THE UNI VE RS I TY OF MI C H I GAN COLLEGE OF ENGINEERING Department of Mechanical Engineering Technical Report VARIABLE-COPRESSION-RATIO PISTON AND ITS EFFECT ON IDEAL AND PRACTICAL ENGINE EFFICIENCY Kamalakar Rao E. T. Vincent ORA Project 05847 under contract with: U. S. ARMY TANK-AUTOMOTIVE CENTER PROPULS ION SYSTEMS LABORATORY CONTRACT NO. DA-20-018-AMC-0729-T WARREN, MICHIGAN administered through: OFFICE OF RESEARCH ADMINISTRATION ANN ARBOR August 1966

TABLE OF CONTENTS Page LIST OF FIGURES v LIST OF TABLES vii ABSTRACT ix I. OBJECT 1 II INTRODUCTION 2 III. METHOD IV. IDEAL PERFORMANCE 7 V. ENGINE PERFORMANCE 12 Charge Dilution and Volumetric Efficiency 12 Performance of VCR Engine 19 VI. DISCUSSION 22 VII, CONCLUSION 23 VIII. REFERENCES 24 iii

LIST OF FIGURES Figure Page 1o Maximum pressure, compression ratio, and boost ratio employed. 4 2, Index of compression. 6 35 Ideal engine performance, Pmax = 2000 psi. 8 4o Ideal engine performance, Pmax = 3000 psi. 9 5. Volumetric efficiency and boost ratio. 13 6. Cylinder air charge, Pmax = 2000 psi. 14 7. Cylinder air charge, Pmax = 3000 psi. 14 8. Corrected performance, Pmax = 2000 psi. 17 9. Corrected performance, Pmax = 3000 psi. 18 10, Performance at fixed compression ratios. 20 11. Comparison of ideal and engine results. 21 v

LIST OF TABLES Table Page I, Heat Flow to Coolant 5 II, Performance with VCR Piston Set for 2000 psi 10 III, Performance with VCR Piston Set for 3000 psi 11 IV. Corrected Performance 16 vii

ABSTRACT This report presents the results of calculations of the ideal performance of a compression-ignition engine over a range of compression ratios from 8 to 22:1, For these calculations it was assumed that the cylinder charge undergoes typical heat losses, but that combustion is perfect. The results are compared with the results of a series of tests of the engine at three different compression ratios, for each of which injection characteristics were varied to achieve the best performance and specific fuel economy. Finally, a typical performance curve for an engine employing a normally operating variable-compression-ratio piston is compared with the above data. The object is to establish the penalty, if any, of employing a fixedinjection system in which the chamber shape varies with the engine load but the maximum gas pressure remains roughly constant. ix

I. OBJECT The object of this analysis is to examine the effect of varying compression ratio upon ideal and practical cycle efficiencies, and thus to determine how specific fuel consumption (SFC) varies as the compression ratio is changed over a wide range by means of a variable-compression-ratio (VCR) piston. 1

II. INTRODUCTION The advent of variable-compression-ratio pistons has introduced another variable into the already formidable array of factors affecting the power output and specific fuel consumption of a compression-ignition engine. It has been established that the variable-compression-ratio piston does permit control of the maximum cylinder pressure as the degree of supercharging varies (Refs, 1 and 2). As a result, one can use high-ratio supercharging in conjunction with a large increase intbrake mean effective pressure without increasing the structural load on the engine mechanism, and yet at the same time retain high compression ratio at slow speeds and light loads; an engine built according to these principles should start well in cold weather, use little fuel while idling, and operate with any of a variety of fuels. Currently available results of engine tests show no major effect of variable-compression-ratio on specific fuel consumption. True, fuel consumption at. low ratios is somewhat higher, but the reason for this increase is not clearo It may be that a change of ratio should be accompanied by a change in the injection system, to allow for the rather large variation in the shape of the combustion chamber as the ratio varies, as well as some changes in the turbulence of the fuel-air mixture and other phenomena, In this investigation we duplicated the range of compression ratios used in earlier tests and analyzed the ideal cycle to obtain the theoretical variation in efficiency, indicated mean effective pressure, specific fuel consumption, etc., as the piston position varies to maintain a constant maximum cylinder pressure. Then with the aid of test results taken with the compression ratio fixed, and injection tailored to that ratio, we predicted the effective ratio between the results of actual tests and ideal calculations. Using this material one can make theoretical comparisons between the performance of a variaible-compression-ratio piston engine and the best performance of a fixedratio engineo Thus, one can work toward estimating how a compromise in injection characteristics will affect engine performance, 2

