I LL IN I UNIVERSITY OF ILLINOIS AT URBANA-CHAMPAIGN PRODUCTION NOTE University of Illinois at Urbana-Champaign Library Large-scale Digitization Project, 2007. UNIVERSITY OF ILLINOIS ENGINEERING EXPERIMENT STATION BULLETIN No. 6 JUNE 1906 HOLDING POWER OF RAILROAD SPIKES BY ROY I. WEBBER, C. E., INSTRUCTOR IN CIVIL ENGINEERING The determination of a proper fastening between the rail and the tie has become a matter of considerable importance. During the period when the supply of suitable hard wood timber was suf- ficient, the ordinary spike satisfactorily fulfilled the require- ments of traffic; but with the increase in the amount of traffic handled, and the heavier weights of cars and locomotives, and also with the use of soft deciduous and coniferous woods for ties, the common spike has proved deficient. Variations in the form of the ordinary spike have been developed, and new forms of spikes have been devised in an attempt to overcome the loss of efficiency attendant upon the use of inferior timbers. In view of these conditions, and the meager supply of pub- lished data on the holding power of spikes in ties, the writer has carried out a series of experiments to determine the resistance to withdrawal offered by the same type of spike in different timbers and by different forms of spikes in the same timber, and also to determine whether or not the preservative has any influence upon this resistance. The writer wishes to express his thanks for the hearty co- operation received from the various persons, firms and corpora- tions mentioned in the text. He wishes also to express his in- debtedness for personal aid, to Mr. Robert Trimble, Chief Engin- eer Maintenance of Way, Pennsylvania Lines; Mr. George E. ILLINOIS ENGINEERING EXPERIMENT STATION TABLE I DESCRIPTION OF THE TIES No. of Kind of Kind of Tie Timber Treatment Zinc-Creosote Zinc-Creosote Zinc-Creosote Zinc-Tannin Zinc-Tannin Zinc-Tannin Zinc-Creosote Zinc-Creosote Zinc-Tannin Zinc-Creosote Zinc-Creosote Zinc-Creosote Zine»-Cre'osonte 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 Blue Ash Blue Ash Sweet Gum Water Oak Water Oak Red Oak Red Oak Red Oak Red Oak Rock Elm Poplar Elm Elm Beech Elm Black Oak Red Oak Black Oak Poplar Loblolly Pine Lob'y Pine Red Oak Black Oak Black Oak Water Oak Water Oak Black Oak Red Oak Water Oak Red Oak' White Oak White Oak White Oak Water Oak Burr Oak Beech Elm Beech Lob'y Pine Chestnut Red Oak Beech Beech Beech Date Treated 1905 1905 1904 1904 1904 1904 1905 1905 1904 1905 1905 1902 190'9 1902 1902 1905 1905 1905 1905 1905 1905 1905 1905 1905 1905 1905 1904 1904 1904 1904 1904 Remarks I Zinc-Creosote Zinc-Creosote Zinc-Tannin Zinc-Tannin Zinc-Tannin Zinc-Tannin Zinc-Tannin Zinc-Tannin Zinc-Tannin Zinc-Tannin Zinc-Tannin Zinc-Tannin Zinc-Tannin Creosote Creosote Creosote Creosote Creosote Seasoned; sound Seasoned; sound Seasoned; sound Seasoned; sound Seasoned; sound Seasoned; sound Seasoned; sound Seasoned; sound Seasoned: sound Seasoned; sound Seasoned; sound Seasoned; sound Seasoned; sound Seasoned; sound Seasoned; sound Seasoned Seasoned Seasoned Seasoned Treated Decem- ber, 1905; sound Treated Dec; '05; sound Treated Dec; '05; split Treated Dec; '05 Treated Dec; '05 Treated Dec; '05 Treated Dec; '05 Treated Dec; '05 Treated Dec; '05 Treated Dec; '05 Treated Dec; '05 (Seasoned; in track two years Indiana Oak; sap I wood showed slight [ decay Georgia Oak; seasoned; sound Sound Sound Sound Sound Sound Seasoned; sound Seasoned; sound Showed tendency to split Sound Sound Sound PLATE I TESTING MACHINE WITH TIE IN POSITION FOR TEST WEBBER-HOLDING POWER OF RAILROAD SPIKES Boyd, Roadmaster of the Illinois Central Railroad; Mr. A. L. Kuehn, Superintendent of Maintenance of Way, of the Cleveland, Cincinnati, Chicago and St. Louis Railway; Dr. Octave Chanute, President of the Chicago Tie Preserving Company, Chicago, Illi- nois; and to Professor Ira 0. Baker and Professor C. H. Hurd of the University of Illinois. THE TIES The ties used in these experiments were furnished gratuitous- ly as follows: Nos. 1 to 11, and 16 to 30 by the Chicago Tie Pre- serving Company, Chicago, Illinois; Nos. 12 to 15 by the Illinois Central Railroad Company; Nos. 31 to 41 by the Cleveland, Cincin- nati, Chicago and St. Louis Railroad Company. Table I gives a description of the several ties used. The ties were taken either from the stock pile of the railroad companies or from those of the treating plant. No attempt has been made to trace their history farther back than the place of growth and the date of treatment. Treated ties were used in a majority of the experiments, since in the future, as the inferior grades are pressed into service, the ten- dency will doubtless be toward the use of preserved timber. EXPERIMENTS Two distinct lines of experiments were undertaken: (1) The determination of the resistance to direct pull of several forms of spikes; and (2) An investigation of the resistance to lateral thrust. Therefore the paper naturally divides itself into two parts: Part I, Resistance to Direct Pull; Part II, Resistance to Lateral Dis- placement. All of the experiments were made in the Laboratory of Ap- plied Mechanics, University of Illinois. PART I RESISTANCE TO DIRECT PULL The experiments were made with a Riehle 100,000-pound testing machine. Plate I shows the machine with a tie in position for a test. The pulling device for ordinary spikes, also shown in Plate I, was a Verona spike-puller threaded into a piece of steel gripped between the lower jaws of the machine; the pulling de- vice for the screw spikes was of the same general pattern and was designed especially for these tests. A scale graduated to 1-16 of an inch was so set that the distance moved through the lower head of the machine could be measured directly. A load of 500 4 ILLINOIS ENGINEERING EXPERIMENT STATION pounds was applied to insure the tie's having a good bearing be- fore any records were taken. The machine was geared to move at the rate of 5-8 of an inch per minute, which allowed time for carefully balancing the machine and for