10 psl. The gage was equipped with a knife-edge pointer so that parallax was eliminated. Careful calibration in the manufacturer's plant assured accuracy being at least equal to one division or + 0.133 percent of the range. (b) The resistance of the circuit was measured by means of a double Wheatstone bridge in which the compensating resistance is adjusted for a zero galvanometer reading. Of the available eight ranges, the following ones were used: Transition Pressure* kb ** AV ** kg/cm2 Vo i-n 25,300 24.8 4.60% n-ni 27,000 26.5 2. 95% rn-rv 44,800 44.0 .6% iv-v 65,000 63.8 .5% V-VI 89,800 88. 1 1.2% * Bridgman data ** Computed from lirid man data. gages were observed visually and simultaneous readings taken by observers. However, the availability of appropriate transducers allowed the use of an X-Y recorder which was in operation during a number of runs. (d) The size of the specimen assembly was measured before and after each run by means of a micrometer. It is interesting to note that for a given structure of the specimen assembly the residual size (after being subjected to pressure) was primarily the function of the anvil size. It depended to a small degree only on the initial size of the specimen assembly. For instance, for La = 1.35 the residual length Lw is given in Table 2. FIRST INVESTIGATION (BISMUTH DISCONTINUITIES) Bridgman observed five volumetric transitions of bismuth shown in Table 3. Previous electrical measurements by Bridgman and Bundy disclosed only three discontinuities. Of these, the first two occurred at pressures identical with those for the first two volumetric transitions, while the last one occurred at pressures in the neighborhood of 125 kb. Kennedy and LaMori pointed out that the high electrical bismuth transition probably corresponds 4 &/SMum -r^fjs/r/aAj 2000 TOA/S CC4S/C UAJ/T A/vv/L eece. /3S" pr/scrwt-jj7F EP6£ -33/ " MAG/s/r/CA-r/c*s ^Across. 2.27.5 S/LVZG CtfUX/PE SL££V£ test KUAJ; APe. /S62 30 35 to SO 35 £>c> e,s TO 73 &o 35 SO Fig. 8 Bismuth transition curve K6 Table 4 Bismuth Transitions Data uaed In the determination of Bismuth Transitions. Pm* during experimental runs Transition Reference Pressure kb Experimental Runs ^ Volumetric.. (Bridgman) Resistance (Bundy)'4' I n m I-U 24. B 25.4 36.8 43.4 37.7 n-rn 26. 5 27.9 42.2 47.9 42.5 in-rv 44.0 — 82.0 82.4 79.4 rv-v 63.8 — 119.2 119.5 117.0 v-vi 88.1 176 178.0 180.0 vi-vni J 89.0 * Pm : p I it. - where p = primary pressure in kb m = magnification factor m area of main ram area of anvil tip Table 5 Volume and Resistance Changes for Bismuth Transitions * Calculated from Bridgman data. Pressure V * R kb (Bridgman) I-II 24.8 4.60% 75-88% n-m 26.5 2.95% 12-25% ni-rv 44 .6% 1-3% rv-v 83.8 .5% 1-2% V-VI 88 1.2% 4-6% to the volumetric transition designated by Bridgman as V-VI and that the difference in observed pressures is probably caused by substantial friction losses occurring only during electrical runs and not occurring during the measuring of volumetric changes. Accepting Kennedy's explanation, Bundy found three of the volumetric transitions accompanied by discontinuities in electrical resistivity,5 while no such correlation was observed for the volumet- After the correction described in test is applied, Bundy's VI-VIII electrical transition becomes identical with Bridgman's volumetric V-VI. ric transitions Bi III-IV and Bi IV-V. In analyzing this apparent discrepancy, we have come to the conclusion that, as Bridgman surmised, the reasons for it lie in the comparatively small volumetric and electrical changes accompanying these transitions. The total initial resistance of specimens used by Bundy were only on the order of 0.060 ohm. One per cent of change in resistivity would amount to less than one milliohm and, most probably, be too small to be observed with the instrumentation used in those investigations. The available cubic machine offered the opportunity to use rather large specimens with a resistance substantially higher than that used in 5 /O /S SO -?5 30 33 40 45 SO Fig. 9 Bismuth transition curve /.ooom