TII. MEIHODS To achieve the above objective, we began with an ideal cycle which would approximate engine conditions, taking into account the effects of heat losses, ATnen boost ratio varies widely, heat losses change so greatly as to completely mask the effects of the variable ratio unless properly corrected for, Some years ago E, T. Vincent charted the results of cycle calculations in which data on variable specific heat and on heat losses during compression, comchustion, expansion, etc., were used in making close approximations of engine cycles in which -ipercharge ratio, heat loss, and compression ratio varied only moderately, With some extrapolation, these charts were extended to cover the range of vale$s in question here. Errors introduced by the extrapolation are considered minor. The ideal cycle was examined for two limiting cases, (1) maximum cylinder pressure of 20003 psi, and (2) maximum cylinder pressure of 3000 psi, For each maximum pressure a compression ratio of 22:1 was used for starting and low-load operation; as the load increased, the ratio was gradually reduced to 12l1, This is the range of variation used in current variable-compressionratio engines. To provide for decrease in compression-ratio variation with further developments in engines, additional calculations down to a ratio of 8 1 were made. Also, to allow for increases in boost ratio above that at which a maximum pressure of 2000 or 3000 psi is reached, calculations were made in which the compression ratio was held at 12:1 while the boost ratio increased from 532, giving 2000 psi, to 5,0:1, giving a maximum pressure of 520C psi for case (]). In the case of 3000 psi, the boost ratio was raised from i4.75 i giving 3200 psi at 12:1 compression ratio, up to 8o0:1, giving 5200 psi at the same pressure ratio, Figure 1 shows the compression and boost ratio ranges employed, and the maximum pressures reachedo To fix the cycle, it was necessary to select a value for the ratio Pmax/Pcomp. This was kept at 1,5 foor all cases examined, since, in general, it is about the minimum at which good combiustion can be secured. One other important factor is the fuel/ air ratio; this was held at 0.058, giving a heat addition of 700 Btu per cycle per lb of air. To include the effect of heat flow to the coolant, some assumptions were necessary. The literature contains little background for the boost ratios investigated, bOut on the basis of that at hand the schedule of heat losses in Table I was employed. It is based upon the established fact that as the boost ratio increases, the percentage of heat flow decreases, but of course the total.tu transferred increases because of the greater heat release per unit volume of cylinder. In the calculations performed, the percentage of heat flow to the coolant was a ssu med to vary as in Taloe I. 5

5000 r CR 12.0 I 4000 -CR 22301 CR 12.0:1 ^-CR 12.0:1 CR8o0.1 e ~-' 3000.E I X - 3000 - CR 12.0': - CR 22:. 01 1 II 1.41 2.05, 1 3.2, i4,75 5.55,& 0 &2 I IiI I I 1 I 1.0 2.0 3.0 40 5.0 6.0 7.0 80 Boost Ratio Figure 1. Maximum pressure, compression ratio, and boost ratio employed.

TABLE I HEAT FLOW TO COOLANT Heat Transfer: Percentage of Btu of Fuel at Phase of Cycle Boost Ratio = 1.0 Boost Ratio = 5.0 Compression Ratio 22:1 Compression Ratio 12:1 Compression 0.5 035 Combustion 1,2 lo 00 Incomplete Combustion 2.0 1.50 Expansion 4,0 3.50 Exhaust 5.0 4.00 Loss to Oil 1,2 1.00 Mi scellanc: Is 4,1 2 65 Total Percentage of Loss 18.0o 14,0% Of the heat losses listed, the only ones affecting the cycle efficiency are the first four, plus one quarter of the sum of the last two per stroke, This assumes that these last losses are distributed uniformly over the whole cycleo Losses at other ratios were assumed to be in linear proportion on the basis of the two conditions given in the table. This is only an assumption, but is considered reasonably close for the present purpose. Actually it is believed that as boost increases, these losses decrease fairly rapidly at first and then tend to level off,.By use of the results thus gained and of the charts already mentioned we determined the value of n in PV = constant, This is the compression for the heat loss assigned over compression ratios of 15 to 17:1, for which the charts were originally designed. The result is plotted as the solid line in Figure 2, in which straight lines were obtained; these were extrapolated over the range of ratios from 9:1 to 22:1, shown by the dotted lines, Examination of Figure 2 reveals that n changes from an average of 13564 to Lo377 over the total range of ratios at the normal air inlet temperature of 6003 and from 1,355 to 1,368 at about 700,~ the gas temperature value employed for high boost with aftercooling. In view of this modest change, and its small effect on the cycle, a constant value of 1.363 was finally used for all cases, For expansion after combustion the value of n varied even less, so that it was taken as 1,265, The heat addition during the cycle was calculated as a heat addition at constant volume but variable specific heat from the end of compression to the maximum cylinder pressure, followed by heating at constant pressure until the total 700 Btu per lb of air had been accounted for, either as state change or as the result of heat loss to coolant, 5