taking the readings of the scales. Five observations were usually taken; viz., when the lower head of the machine had moved through 1-8, 1-4, 1-2 and 3-4 of an inch, and also at the point at which the maximum fiber resist- ance was developed. No observations were made after the spike had been pulled 3-4 of an inch, as it would have lost its usefulness long before that point had been reached. Further consideration of this part of the paper will be contin- ued under the following heads: Art. 1, Holding Power of Ordinary Spikes; Art. 2, Holding Power of Screw Spikes without Linings; and Art. 3, Holding Power of Screw Spikes with Helical Linings. ART. 1 HOLDING POWER OF ORDINARY SPIKES The ordinary spikes were received from the following com- panies, the numbers in this list being the designations in the sub- sequent tables: Nos. 1 and 2 from the Pennsylvania Railroad Com- pany; Nos. 3 and 4 from the American Iron and Steel Manufactur- ing Company, Scranton, Pennsylvania; Nos. 5 to 10 from Dill- worth, Porter and Company, Pittsburg, Pennsylvania; No. 11 from the W. A. Zelnicker Supply Company, St. Louis, Missouri, and Nos. 12 to 14 from the Illinois Steel Company, Chicago, Illinois. The nominal dimensions of the four sizes of spikes are shown in Table II. The actual lengths varied considerably from the nominal lengths, usually being less. This was particularly true concerning the 6-inch spike. The actual cross sections were nearly the same as the nominal, the variation in thickness rarely being over 1-64 of an inch. As the range in thickness of the spikes was only 1-16 of an inch, some experiments were made with plain, square and chisel-pointed bars 1-2, 3-4, and 7-8 of an inch thick to determine the relation between the holding power and the cross section. The spikes had differently shaped points, as shown in Table II. Three spikes were used for each experiment, and these three were always of the same size and lot number. The spikes were driven by Mr. M. Flood, an experienced track foreman detailed for this purpose by the division engineer of the Cleveland, Cincinnati, Chicago and St. Louis Railway. WEBBER-HOLDING POWER OF RAILROAD SPIKES DESCRIPTION S.a') 6 5 1-2 5 1-2 5 1-2 6 5 1-2 5 1-2 6 5 1-2 5 1-2 5 1-2 5 1-2 5 1-2 6 5-8 5-8 5-8 5-8 5-8 19-32 19-32 5-8 9-16 9-16 9-16 9-16 9-16 5-8 TABLE II OF THE ORDINARY SPIKES F~~) I- ~ 0.372 0.372 0.372 0.372 0.372 0.352 0.352 0.372 0.316 0.316 0.316 0.316 0.316 0.372 Chisel Chisel Blunt Blunt Sharp Sharp Chisel Blunt Blunt Sharp Chisel Sharp Chisel Chisel 50 S 5a 5 0 5 5 Whole ties were used to insure freedom from splitting in driving the spikes, and care was exercised to avoid driving the spike into knots or cracks. The spikes were driven into the tie to a depth of 5 inches. In some instances, as shown in the record, holes were bored for the ordinary spikes, the hole being 1-16 or 1-8 of an inch less in diameter than the cross sectional dimensions of the spike. The depth of boring was not quite as great as the depth of insertion, so that the pointed end of the spike was forced into the undisturbed wood. Table III gives the detailed numerical re- sults of the tests and Plates II and III show graphically the curves of average resistances of the different ties. 0 1^ Smooth Smooth Smooth Smooth Smooth 6 ILLINOIS ENGINEERING EXPERIMENT STATION PLATE II ¾/÷ / a ,7 A 7/7e 0 Al y- '14 / S.. -I Whit - e oi Water[- Oak. : Oak. -I- + - - S I : i / I SI I I I r ' I Red I I \ I I Oak. I /11| Ii Beech. C/",wes ,/-oy 9Resa,,ce 'o /1h./bdar',/ of /'e So/l,e fv,, /,Ae ae. /Zo00 - -ye^re Sp /es - - -ScrewSpes ood Lw-rfs - SBm _Oak zzzz's'^^z - - - - , 3 - - - - _ /\ ._ _ / \==== _ _ _ _ y _ _ - - / - - - ^*//C I WEBBER-HOLDING POWER OF RAILROAD SPIKES PLATE III Z.D1tance Sp40Ae was W,1h/draw? .? /Ance' 0 14 S-, kes ,: I i ll 3/4 / 0 Z/, Ash I :I I 1l4 I I PL0bIn 1 "r ~~ ple , ~ '/2 3/4 / Chestnut Curves JhomaWy 1e/auwe lto aM/rr/ of Mte J.lole 91'z /e 77e. Elm /00oo S000 /lox Z: (3 30-00 5000 Sweet Gum -+-+-4 - I I !A 1 1.1 1L It l I I N ' 1 1l 1- ! 1i l l iI --- Poplar , 8 ILLINOIS ENGINEERING EXPERIMENT STATION TABLE III DETAILED RECORD OF TESTS OF DIRECT PULL OF ORDINARY SPIKES Kind of Tie Blue Ash Sweet Gum 3 3 ¢$ ° 12 6 14 11 3 6 H 1 2 3 Av. 1 2 3 Av. Av. 1 2 3 Av, 1 2 A3 A Ir Resistance in Pounds for Pull of 1800 4040 4270 4150 2220 3390 2860 3000 2630 3940 5180 3920 2900 3470 3540 3300 3030 2690 5030 3580 2110 2780 1680 9120 4460 5060 4340 4630 4700 6940 5670 5770 1930 4010 3920 3960 4030 4100 3580 3900 5100 5570 3400 4370 4030 3190 4100 3770 2650 6500 3890 4100 2910 6180 -15fl 5590 5220 4510 3860 4530 5250 4710 4890 4890 2010 3000 4620 2690 3260 2750 3030 3010 2930 4040 3440 2340 2320 3730 2790 4410 3590 4800 4190 4450 3990 3370 3970 5230 4710 4830 4830 2220 2550 4560 2470 2720 2780 2500 2640 2930 3100 3420 1680 3340 2510 4030 3340 4070 3810 Maximum Resistance 0 6840 7260 6330 6810 8740 8020 8540 8640 4300 5640 5180 5040 5610 5370 4900 5330 5100 5570 5700 5440 4030 4810 4980 4610 6500 5460 6180 6050 a ss §3u 3-8 3-16 3-16 3-16 3-8 5-16 3-8 3-8 3-16 3-16 1-8 3-16 3-16 3-16 3-16 3-16 1-4 1-4 3-16 1-4 1-4 3-16 5-16 1-4 1-4 3-16 1-4 1-4 4190 3810 3 A 310 59 WEBBER-HOLDING POWER OF RAILROAD SPIKES TABLE III-Continued 14 1 2790 2 3300 3 2220 Av. 2770 14 1 2 3 Av. 14 1 2 3 Av. 14 1 2 3 Av. 6 1 S9 7 6 3 Av. 1 2 3 Av. 1 2 3 Av. 2870 1610 2240 2560 3440 3160 Resistance in Pounds For Pull of o 0 0 6580 7060 5330 6320 6040 4460 5250 5430 3340 3050 4380 1580 3900 1470 3550 2190 4070 1740 3840 1960 6030 2390 5320 3200 6380 2520 5920 2750 6070 4330 4890 1610 4360 2930 5240 3370 3860 1800 5440 2550 4490 2570 4600 3200 5300 2130 5710 3500 5820 2940 5610 4190 2970 3920 3660 4270 5060 4660 3610 3050 3200 3290 3970 3450 2990 3470 5420 4380 4900 5260 3430 3190 3960 3380 3370 3680 3380 4020 4200 4620 S4280 3930 2940 3200 3360 3400 4240 3820 3530 2590 3210 3110 3160 3090 3130 4530 4100 4320 4560 3040 3020 3870 3180 3130 3230 3180 3820 3700 4340 3950 Maximum Resistance Kind of Tie Water Oak 25 26 7560 7060 7740 7450 7720 7780 7750 6150 4960 5810 5640 6000 5110 4070 5060 8690 8040 7320 8020 8580 5270 4760 6200 4910 5440 4490 4940 5300 5710 5820 5610 5-16 1-4 5-16 5-16 5-16 3-8 3-8 1-4 3-16 3-16 3-16 5-16 5-16 1-4 5-16 5-16 3-8 5-16 5-16 3-8 3-16 1-4 1-4 3-16 1-4 1-4 1-4 1-4 1-4 1-4 1-4 10 ILLINOIS ENGINEERING EXPERIMENT STATION TABLE III-Continued 281 462( 372( 372( 282 313t 343( 316( 300( 320( 323( 314( 308( 227( 199( 245( ') At Kind of Tie 2 344( 3 184 Av. 3571 1 2341 2 170( 3 3361 Av. 2471 1 4091 2 3091 3 3181 Av. 3451 1 2371 2 3011 3 3901 Av. 3131 Resistance in Pounds for Pull of 0 4480 3750 3 0 .... 