HEAT LOSS 2.1 1.39- PEn STROKE 2.8 4.1W ~3.5 BTU 0O 1.38 - > 1.37. F:.. 1.35 1.34 1.39 10 12 14 16 18 20 22 C. R Figure 2. Index of compression. With this information the work done, heat added, change of volume, etc., were obtained, and the ideal cycle efficiency, IMEP, IHP/lb of air, SFC/lb/ IHP/hr, etc., were calculated for the complete range of boost and compression ratios desired. The results so obtained represent the practical ideal cycle without allowance for volumetric efficiency, mechanical losses, rounded corners of indicator diagram, etc. 6

IV. IDEAL PERFORMANCE The results obtained are recorded in Table II for a limiting Pmax of 2000 psi and in Table III for 3000 psi. They are plotted in Figures 3 and 4. Using a naturally aspirated engine with 22:1 ratio, for easy starting, and a Pmax/Pcomp = 1.5, the boost was increased until Pmax = 2000 rpm at a ratio of 22:1. After this point the compression was reduced as boost increased; as before, Pmax was held at 2000 psi and Pmax/Pcomp at 1.5. Similar methods were employed for maximum pressure of 3000 psi. Note that the rpm of the engine enters into these relationships only indirectly, in the air flow. The data of Tables II and III, as far as IHP, etc., are concerned, are based upon an air flow of 1 lb/sec, which means that the rpm varies with cylinder displacement. To obtain the actual output per cubic inch of displacement for a given engine, the value given in the tables must be multiplied by the lb/sec of air flow and the ratio of total volume to displacement volume. It must be remembered that the above results are for a volumetric efficiency of 1004, with no dilution by retained exhaust gas. In other words, we assumed that the clearance is scavenged perfectly, so that one pound of air completely fills the volume of displacement plus clearance. 7

700 PMoxz 2000 psi 1. H. P. I Ibsl Hr 600 600 a: 600- g^ ^^ -500 Compression Ratio at 12: 1 500 400.- 400- - x M. E. P. L 300- x -X 0.30 200! m ~ 200^ Comp. Ratio at 22:1 X X~ j x 0.25 -0.25 100 - - 6.0 c3 S. F. C. Ibs/I.H.P. Hr 0.20 0.20 0 5.0 70 - 4.0 I I I I, Efficiency % o >60' ~^^ a) ~3.0 50 oost Ratio 2.0 40 o 1.0 6.0 8.0 10.0 12.0 14.0 16.0 18.0 20.0 22.0 Compression Ratio Figure 3. Ideal engine performance, Pma = 2000 psi. 8

PMx -3000 psi 700 600 I. H. P./lbs Airlsec. 1000 500, - Comp. Ratio 12:1 I 800 \ 400 CL, 600- - 300 " I M. E P. psi I 400 """ - 0.50 I!~~200 - w^^^ ^^^k Comp. Ratio 22:1 200 0.25 o 0.20 i 9.0 ~~~70 -~~~ -8.0 7.0 Efficiency % 6.0 60 -5.0 40 I l l ll- 4.0 - cr^~ Cm pressionBoost Ratio o'-3~,,.......,.~~~~ 3.0 O".~ 50 - L 2.0 1.0 40- - I - I { 6.0 8.0 10.0 12.0 14.0 16.0 18.0 20.0 22.0 Compression Ratio Figure 4. Ideal engine performance, Pmax = 3000 psi. 9

TABLE II PERFORMANCE WITH VCR PISTON SET FOR 2000 PSI Compression ratio 22.0 22.0 22.0 22.0 20.0 18.0 16.0 14.0 12.0 12.0 12.0 10.0 8.0 Boost Ratio 1.0 1.2 1.3 1.4 1.58 1.84 2.11 2.43 3.20 4.0 5.0 4.12 5.55 Pcomp 932 1156 1244 1330 1330 1330 1330 1330 1330 1695 2130 1330 1330 Pmax 1400 1735 1868 2000 2000 2000 2000 2000 2000 2545 3198 2000 2000 Net Work 458.0 461 459 474 469 455 446 417 389 382 374 371 331 (Btu) Thermal efficiency (^) 65.6 65.8 65.6 67.6 67.0 64.9 63.6 59.6 55.6 54.6 52.4 52.9 47.2 IMEP (psi) 157.5 193.6 205 225 246 272 304 346 392 483 586 483 585 IHP (lb air/sec) 650 654 649 670 663 643 630 588 550 542 530 524 468 SFC (lb/IHP/hr) 0.210 0.209 0.211 0.204 0.207 0.213 0.218 0.233 0.249 0.252 0.258 0.261 0.292 Cylinder volume 16.51 13.52 12.67 11.92 10.86 9.57 8.46 6.99 5.85 4.66 3.76 4.60 3.49 i-J (cu ft/lb/sec) 0 Pressure at exhaust 52 62 67 71 82 95 111 134 166 203 248 218 298 HP (cu in of total 0.273 0.336 0.356 0.390 0.424 0.467 0.518 0.582 0.653 0.808 0.98 0.792 0.932 vol/lb/sec)

TABLE III PERFORMANCE WITH VCR PISTON SET FOR 3000 PSI Compression ratio 22.0 22.0 22.0 22.0 22.0 20.0 18.0 16.0 14.0 12.0 10.0 8.0 12.0 12.0 Boost ratio 1.0 1.2 1.3 1.4 2.05 2.34 2.74 3.21 3.84 4.75 6.7 8.17 6.7 8.0 PComp 932 1156 1244 1330 2000 2000 2000 2000 2000 2000 2000 2000 2570 3410 Pmax 1400 1735 1868 2000 3000 3000 3000 3000 3000 3000 3000 3000 3855 5170 Net Work 458 461 459 474 464 454 446 433 414 391 370.5 336 404 402 (Btu) Thermal efficiency (%) 65.6 65.8 65.6 67.6 66.3 64.8 63.7 61.8 59.2 55.9 53.0 48.o 57.7 57.4 IMEP (psi) 157.5 193.6 205 225 343 371 414 459 516 593 716 878 752 979 IHP (lb air/sec) 650 654 649 670 657 640 629 611 574 552 523 475 569 567 SFC (lb/IHP/hr) 0.210 0.209 0.211 0.204 0.209 0.214 0.218 0.224 0.238 0.248 0.262 0.288 0.241 0.242 Cylinder volume 16.51 13.52 12.67 11.92 7.65 6.96 6.15 5.43 4.67 3.89 3.11 2.365 3.16 2.42 H- (cu ft/lb/sec) H-J Pressure at exhaust 52 62 67 71 109 123 145 166 199 248 327 450 310 411 HP (cu in of total 0.273 0.336 0.556 0.390 0.596 0.659 0.710 0.782 0.854 0.964 1.17 1.398 1.252 1.615 vol. per lb/sec.)