4070 3 0 3450 2910 0 3970 3240 3 0 5920 4360 4 0 5600 4170 3 0 6330 4440 4 0 5980 3320 4 0 6020 3340 2 0 8010 4720 0 5800 3900 3 0 6610 3950 3 0 .. .. 2650 2 0 5090 3360 0 5420 3610 3 0 5260 5210 2 0 5100 4230 3 0 6680 4000 3 0 3710 2830 2 0 5160 3680 3 0 5620 5770 5 0 3730 2830 2 0 6560 3600 3 0 5300 4070 3 0 7000 4070 4 0 6780 3550 2 0 7280 4660 2 0 7020 4090 3 0 4720 4940 4 0 6670 5210 4 0 8130 5060 4 0 6510 5070 4 Water Oak Maximum Resistance rm g ^~' W w0 a o 1 2 3 Av. 1 2 3 Av. 1 2 3 Av. 1 2 3 Av. 1 29 4 5 26 25 34 34 6 13 11 11 13 13 14 160 720 440 360 460 060 -260 !750 300 890 980 i270 2940 1000 2770 640 980 2540 380 5080 2260 1010 950 :020 2900 2870 1260 r730 1930 540 i740 4480 4760 3720 4320 9000 7450 9000 8380 6240 9180 6490 7300 4240 5090 5420 4920 6450 6680 4550 5860 9070 4970 6560 6870 8430 6780 8040 7750 6400 7360 8130 7290 1-4 3-16 1-8 3-16 3-8 5-16 3-8 3-8 5-16 5-16 5-16 5-16 3-16 1-4 1-4 1-4 5-16 1-4 5-16 5-16 3-8 5-16 1-4 5-16 5-16 1-4 5-16 5-16 5-16 5-16 1-4 5-16 WEBBER-HOLDING POWER OF RAILROAD SPIKES TABLE III-Continued Resistance in Pounds Maxi for Pull of Resis o 0 0 0 0 0 Kind of Tie Black Oak 4600 3340 5040 4190 3330 3170 2830 3110 4010 3790 5320 4370 3080 8620 2670 4440 3550 2570 2900 3090 2850 3410 3550 4380 3780 7130 8250 7690 5860 5660 4000 4880 5890 6170 8620 6890 imum stance 0C R 3c 5950 4930 9000 .... 4380 9100 4620 3180 8700 5300 4160 8940 3270 2890 6110 4120 3760 6540 3980 3540 7760 3790 3390 6810 3460 3290 8210 4290 3850 7940 .... 4050 9060 3870 3730 8070 5300 4740 10000 4500 4170 8970 4200 3850 7070 4670 4250 8680 3370 3370 8470 2900 2930 8780 4480 2940 .... 3010 6880 .... 7220 3880 8500 3450 7530 3230 6110 ... * 6280 2090 4390 2660 5590 2980 6740 3380 7940 1220 2920 2200 5130 3430 8070 5910 2870 7570 0 - z 0 ~) cJ H -~ 0 0 I -~ 16 8 16 3-8 3-8 5-16 3-8 1-4 5-16 3-8 5-16 5-16 1-4 1-2 3-8 5-16 5-16 1-4 5-16 1-4 5-16 1-4 1-4 5-16 9-16 5-16 3-8 3-8 1-4 3-8 1-4 3-16 5-16 1-4 11 3150 3510 2750 2940 3070 2650 2660 2660 1700 2120 2250 2020 3650 4370 2400 3470 7020 8470 7130 6690 7430 5240 6190 5720 3410 3900 4000 3770 5890 4430 6230 5520 12 ILLINOIS ENGINEERING EXPERIMENT STATION TABLE III-Continued 27 24 10 1 2 3 Av. 1 3 Av. 4 1 2 3 Av. 7 1 2 3 Av. Black Oak Red Oak Resistance in Pounds fo.r Pull of 4900 7070 3780 5740 2730 6550 3800 6450 3950 6580 1810 4050 2960 5390 2910 5340 2330 5070 1880 5570 3450 6500 2550 5710 1820 4690 2250 4110 2960 7120 2340 5760 2520 6110 1810 5710 3020 6480 2650 1870 2320 2050 2210 2940 3170 2640 6130 4750 6750 5750 5460 6840 6570 6290 3890 3670 3780 4150 3460 3600 3740 5820 4320 4300 4810 2800 5520 4590 3930 4040 4160 3980 4030 4410 4150 4280 4310 3730 3410 3820 u 3140 3260 3440 3280 3650 2780 3410 3280 5710 3740 3800 4740 2880 3880 3620 3120 3490 3490 3710 3560 4190 3760 3980 4100 3370 3360 3610 Maximum Resistance 7070 5740 6550 6450 6880 6510 6500 6530 7740 7010 7360 7240 8790 7120 7700 7070 7070 7360 7130 7190 8300 7750 7300 7200 6570 6790 .^-* Q^.S 1-4 1-4 1-4 1-4 1-4 5-16 5-16 5-16 3-8 5-16 5-16 5-16 5-16 7-16 1-4 3-8 5-16 5-16 5-16 5-16 3-8 5-16 3-8 5-16 5-16 1-4 5-16 Kind of Tie 6 I I WEBBER-HOLDING POWER OF RAILROAD SPIKES TABLE III-Continued Kind of Tie Red Oak -z 0 cjD ,o Resistance in Pounds for Pull of 3300 8700 4 4200 7780 4 3750 8240 4 3100 2910 2 3750 3220 2 4890 4200 3 0 3910 3440 2 0 3940 2760 2 0 2890 2510 2 0 3490 2460 2 0 4670 2570 2 0 3250 2620 2 0 5490 3780 3 0 3950 2800 2 0 4400 4540 3 0 3650 2230 2 0 4300 2690 2 0 4110 3150 2 0 5030 5420 6 0 5240 9900 6 0 6400 7550 7 0 5560 7620 6 0 4480 3300 2 0 .... 3500 3 0 .... 3710 3 0 4350 3550 2 0 4410 3520 3 0 5330 4640 3 0 7830 4570 3 0 7000 4710 3 3 6740 4640 3 920 220 570 600 990 220 920 770 370 370 '550 440 400 650 630 420 530 2860 260 710 020 660 2910 3640 3080 2990 3150 3420 3200 3670 430 1 145( 2 203( 3 Av. 1740 1 1570 2 1730 3 1950 Av. 168( 1 250( 2 297( 3 349( 4 221 5 377( 6 262 Av. 293( 1 217 2 390 3 193( Av. 266( 1 171 2 224 3 3281 Av. 241 1 352 2 370 3 369( 4 332 Av. 3551 1 215 2 299C 3 324 Av. 276 22 6 6 8 12 11 Maximum Resistance 9210 5-8 8800 7-16 9000 1-2 7330 7-16 7230 7-16 8970 7-16 7840 7-16 5120 3-16 4990 3-16 5270 3-16 4670 1-4 5150 3-16 5490 1-4 5120 3-16 7040 5-16 6040 3-16 5650 5-16 6240 1-4 9720 3-8 11900 1-2 10940 7-16 10850 3-8 5950 3-16 6930 1-4 4460 3-16 6240 3-16 5900 3-16 7580 5-16 7830 1-4 8280 5-16 7890 5-16 14 ILLINOIS ENGINEERING EXPERIMENT STATION TABLE III-Continued Resistance in Pounds for Pull of 4± Cl CO Maxi Resis 0 P_ Kind of Tie Red Oak *This was the first tie tested, and gave unusually high results. 5010 7120 5630 9080 6110 4930 4300 7020 5560 6030 4960 8030 5410 3790 3680 5410 6280 3950 3510 6280 7140 4350 4300 7140 6280 4030 3830 6280 .... 3790 3630 7030 4600 3730 3110 6660 4610 3540 3180 6130 4600 3680 3320 6610 .... 11620* .... 11230 ... .... 10630 .... 11490 .... .... . 3430 3130 6390 7150 4410 3480 7150 5000 4270 3670 6760 6080 4040 3530 6770 .... 5200 3710 8200 4000 3280 2600 6250 6870 3740 3280 6870 5440 4070 3200 7100 6800 6080 4950 9420 6250 6680 4640 9240 6530 6380 4790 9330 9 9 17 28 28 8 1 2430 2 3430 3 Av.; 2930 1 3530 2 3000 3 3720 Av. 3420 1 3270 2 3740 3 3690 Av. 3560 1 2 .... 3 .... v .... 2 3910 3 3810 Nv. .3860 1 2 6250 3 2880 Nv. 4570 1 3030 2 2680 3 3580 Av. 3060 12 13 12 12 11 12 imum tance P'u ses- c) ^* 3-8 5-16 3-8 1-4 1-4 1-4 1-4 1-4 3-16 3-16 3-16 1-4 1-4 3-16 1-4 5-16 1-8 1-4 1-4 5-16 3-8 5-16 WEBBER-HOLDING POWER OF RAILROAD SPIKES TABLE III-Continued Resistance for P 22 Maximum Resistance az 13) 13 Kind of Tie Red Oak Burr Oak 6 0) GO 2440 2850 1880 2390 3700 1700 2700 1680 2020 2540 2040 4500 4750 1930 3730 3670 3760 4620 4010 2960 1410 3090 2490 4020 2200 2230 30 30 5810 5040 4530 5130 3960 3560 3760 3580 4070 4590 4070 7690 5840 6760 7950 7110 4500 6520 4960 3570 6850 5130 8640 3390 5020 coo~r' 12 10 in Pounds ull of 3270 3340 2770 2170 3700 3450 3250 2990 2950 2460 2920 3150 2940 3010 3070 3030 2490 2480 3070 2890 2870 2760 3530 3340 3200 3300 3370 3750 3360 3460 3100 2420 3450 3300 4450 3960 3660 3220 5290 4000 4940 4420 4190 3510 4810 3970 8240 5560 9450 6220 5930 5600 7540 5780 6210 5770 9240 4500 6290 5250 7250 5170 Av. 1 2 3 Av. 1 2 3 A 1-4 1-4 3-8 1-4 3-16 3-8 1-4 5570 3-8 5220 5-16 6080 3-8 5620 3-8 7690 1-4 7210 3-16 7430 5-16 7440 1-4 7950 1-4 7110 1-4 6230 3-16 7090 1-4 8290 5-16 7840 5-16 4940 1-8 7020 5-16 8240 1-2 9450 1-2 9440 3-8 9040 1-2 10560 3-16 9240 1-2 9000 3-8 9600 3-8 v. 2820 5680 5810 5040 6280 5710 5500 4720 5110 7250 5170 16 ILLINOIS ENGINEERING EXPERIMENT STATION TABLE III--Continued Resistance in Pound for Pull of -t Burr Oak White Oak s Max Resis Kind of Tie 8 1 2690 7040 4820 2 2740 5840 4060 Av. 2710 6440 4440 1 1 3240 7030 3400 2 2430 5870 4390 3 3700 7500 4180 Av. 3150 6800 3990 14 1 .... 4020 3600 2 3960 7100 4000 3 2250 5580 3650 Av. 3110 5560 3750 1 1 4220 3570 3810 2 1950 3670 4640 3 3190 5260 3810 Av. 3120 4160 4090 7 1 3860 9440 5930 2 3460 6400 3710 3 1610 3740 4670 Av. 2980 6530 4770 7 1 4790 3910 2860 2 4150 4930 3510 3 4630 3840 3070 Av. 4520 4230 3150 10 1 2400 3490 8280 2 3100 4840 5410 3 2570 6410 10670 Av. 2690 4910 8120 10 1 2930 5490 2390 2 3080 6360 2860 3 3890 6810 3540 Av. 3300 6220 2930 Q 0 31 imum tance a(.S " 0 4110 10090 3700 7920 3950 9000 3140 7030 3850 7580 3330 7500 3440 7370 3280 7830 3750 7100 3200 8980 3410 7940 3040 7520 3340 6940 3500 6410 3290 6990 4650 9440 3680 8650 5570 9360 4300 9150 2530 5750 3270 6500 2450 6030 2750 6090 3820 8280 3880 10190 4400 10670 4030 9710 2330 5490 2460 6360 3360 6810 72- 20 22OQ 3-8 3-16 1-4 1-4 3-8 1-4 3-16 3-16 1-4 3-8 3-16 3-16 3-8 5-16 5-16 1-4 5-16 1-2 3-8 3-16 3-16 3-16 3-16 1-2 3-8 1-2 1-2 1-4 1-4 1-4 1-4 WEBBER-HOLDING POWER OF RAILROAD SPIKES TABLE III-Continued TCind nf a ~ White Oak 32 31 33 Rock Elm 10 Red Elm 10 10 13 a 9 1 363 2 296 3 3494 Av. 336( 3 1 2 4001 3 4101 Av. 405 4 1 420' 2 420 3 590' Av. 483 5 1 225 2 326 3 291 Av. 281 2 11 14 1 192 2 ... 3 196 Resistance in Pounds for Pull of Q3 "Q ' n e 0 0 0 0 3 0 0 0 0 0 0 06 Av. 1940 1 3730 2 2800 3 3300 4 1600 5 ... Av. 2800 1 1240 2 3000 3 1760 Av. 2000 7500 6760 8270 7510 7490 8450 7920 7530 3850 5690 6530 7160 5880 6520 3770 4510 4140 7760 5820 6270 6070 7810 6950 6500 5970 6235 S 5340 4650 4250 4500 4800 5200 3980 5010 4730 4330 4390 3100 3940 4460 4620 4650 4580 6060 4420 5240 4310 4460 4120 5030 5210 4620 6430 5080 , 5750 4430 4330 3770 4240 4080 3790 3630 2790 3410 4420 3850 4340 3210 5310 4160 4730 3930 3580 3550 4140 4490 3340 4380 4960 4040 4460 Maximum Resistance 9640 10650 10750 10350 8380 7490 8450 8770 7330 7530 6590 7150 8280 7160 7300 7910 7410 7730 7570 7760 6800 7840 7700 7810 7600 9230 10040 8810 9350 Tie 3-8 3-8 5-16 3-8 3-8 1-4 1-4 3-16 3-16 1-4 3-16 3-16 5-16 1-4 3-8 5-16 7-16 3-8 3-8 1-4 5-16 5-16 5-16 1-4 5-16 7-16 7-16 3-8 5-16 623 18 ILLINOIS ENGINEERING EXPERIMENT STATION TABLE III-Continued Kind of Tie 6 Red Elm 13 13, Resistance in Pounds for Pull of 66 S 2 1 1930 3990 2 2240 3860 3 1960 4200 Av. 2040 4020 10 1 2810 4930 2 2450 4800 3 2140 5030 Av. 2460 4920 VV Inte Ell 12 1 I1 2 2810 5500 3 1890 5610 Av. 2150 5200 12 5 1 2460 5790 2 2140 5410 3 1770 5270 Av. 2120 5490 15 5 1 2630 6330 2 3810 7260 3 2810 6100 Av. .3080 6560 37 5 1 2490 8130 2 .... 5760 3 2790 6540 Av. 2640 6810 15 13 1 1600 4600 2 2230 5770 3 1450 4200 Av. 1760 4860 12 10 1 1990 3560 2 1920 4320 3 1830 3930 Av. 1910 3970 4540 3760 5250 3970 4540 3850 4770 3890 4850 3510 4 4 ±;O0k I 0 k, 3790 3210 4520 3490 3270 2720 3530 3010 3620 2770 3470 2830 3590 2980 2830 2770 2520 2430 2980 2720 5580 4680 4310 4160 3620 4770 4000 3900 3660 4040 3330 3770 3420 3910 3470. 5900 5530 4820 9310 6190 7420 5630 3350 2700 3540 2690 2360 1650 3080 2010 Maximum Resistance I S 7730 8100 7120 7650 7550 7430 8690 7890 5330 5590 5610 5510 6280 5410 5270 5650 9500 9560 8050 9030 8130 6650 7460 7410 9670 8760 9310 9250 5370 5450 4100 4970 5-16 7-16 3-8 7-16 3-8 3-8 7-16 3-8 3-8 5-16 1-4 5-16 '5-16 1-4 1-4 1-4 3-8 3-8 5-16 3-8 1-4 5-16 5-16 5-16 7-16 3-8 1-2 7-16 3-8 3-8 1-4 3-8 I WEBBER-HOLDING POWER OF RAILROAD SPIKES .TABLE III-Continued Resistance in Pounds Maximum for Pull of Resistance Kind of Tie 6 - a White Eln 15 12 1 .... 5590 4670 3810 7860 3-8 2 2430 5630 3870 3350 6140 5-16 3T. 2600 5550 3140 .... 5550 1 -4 Av. 2510 5920 3890 3580 6520 5-16 37 2 1 2100 4710 3390 3330 7170 3-8 2 2920 5160 4070 3840 6310 5-16 -~ - 3 3150 5900 4300 3860 757860 3-8 3. 2600 5550 3140 .... 5530 14 Av. 2390 5290 3920 3680 7020 3-8 Beech 14 6 1 .... .... 5390 4670 7680 1-4 2 2870 5150 4870 4320 7190 3-8 3 2230 5660 5310 4940 7820 5-16 Av. 2550 5400 5190 4470 7560 5-16 36 6 1 4330 4740 4400 4510 7120 3-8 2 3610 8100 6230 5320 8560 1-4 3 2640 8120 5470 4640 9080 5-16 Av. 3530 6990 5030 4820 8250 5-16 14 2 1 2550 4740 4570 4100 7670 5-16 2 2200 5570 5690 4190 8170 3-8 3 2120 4910 4800 3970 7860 3-8 Av. 2290 5070 4700 4090 7900 3-8 36 2 1 3210 5940 4010 3840 8460 3-8 2 3110 6900 4170 3900 10400 3-8 3 2120 5440 5240 4130 8270 5-16 Av. 2850 6090 4470 3960 9040 3-8 14 9 1 2660 5070 .... 3560 8130 3-8 2 1500 2820 9910 5060 9910 1-2 3 1490 3810 8900 5280 9220 7-16 Av. 1880 3900 8960 4630 9090 7-16 36 9 1 2130 4900 4240 4210 9890 3-8 2 2940 6640 3860 3650 9430 5-16 3 2370 4920 3830 3600 8900 3-8 Av. 2480 5490 3980 3820 9410 3-8 20 ILLINOIS ENGINEERING EXPERIMENT STATION Kind of Tie Poplar TABLE III-Continued 6 11 11 19 Chestnut 40 40 40 40 0 0 Resistance in Pounds for Pull of Maximum Resistance .0 .0 .0 0 0 0 0 0 0 0 0 0 p -± ~. 2 1 2700 4690 3980 3520 4690 2 4690 .... 