V. ENGINE PERFORMANCE The above ideal performance characteristics include an approximation for the cyclic heat losses, as they affect power output and fuel economy. To obtain a description of an actual engine that can be compared with variable-compression-ratio tests, the results of dilution of the charge by exhaust left over from the previous cycle must be allowed for, as must the rounded (rather than square) corners of the indicator diagram, the effects of volumetric efficiency, and the mechanical efficiency of the engine. For convenience, the performance was based upon results reported hereo CHARGE DILTION AND VOLUMETRIC EFFICIENCY The actual cylinder charge is a mixture of air, fuel, and exhaust gases left in the clearance space from the preceding cycleo The amount of residual gas present is difficult to determine, for it varies greatly with valve overlap, pressure difference between inlet and exhaust manifold, dynamic effects of the inlet and exhaust system, etc, To approximate this value for an engine with well designed manifolds, valves, etCo we have assumed that scavenging of the clearance space is uniform, leaving 25% of the residual gas trapped. The maximum air charge in the cylinder, then, is given by Equation (1), for a total cylinder volume containing 1 lb of air. Air charge per cu ft = loO - l0 x 0.75 lb ( of total volume = 1.0 -075 lb R- 1 where R = compression ratio. It is recognized that the above equation includes no factor for the varying boost ratio, which tends to improve scavenging as it increases. However, as the boost increases, the compression ratio decreases; hence the clearance volume to be scavenged increases, and the air flow must be greater during valve overlap to scavenge the assumed 25% of the clearance. Thus it is hoped that between these two effects-boost ratio and clearance volume-a good approximation will be obtained, The effect of volumetric efficiency is mainly one of pressure reduction and temperature increase between the atmospheric conditions and the charge in the cylinder when the engine is operating naturally aspiratedo For the unboosted condition, atmospheric pressure in the cylinder was assumed to be 14o7 psi, and the temperature 100~Fo If at BDC these values become 13o7 and 150~F, then without scavenging the maximum volumetric efficiency is 12

- 13.7 x 56 100 = 85.5 (2) T~v -14.7 x 610 =weight trapped in cylinder Now engine tests indicate that Tv increases with boost, to over 100% at high ratios. We will assume that Wv reaches 98% at a boost ratio of 5:1, and that it changes linearly with ratio. Thus Figure 5 would represent an approximate plot of vr vs. boost. The relation between rv and x for this line is 120 - 110 - 100 E 90- g 1.0 2.0 3.0 4.0 5.0 6.0 7.0 8.0 9.0 Boost Ratio x Figure 5. Volumetric efficiency and boost ratio. given by v = 3.25x + 81.75 (3) It follows that the new air change in lb per cycle will be given approximately by air change = (1.0 - 0275) (3.25x + 81.75) lb/cycle (4) R-l Equation (4) applies to a cylinder which contains 1 lb of air per cycle when scavenging is complete. The results of plotting Equation (3) against boost ratio are also shown in Figure 5. Note that when the boost ratio is 9:1, Wv, calculated normally, is 111%. This may seem high, but at such a high ratio the pressure at the inlet and exhaust manifolds will differ by about 20 psi when normal pressure drop for turbocharging is used. With valve overlap, good scavenging of the clearance space is then possible. If the compression ratio is 8:1 for a boost 13

ratio of 9:1, a volumetric efficiency of 115% could be achieved. From Figure 5, in conjunction with Equation (4) and the boost-compression relationship of Figures 3 and 4, we have calculated the expected mass air flow for a cylinder having a total capacity of 1 lb of air per cycle. The results are shown in Figure 6, for a Pmax of 2000 psi, and in Figure 7 for 3000 psi. dr. PMoa' 2000 psi - 100 -. I I I I. Volumetric Efficiency % C-) u Total Air Charge 80 -c > 70 6.0 8.0 10.0 12.0 14.0 16.0 18.0 20.0 22.0 Compression Ratio Figure 6. Cylinaer air cnarge, Pmax = 2000 psi. 120 - PMax 3000 psi rxCA Ui Efficiency _ 90 I-> Total Air Charge 80 I I I I I I I I 6.0 8.0 10.0 12.0 14.0 16.0 18.0 20.0 22.0 Compression Ratio Figure 7. Cylinder air charge, Pmax = 3000 psi. These values were calculated as follows: Example. Let boost ratio be l1.4:1 at a compression ratio of 22:1, calculate the mass of new air inducted per cycle. 14