3240 2890 4980 3 3000 5100 3240 2900 5100 Av. 3460 4890 3490 3100 4920 2 1 2750 4510 4400 3980 6990 2 2710 5270 2840 2610 5270 3 3100 6050 4080 3650 6050 Av. 2850 5240 3760 3410 5900 12 1 2220 5130 2960 2750 5350 2 2750 4940 3800 3590 5070 Av. 2480 5040 3280 3170 5210 12 1 2610 5670 .6250 2 2460 6220 4400 4170 7040 Av. 2530 i 5990 4400 4170 6650 14 1 2300 3100 2410 2260 4300 2 2330 2600 2860 2460 4060 3 3730 3370 2370 2100 5050 Av. 2490 3060 2540 2270 4470 5 1 .... .... 5830 2 3010 2720 2650 2650 5180 3 3300 3570 2950 2400 5500 Av. 3150 3150 2800 2520 5510 12 1 3320 6230 3050 2270 6230 2 5110 .... .... .... 5110 3 2000 4000 2490 2430 4000 Av. 3480 5110 2770 2350 5110 4 1 1300 3780 3170 2940 5420 2 2300 5420 3360 2780 5420 3 2440 5640 3190 2590 6220 Av 950 I 4950 3240 2770 5690 CI~ u ~ 1-4 3-16 1-4 1-4 3-8 1-4 1-4 1-4 5-16 5-16 5-16 5-16 5-16 5-16 3-16 3-16 3-16 3-16 3-16 3-16 3-16 3-16 1-4 1-8 1-4 1-4 3-16 1-4 5-16 1-4 .. . 2 . .. . .. . . . . . .. . .... ... . I WEBBER-HOLDING POWER OF RAILROAD SPIKES TABLE III-Concluded Resistance in Po for Pull of Kind of Tie 6 I - Loblolly 39 14 1 3390 2970 2620 2 3760 3980 279( Pine 3 4050 2860 202( Av. 3730 3270 248( 21 14 1 2880 4550 237( 2 1980 2110 189( 3 4510 3910 334( Av. 3120 3520 256( 20 5 1 2250 4550 293( 2 2810 2670 272( 3 3610 .... ... I 4 1890 2690 229 Av. 2640 3270 265( 21 10 1 3570 4450 250 2 2550 4890 340( Av. 3060 4670 295( 20 3 1 3090 4800 273( 2 2610 2330 230< 3 1870 3810 251 Av. 2860 3650 251 39 6 1 3110 2120 2171 2 1560 3880 3061 3 1630 3330 2641 Av. 2100 3110 296' unds 0 0 0 0 0 0 0 0 0 0 14 2590 2420 1850 2290 Maximum Resistance 1870 4550 1-4 1570 3520 3-16 2880 5200 3-16 2110 4420 3-16 2540 4550 1-4 2640 4570 3-16 ... 3610 1-8 2030 3710 3-16 2410 4110 3-16 2230 4450 1-4 3020 4890 1-4 2630 4670 1-4 2320 4800 1-4 2030 3440 3-16 2280 3810 1-4 2210 4020 1-4 1710 3110 1-8 2380 3880 1-4 2650 3330 1-4 2250 3440 1-4 A study of the results of Table III has been made to determine: (A) Comparative holding power in untreated ties; (B) Comparative holding power in treated ties; (C) Comparative holding power of the same timber, treated and untreated; (D) Effect of preservative on the holding power; (E) Relation between the cross section of the spike and holding power; (F) Relation between the. depth of pene- I 22 ILLINOIS ENGINEERING EXPERIMENT STATION tration and the holding power; (G) Effect of the point of the spike on the holding power; (H) Effect of bored holes on the holding power; (I) Effect upon the holding power of re-driving the spike. A Comparative Holding Power in Untreated Ties Table IV is compiled from Table III to show the average holding power for different untreated ties. Each result in Table IV is the average of the corresponding results in Table III. TABLE IV AVERAGE HOLDING POWER IN UNTREATED TIES Resista a Pounc Kind a Pu ofTie I--- White Oak 10 30 3510 Elm 11 33 2310 Beech 3 9 2240 Chestnut 4 12 2990 Loblolly Pine 2 6 2920 nce in Is for 11 of Maximumi Resistance a) 0'Z f§ fi - Resistance in per cent of that in White Oak c a g 100 100 100 66 136 93 64 96 104 86 103 66 85 81 46 Table IV shows the comparative holding power of five kinds of timber. The last three columns show the holding power in terms of that of white oak. It is thought that a pull of 1-4 of an inch gives results which are of more value in comparing the hold- ing power of the different kinds 'of ties than the results for either greater or less distances, since the results for the 1-4-inch pull re- present the resistances of the various timbers to the withdrawal of the spike for a distance which should not be exceeded in practice, and since the maximum resistance and the results for a pull of 1-8 of an inch represent the resistances for distances which are there- fore not of so much consequence as the 1-4-inch pull. Notice that with chestnut and loblolly pine the maximum resistance oc- curs at 3-16 of an inch, which is a reason for comparing their max- imum resistance with that of white oak at 1-4 of an inch instead of with its maximum resistance, as in Table IV. If this is done, the efficiencies of chestnut and loblolly pine for a 1-4-inch pull or less are 131 and 85. per cent respectively. WEBBER-HOLDING POWER OF RAILROAD SPIKES The fact that the maximum resistance did not occur until the spike had been pulled from 3-16 to 3-8 of an inch is interesting. While the spike is being driven the fibers of the wood are bent downward and are pressed outward, and as the spike is withdrawn the friction between the spike and the wood tends to draw the fib- ers into their original position, which causes them to crowd lat- erally against the spike and also toward the surface of the tie, until finally the external pull exceeds the internal resistance and the spike slips. When the fiber structure is open, there is con- siderable cellular space for the displaced fibers to occupy, and therefore the maximum resistance is low, and is quickly attained; but when the fiber structure is compact, the reverse is true. As the loblolly pine ties should always be preserved, the re- sults in Table IV for this timber are of doubtful value. For the best results elm ties also should be treated; but as some species of elm do not absolutely require treatment, elm is properly included in Table IV. Arranging these timbers in the descending order of their resistances for a 1-4-inch pull, we have elm, chestnut, white oak, beech and loblolly pine. The maximum holding power for the first three timbers in Table IV is satisfactory, but that for the last two is quite low. The last fact indicates that when timber of the softer varieties or timber having loose fiber structure is used for ties, some more efficient form of fastening should be devised. B Comparative Holding Power in Treated Ties Table V is compiled from Table III to show the average holding power obtained with various treated ties, each result in this table being the mean of the corresponding values in Table III. The average results obtained with untreated white oak are also included so that comparisons can be made. The average for the resistances for all of the treated timbers is shown at the foot of the table. Excluding the last two timbers, the average resistance for the 1-4-inch pull is 5690 pounds. The maximum resistance of the last two timbers should be averaged with the resistances of the others for the 1-4-inch pull, in which case the average resistance for all of the timbers for a 1-4-inch pull or less is 5400 pounds. Table V shows that the resistances of the several timbers do not differ widely, and that the soft timbers give results which 24 ILLINOIS ENGINEERING EXPERIMENT STATION TABLE V AVERAGE HOLDING POWER IN TREATED TIES Kind of Tie White Oak (Untreated) Water Oak Black Oak Red Oak Burr Oak Ash Elm Beech Poplar Loblolly Pine Sweet Gum Av. cc 10 16 13 20 3 2 .5 3 4 4 5 3. . -n . . Maximum . Resistance in Resistance Resistance in Pounds for Resistance per cent of that , a Pull of of White Oak vi f. 0? 0 6 .5 --S g ^g - ^ ~ - 0of ^ Q ^ c ____ r-_ i I _ _________ - 30 3510 48 2870 SQ 901nA 60 9 6 15 9 12 12 15 2950 2670 3570 2590 2950 2830 2920 3230 2950 3950 5730 5890 5350 5750 5200 5940 6190 5290 3780 5320 5320 7870 5-16 6780 5-16 7230 5-16 7730 5-16 9210 3-8 7730 5-16 7500 8900 5670 4310 5300 7040 5-16 3-8 5-16 1-4 3-16 100 82 83 84 76 101 74 84 81 83 0 100 145 149 135 145 131 150 157 134 109 92 96 84 135 100 86 92 98 117 98 96 113 72 55 67 89 compare favorably with those obtained for the hard woods. This table also shows that the range for the maximum resistances is much greater than that for either the 1-8-or the 1-4-inch pull. The resistances for the different species of oak are very nearly the same, the mean for a 1-8 inch pull being 2850 pounds, for a 1-4-inch pull 5680 pounds and for the maximum 7740 pounds. Notice that with nearly all of the timbers the maximum resistance was obtained after the spike was pulled more than 1-4 of an inch, but there is no apparent relation between the amount of the holding power and the distance through which the spike has been pulled. Comparing the resistances of treated timbers with that of un- treated white oak, we see that the initial resistance of the white oak is higher than any of the other woods except one; while on the other hand, the resistance at 1-4 of an inch in white oak is less than in any of the other woods save one. The maximum resis- tances of all but the last three timbers are practically the same. Considering the uniformity of the results obtained with a pull of 1-4 of an inch in the few timbers which were available, there appears to be no strong reason for much discrimination between the different treated timbers. WEBBER-HOLDING POWER OF RAILROAD SPIKES C Comparative Holding Power of the Same Timber, Treated and Untreated Table VI has been compiled from Table III for the purpose of studying the effect of the treatment upon the holding power of a timber. TABLE VI RELATIVE HOLDING POWER IN TREATED AND UNTREATED TIES Kind of Tie , Elm 3 27 2 15 Beech 1 9 1 9 Loblolly 1 6 Pine 2 12 Red Oak 3 15 4 21 Condition of Tie Untreated Treated Untreated Treated Untreated Treated Untreated Treated Resistance and Gain in Pounds Due to Treatment 2310 2590 2240 2950 2920 2920 280 710 000 5390 5940 550 3790 6190 2400 3190 3730 640 7290 7500 210 8180 8900 820 3630 4310 680 6460 7730 1270 Table VI shows that higher resistances are developed in treated than in untreated ties. The average increase due to treat- ment for a 1-8 inch pull was 330 pounds; for a 1-4 inch pull, exclud- ing the seemingly unreasonable increase in beech, 685 pounds; and for the maximum resistance 747 pounds. Considerable reliance is placed upon the conclusions drawn from Table VI, inasmuch as the methods of making the tests were exactly the same for the treated and untreated ties, and since the same number of spikes, fifty-seven, was used in both cases, and also since the preserved ties were treated by different processes and at different plants. The increased resistance due to treatment has two causes: (1) The presence of the preservative in the cells, thus reducing the space into which the fibers can crowd as the spike is withdrawn; and (2) The hardening of the fibers by the steaming, preparatory to treatment, which renders them less pliable. 26 ILLINOIS ENGINEERING EXPERIMENT STATION The movement which took place among the fibers near the surface of the tie is interesting. In the untreated ties there was a crumpling of the fibers close to the spike, while the fibers in the treated ties were torn out in deep slivers extending from the spike to the blocks which supported the tie. D Efect of the Preservatives on the Holding Power Three distinct kinds of preserving solutions were used in the ties tested,-creosote, zinc-creosote and zinc-tannin. Table VII has been compiled from Table III to study the effect produced by the treating solution upon the holding power of the tie. Table VII does not show any marked difference between the resistances in ties treated with the different preservative solu- tions. For example, the maximum resistance of the red oak is lower when treated with zinc-tannin than when treated with zinc- creosote, but the reverse is true of the initial resistance of the red oak and also of the maximum resistance of black oak. With elm the initial resistance is higher in creosoted ties than in those treated with zinc-creosote, but the maximum resistance is lower. If any rating were made in order of efficiency, it would appear about as follows: (1) creosote, (2) zinc-creosote, and (3) zinc-tan- nin. However, there are too many uncertain quantities involved to make such a rating reliable; and morever, the effect of the treating solution upon the holding power is only one of the many elements which must be considered when choosing between the different treating solutions. E Relation between the Cross Section of the Spike and the Holding Power The question to be answered here is, which size of spike will develop the highest holding power. To answer this question, Table VIII showing the relation between the cross section and the holding power has been compiled from Table III. From a study of the results of Table VIII it will be noticed that no general rating can be made for the