From Figure 5 or Equation (3), volumetric efficiency = 86.2 = nv. This is based upon the usual definition for nv and thus refers to the engine displacement volume, not to total volume. Translated to total cylinder volume, Equation (4) gives air change = (1.0 - ~.75) (3.25x + 81.75) R-l = (1.0 - 275) 86.2 = 0.964 x 86.2 21 = 83.20 For a total cylinder volume equal to 1 lb of air per cycle, then, the volumetric efficiency becomes (8302 x 22)/21 = 87%, rather then 86.2o as given by Equation (3)5 The correction for clearance volume scavenging is seen to be small in this case, where the boost ratio is low and the compression ratio high. From Figure 6 for Pmax = 2000 psi the difference becomes (100.3 - 98.2) = 2,1% when the boost ratio is 5,55:1 and compression ratio is 8:1, whereas at Pmax = 3000 psi with 8,17:1 boost and compression of 8:1, the difference between the values in 5 and 7 becomes (110.5 - 108) = 2.5% It is now possible to obtain the effective air charge to the cylinder for all the supercharge ratios used over the range of compression ratios given by the VCR pistono The results of these calculations are given on lines 6, 7, and 8 of Table IV, Line 9 gives a correction factor, necessary because up to this point the indicator diagram has been assumed to have square corners rather than the actual rounded ones. The correction factor has been set at 95% for a naturally aspirated engine, and increases, with the diagram area, up to 99% at high boost ratios, By this means results predicted for the engine were calculated; they are given on lines 10, 11, and 12 for Pmax = 2000 psi, and on lines 22, 23, and 24 for Pmax = 3000 psi, All these figures, as corrected, are diagrammed in Figures 8 and 9, PERFORMANCE OF VCR ENGINE Let us now compare the predicted performance characteristics with the results of engine tests, These data are given in two different ways: 1, Engine performance was obtained with the piston locked at a series 15

TABLE IV CORRECTED PERFORMANCE (Pmax = 2000 psi, F/A = 0.038) Compression ratio 22 22 22 22 20 18 16 14 12 12 12 10 8 Boost ratio 1.0 1.2 1.3 1.4 1.58 1.84 2.11 2.59 3.20 4.0 5.0 4.12 5.55 Charge density 0.0605 0.074 0.079 0.0839 0.092 1.047 1.184 1.452 1.735 0.215 0.266 0.2175 0.286 (lb/cu ft) Cylinder volume 16.51 13.52 12.67 11.92 10.86 9.57 8.46 6.99 5.85 4.66 3.76 4.60 3.49 (1 lb of air) Charge wt (lb) 0.82 0.824 0.828 0.830 0.832 0.833 0.838 0.840 0.850 0.872 0.900 0.862 0.876 IMEP (psi) 129.1 159.5 169.6 186.6 204.5 226.5 254.5 290.6 333.0 421.0 526.0 417.0 512.0 HP (lb air/sec) 532 538 568 556 552 535 528 494 468 472 476 453 409 SFC (lb/IHP/hr) 0.257 0.254 0.241 0.246 0.249 0.256 0.260 0.277 0/.292 0.289 0.287 0.301 0.334 Correction factor 0.95 0.953 0.956 0.96 0.963 0.966 0.97 0.973 0.976 0.98 0.983 0.986 0.99 Corrected IMEP 123 152 162 179 196.8 219 247 283 325 412 517 411 507 HP (lb air/sec) 505 512 543 533 531 516 512 480 457 462 468 447 405 SFC 0.271 0.267 0.252 0.257 0.259 0.265 0.268 0.285 0.299 0.295 0.292 0.305 0.337 ca Compression ratio 22 20 18 16 14 12 12 12 10 8.0 Pmax = 3000 psi, F/A = 0.038 Boost ratio 2.05 2.34 2.74 3.21 3.84 4.75 6.7 8.0 6.7 8.17 Charge density.1307.1437.1625.1842.2141.257.3165.4132.3217.422 Cylinder volume 7.65 6.96 6.15 5.43 4.67 3.89 3.16 2.42 3.11 2.37 (1 lb of air) Charge wt (lb 0.852 0.859 0.866 0.877 0.887 0.906 0.965 1.005 0.950 0.967 IMEP (psi) 292 319 358 402 458 537 725 984 681 848 HP (lb air/sec) 560 550 544 536 509 500 549 570 498 549 SFC (lb/ IHP/sec) 0.245 0.249 0.252 0.255 0.268 0.274 0.250 0.240 0.275 0.298 Correction factor 0.96 0.963 0.966 0.97 0.973 0.976 0.98 0.983 0.986 0.99 Corrected IMEP 280 307 346 390 445 524 710 967 671 840 HP (lb air/sec) 538 529 525 520 495 488 538 560 491 455 SFC 0.255 0.259 0.261 0.263 0.275 0.281 0.255 0.244 0.279 0.301

600500PMax- 2000 psi 400 - 0 600 aLi 300 - I.M.E. P. 550 200 500 < I. H. P. lb. Air /sec - | 100 t- X 150 I.M.P. 49" Boost F/A =0.035 ~0.,40 x / x 230 " 71" " " =0.0375 450' 0.40 / 260 " 81" " " =0.0385 = 0.35 \~ -.. — _ ~X -I | \~-' X lJ Min. S. F. C. ^0.30^ 0 X x, J Average " --- ^ 0.30 - at^ x Max. at 260 psi F/A =0.0385 |,' A "S. F. C. L 0.25 0.25 - *6.0 8.0 10.0 12.0 14.0 16.0 18.0 20.0 22.0 Compression Ratio Figure 8. Corrected performance, Pmax = 2000 psi. 17

900 800 PMOax3000 psi 700 \0 600 600 6 L"i 500 E 500- I. M.E.P. ~t' ok ^~.q 550 400 - 300 - 500 H. P.:. 200 - 450 100 - 0.35x x | 400 0 0.30S. F. C. 0.25 Ll' 0.20. I I I I I *6.0 8.0 10.0 12.0 14.0 16.0 18.0 20.0 22.0 Compression Ratio Figure 9. Corrected performance, Pa = 3000 psi. 18

of fixed compression ratios, for various injector adjustments at each position. The results thus represent the best performance, the average, and the worst, obtainable for the given combustion shape at each compression ratio, 2. Engine performance was obtained with the VCR piston in operation, so that the ratio was varied automatically with load; the same fixed-injection system characteristic was used for all ratios and loads. In other words, the second set of results is for normal operation of the VCR system. Figure 10 shows the results of the first series of tests in which compression ratios of 11.7, 14.1, and 16.1 were used, with the engine set for a gas pressure of 2000 psi. Each graph consists of three lines, a solid one and two broken ones, The solid line is drawn through the points obtained by averaging all the data available for different injection characteristics, etc.; the lower broken one is drawn through the best specific consumption data, and the upper one through the worst, Figure 10 is not exactly comparable with Figure 8, because of variations in F/A and turbocharger ratio in the VCR engine tests, but the points are shown in Figure 8, In general the F/A ratio is higher than the theoretical curves because the latter are obtained for complete combustion whereas the former is naturally affected by some incomplete combustion products. If follows that for any given turbocharger ratio the IMEP developed will be somewhat less then the calculated values even if the F/A ratio is same. The next step was the combination of the data on the VCR engine in normal operation with the data of Figures 8 and 10. The results are shown in Figure 11, together with some material on a smaller engine. On this diagram the performance of the test engine has been plotted at the IMEP of the engine test; the mean pressure curve of the theoretical calculations was used to locate the pointo This was necessary because the engine-compression ratio was not known exactly, being a variable affected by many factors, Hence the ratio scale of Figure 11 is not exactly true except for the ideal case, but it is considered approximately so. The results shown are thus a first indication of the effect of an automatically variable compression ratio on ideal engine performance. Note that Figure 11 gives the average F/A ratio for the test results; it is considerably higher at 410 IMEP than that obtained in the ideal calculations, approximately the same at 300 psi, and lower at 200 psi and 140 IMEP; these differences, and their effect upon MEP, account for the change of slope of the curves. 19

~0.38 -Comp. Ratio 16:1 x 0.38 _ - -'r' X ~- - = 00 30 as _ 01 -0.38 - XX X I [Comp. Ratio 14:1 W X O XX 0.26 X X,,, 0.38 Comp. Ratio 14:m1 R ^~~~~~~ 0.36~ ~ ~ ~ ~ ~ ~ ~~~0.34 -o,,X..-w-. a a-l x 0.30 d 0.26 L(I 100 150 200 250 -r- 0.38. Comp. Ratio 117 Figure 10. Performance at fixed compression ratios. v,~ ~~~0.36 b20 100 150 200 250 I. M. E. P. psi Figure 10. Performance at fixed compression ratios. 2O