various sized spikes in order of the resistances developed, since the spike which develops the lowest holding power for the 1-8- inch or the 1-4-inch pull seldom develops the highest maximum resistance. For example, in white oak, the 19-32-inch spike developed the highest resistance for the WEBBER--HOLDING POWER OF RAILROAD SPIKES TABLE VII EFFECT OF DIFFERENT PRESERVATIVES ON THE HOLDING POWER Resistance in Pounds for Kind of a Pull of Maximum Kind of Tie No. Preservative Resistance, T - Pounds Comparison of Zinc-Tannin and Creosote Water Oak 4, 5, 25, 26, 29 Zinc-Tannin 2380 5010 6260 34 Creosote 3020 6270 7310 Red Oak 6, 9, 22, 28, 30 Zinc-Tannin 3170 5470 6580 41 Creosote 3120 5800 6920 Comparison of Zinc-Creosote and Creosote Red Oak 7, 8 Zinc-Creosote 2350 4940 8500 41 Creosote 3120 5800 6920 Elm 10 Zinc-Creosote 2520 5870 7690 37 Creosote 2600 6350 7210 Comparison of Zinc-Tannin and Zinc-Creosote Red Oak 6, 7, 8, 9, 22 Zinc-Creosote 2350 4940 8500 28, 30 Zinc-Tannin 3170 5470 6580 Black Oak 16, 18 Zinc-Creosote 2850 5620 7040 23, 24, 27 Zinc-Tannin 2830 5620 7550 1-8-inch pull, but the 9-16-inch spike developed the highest resis- tance for the 1-4-inch pull, and also the highest maximum resis- tance. In black oak the highest resistance for the 1-8-inch pull was developed by the 9-16 spike, but that for the 1-4-inch pull was developed by the 19-32-inch size and the maximum resistance by the 5-8-inch spike. Averaging all of the resistances for the 1-8-inch pull, the 1-4-inch pull and the maximum resistance collectively, we see that the average holding power of the 9-16-inch spike is 4990 pounds, for the 19-32-inch spike 5420 pounds and for the 5-8-inch spike 5290 pounds. Because of the large number of spikes tested, seventy-two 9-16-inch, thirty-six 19-32-inch, and one hundred and two 5-8-inch, and the irregularity of the results, it was decided that no conclusions could be drawn from Table VIII as to the relative holding power of the different sizes of spikes. However, the thick- 28 ILLINOIS ENGINEERING EXPERIMENT STATION TABLE VIII RELATION BETWEEN THE CROSS SECTION OF THE SPIKE AND ITS HOLD- ING POWER Resistance to With- drawal, Pounds Kind H Condition of Tie of Tie . 0 CC S.i  ' .0 ___ _^^ ___ m l^ Seasoned Treated Treated Treated Seasoned Treated Treated 9-16 19-32 5-8 9-16 19-32 5-8 9-16 19-32 5-8 9-16 58 9-16 19-32 5-8 9-16 19-32 5-8 9-16 5-8 3110 6280 8760 3750 5380 7620 3650 6030 7620 2910 5340 6530 2650 6130 7130 2550 5710 7240 2960 5560 6670 2970 5310 6010 2650 5360 6730 2300 4760 7650 3260 5990 6780 1880 3900 9410 2550 5400 7660 2290 5070 7900 2480 5490 9410 3530 6990 8250 2850 6090 9040 2190 3770 4610 3490 4450 5460 ness of the spikes varied by only 1-16 of an inch or about 10 per cent, and their areas by only 0.075 of a square inch or about 20 per cent. To test still further the relationship between the size of the spike and the holding power, a series of experiments was made with plain square rods with the results shown in Table IX. Each result is the mean of fifteen tests in a single kind of timber. White Oak Black Oak Water Oak Red Oak Beech Sweet Gum 2 2 3 4 2 4 5 6 5 7 9 1 1 1 1 1 1 1 1 9 6 15 15 6 18 15 18 15 21 36 3 3 3 3 3 3 6 12 WEBBER-HOLDING POWER OF RAILROAD SPIKES TABLE IX EXPERIMENTS WITH PLAIN SQUARE RODS IN BEECH TIMBER Increase .for each Increment Size of Area, Area Resistance Rod * sq. in. Z- - - Zo  square per ounds per inches cent cent Successive increments in the size of the rod = 1-8 inch 1-2 inch square 0.250 6280 ..... 5-8 inch square 0.391 6970 0.141 53 690 11 3-4 inch square 0.562 9070 0.171 44 2600 37 7-8 inch square 0.765 9380 0.203 35 310 3 Successive increments in the size of the rod = 1-16 inch 8-16 inch square 0.250 6280 ..... 9-16 inch square 0.316 6450 0.066 25 170 3 10-16 inch square 0.391 6970 0.075 23 520 8 It will be seen from the results in Table IX that there is an irregular increase in the holding power as the size of the rod is in- creased. Notice that with increments of 1-8-inch, the successive increments in the resistance are at first large, but with the last rod this increment suddenly falls to practically nothing. This drop in the increment is principally due to the tendency of the large rod to split the tie. The results with 1-16-inch increments do not differ materially from those in the first part of the table. The deduction for Table IX is that the holding power will be increased as the size of the rod is increased, but that it is not ex- pedient to use rods (or spikes) larger than 3-4 of an inch unless holes are bored for them. F Relation between the Depth of Penetration and Holding Power A series of experiments was made to determine the relation between the depth of penetration and the holding power. The results are given in Table X. 30 ILLINOIS ENGINEERING EXPERIMENT STATION TABLE X HOLDING POWER IN A WHITE OAK TIE WITH VARYING DEPTHS OF PENETRATION Resistance, Pounds Test Number 1 150 1000 2250 3840 3800 5980 7190 7510 9070 3 1 2 3 4 4 6 7 8 4 160 510 1320 2900 2940 4220 4500 5850 7340 7790 140 500 760 050 050 200 210 310 720 540 Average 170 150 490 500 950 1290 2760 2450 3570 3360 4810 4210 5860 5060 6080 6270 7520 7900 8340 The spikes had a taper point approximately 1 inch long. Plate IV shows that the holding power varies directly with the penetration, not counting the taper point. It is impracticable to use a spike longer than 5 1-2 inches in a 6-inch tie, since a longer spike would either pass entirely through the tie or sliver it on the under side. In either case the fiber adjacent to the spike would quickly decay owing to the access of water. In a thicker tie, however, a longer spike could be used advantageously. The main precaution is to keep the spike from damaging the under surface of the tie, otherwise-the longer the spike the greater the hold- ing power. Depth of' Penetration 1-2 in. 1 in. 1 1-2 in. 2 in. 2 1-2 in. 3 in. 3 1-2 in. 4 in. 4 1-2 in. 5 in. WEBBER-HOLDING POWER OF RATLROAD SPIKES PLATE IV £7e&/o/ of 0 ene/ra'on,7, /rnches. e000 7ooo 7000 6000 4000 3000 6000 Curv~e ///uasfroy