Mansfield Data at 7.8to Compression 500480 450 400395 300 C' a I 1I.M.VE. P. - 200 0-.. 100 Gardner Engine OEngine Data F/A 0.047' 0 Fixed Ratio Data FIA 0.046- v F/ A 0. 0464 3 FIA =0.043 * Calculated Data - A FI/A = 0.0385 Fixed Ratio 0 o e" s:. 0.35 0 ~ oFIA=0.036 Figure 11. Comparison of ideal and engine results. 0 03215 0.30 t —ft,/ IDEAL S. FC. 0.25 6.0 8.0 10.0 12.0 14.0 16.0 18.0 20.0 22.0 Compression Ratio Figure 11. Comparison of ideal and engine results. 21

VI. DISCUSSION This report is summarized in Figure 11, a comparison of the performance of the normally operating VCR engine, the fixed-compression-ratio engine, and the ideal one, To this diagram have been added Mansfield's results for the Gardner engine (Refo 3), converted to IMEP as closely as possible. This engine was operated as a fixed-ratio one, with an injection characteristic carefully developed to suit the combustion chamber. It is seen that the results line up quite well with those for the assumed fixed-ratio engine, considering that the F/A ratio is different and that Pmax = 1,600 psi rather than 2,000 psi. The ideal data, at a F/A ratio of 0.038, naturally give the lowest SFC. The ratio between the ideal and fixed-ratio engines has an average of 0.80 when based upon the mean SFC curve for fixed ratios, and of 0.86 when evaluated from the minimum SFC curve. When the VCR engine curve is examined, this ratio ranges from about 75%, at F/A = 0.043, to 85%, at F/A = 0.027. Where the VCR engine F/A ratio is the same as the calculated results, the engine data roughly correspond with those for the fixed-ratio tests. Correction of the engine results at F/A = 0.043 to F/A = 0.038 gives a SFC of 0.36 lb/IHP/hr, which also gives substantial agreement with the fixed-ratio results, 22

VII. CONCLUSIONS It can be concluded that: 1, As the compression ratio changes from 22:1 to 12:1, ideal thermal efficiency for the cycle decreases by about 11%, resulting in an increase of SFC from 0.258 to 0.292 lb/BHP/hr. In view of the great change in inlet air density, heat loss from cycle, volumetric efficiency, etc., accompanying this change of ratio and IMEP from 140 to 400 psi, the change in cycle efficiency is considered small, 2. The engine data for different fixed ratios shows the same general trend as the theoretical calculations. An efficiency ratio of about 0.86 is shown when based upon the performance of the best fixed-ratio engines. 3. The VCR engine results examined had a widely varying F/A ratio over the range of compression ratios employed, so that direct comparison with the theoretical data was difficult. However, correcting the VCR data in direct proportion to the F/A ratio in the high-power range resulted in substantial agreement with the results for the fixed-ratio engine. 4, In the present engines, it can be concluded, little performance (possibly 10%) is lost when the injection characteristics are correctly adjusted to allow for the great change in chamber shape that occurs as the compression ratio varieso 5. Eliminating the multi-fuel requirement, or even just gasoline starting, would permit a lowering of the maximum compression ratio from 22:1 down to 17 or 18:1;this would be a considerable improvement in the usable mean pressure, and smaller changes in SFC. Further, mean effective pressures of 500-600 psi could at least be considered. Probably no increase in peak cylinder pressure would occur in the process. 6, The increased SFC at full load and speed (about 10% because of the use of the VCR principle) has an extremely small effect upon the overall fuel consumption per 24 hr when the engine is operating on the present duty cycle since the VCR engine is more efficient at part load than is the standard one. 7. Increasing the maximum cylinder pressure from 2000 to 3000 psi would reduce the variation in SFC with compression ratio. It would also make possible an IMEP of 700-800 psi, if the necessary cooling and injection systems and turbochargers could be developed. 23

VIII. REFERENCES 1o W. A. Wallace and J. B. Lux, " A Variable Compression Ratio Engine Development," SAE Transactions, 72, 680 (1964). 2. J. C. Basiletti and E. F. Blackburne, "Recent Developments in Variable Compression Ratio Engines," SAE Paper 660544 (1966). 35 W, P. Mansfield and W. S. May, "Diesel Combustion at High MEP With Low Compression Ratio," SAE Paper 660343 (1966), 24

UNIVERSITY OF MICHIGAN 3 9015 03695 6772