ResI/sance /o V///,/j'rawa/ or , V/r'ous De/olAs of /ene/ra/,on. 32 ILLINOIS ENGINEERING EXPERIMENT STATION G Ef'ect of the Point of the Spike on the Holding Power There were three distinct types of points on the spikes,-blunt- point, chisel-point and bevel-point. *S/i/'fe. 5p /5fe. pi"e. Fia. 1 FORMS OF POINTS OF SPIKES The average results obtained with spikes having these types of points have been compiled from Table III, and are shown in Table XI. The average and relative resistances of each type of spike for all timbers are shown at the foot of the table. These averages show that both the blunt-pointed and the bevel-pointed spike are higher in holding power than the chisel-pointed spike. Since the average resistances of the blunt and the bevel-pointed spikes are practically the same, and since the blunt-pointed spike develops the highest resistance for the 1-8-inch and the 1-4-inch pull the greatest number of times, the blunt-pointed spike is first in point of efficieicy, although the bevel-pointed spike is a close competitor under all conditions. The chisel-pointed spike is last. The two upper figures of Plate V are the two halves of a red- oak tie showing the position of the fibers adjacent to the spike; and the lower figure is a portion of the other end of the same tie split after the spikes had been pulled out. The photograph was taken immediately after the tie had been split. The figures are too small to show details clearly, but an examination of the tie showed that the blunt-pointed spike disturbed more fiber than either the chisel or the bevel-pointed spikes, the last two disturbing about the same amount. The examination also showed that the blunt- pointed spike tore rather than cut the fibers, and deposited them in unequal bundles along its faces, while the chisel-pointed spike cut the fibers and deposited them quite uniformly both across and U ~V tV PLATE V EFFECT OF SPIKES IN DISPLACING THE FIBERS OF THE TIE WEBBER-HOLDING POWER OF'RAILROAD SPIKES TABLE XI EFFECT OF THE FORM OF THE POINT OF THE SPIKE ON THE HOLDING POWER Type of Point Chisel Bevel Blunt Chisel Bevel Blunt Chisel Bevel Blunt Chisel Bevel Chisel Bevel Blunt Chisel Bevel Blunt Chisel Bevel Blunt Chisel Bevel Blunt Chisel Bevel Resistance in Pounds for 1-8 in. Pull 1-4 in. Pull Pounds iva- Pounds Riela- 5520 5440 6890 5690 5560 4400 5350 5580 '7040 5190 5610 5240 5740 4670 5580 6190 4950 3060 4130 3650 3390 5010 5340 4810 5490 Maximum Resistance Pounds 6540 6330 8280 6930 6800 5760 7630 7370 8760 7090 8010 7710 7050 9250 8470 7900 5690 4470 5310 4020 4120 5520 6960 6610 6800 Rela- tive 100 97 119 100 98 76 100 97 123 100 113 100 92 109 100 93 127 100 119 97 100 134 105 100 103 in front of each face. The bevel-pointed spike forced a majority of the fibers to the front face and toward the corners. The rela- tively high holding power of both the blunt and the bevel-pointed spikes is due to this unequal concentration of the fibers. Kind of Tie Water Oak Black Oak Red Oak White Oak Elm Beech Chestnut Loblolly Pine Average for all Timbers 34 ILLINOIS ENGINEERING EXPERIMENT STATION H Effect of Bored Holes on the Holding Power A series of tests was made to study the effect of boring holes for the spike. The first step was to determine the proper size of the hole. Table XII shows the summary of a series of tests made at the University of Illinois in 1891* to determine the relationship between the holding power and the "drift". TABLE XII RESULTS OF EXPERIMENTS WITH SQUARE DRIFT-BOLTS IN PINE TIMBER Size of Holding Power, Pounds Size of Drift-Bolt Hole, Drift, - inches inches 6-inch Per inch depth depth 1 inch square 16-16 .... 3972 662 1 inch square 15-16 1-16 4260 710 1 inch square 14-16 1-8 4660 777 1 inch square 13-16 3-16 4050 675 This table shows that with 1-inch square drift-bolts a drift of 1-8 of an inch gives a maximum holding power, but that a drift of 1-16 of an inch gives nearly as much resistance. It is not known that this relation holds with bolts less than 1-inch square, but the author assumed that this was sufficient reason for using a drift of 1-16 and 1-8 of an inch in this investigation, which conclusion is in accord with the usual railroad practice. The second step was to determine the resistance to the differ- ent sized spikes in different kinds of ties. The detailed results for these experiments are given in Table XIII. Notice that the results are arranged according to the drift. The average results from Table XIII are shown in Table XIV along with the results from Table III for the same spike driven in the ordinary way. The average resistances for all timbers, recorded at the foot of Table XIV, show that for a pull of 1-4 of an inch or less the spike driven into a bored hole develops higher holding power than one driven in the ordinary way. For a 1-4-inch pull or less the rela- tive resistances show a marked increase in a majority of cases, but the maximum resistance for spikes driven into bored holes is usually the lowest. * Technograph No. 5, 1891. University of Illinois WEBBER-HOLDING POWER OF RAILROAD SPIKES TABLE XIII HOLDING POWER OF ORDINARY SPIKES IN BORED HOLES Kind of Tie Water Oak c3.2 9-16 Maximum Resistance 0 0 1-2 Av. 1-2 Av. 1-2 Av. 9-16 Av. Resistance in Pounds for Pull of 00 0 0 Hole 1-16 in. Smaller than Spike 2330 3860 3660 3180 2050 3860 3970 3320 2020 6470 4740 4010 1660 4450 4090 3890 2500 6400 4120 3600 3250 3750 3440 3120 2390 4890 3930 3080 2310 4810 3990 3410 3460 6770 3570 2850 3000 7120 3810 3360 4590 6810 3550 3350 2670 6350 3850 3560 2910 6710 3390 2970 2260 6720 3810 3270 3150 6750 7310 3660 3970 6550 3500 3140 3920 6930 3250 3720 2180 5920 4590 3900 2830 6900 3770 3320 2660 4310 3440 2720 2870 5710 4090 3410 2900 6100 3380 3100 3950 6680 4690 4040 2700 7430 3410 3420 2680 7410 3950 3420 3070 6390 3810 3420 3000 5380 3610 ........ 3300 5010 3360 ........ 3130 6240 3540 3510 2710 6530 4070 3600 2600 5460 5160 4170 2850 5810 4860 4400 3130 6800 6980 4950 2950 5890 4390 4140 5740 5730 6750 6460 6400 4940 6740 6110 7190 8190 6810 6350 6710 8630 3230 6830 6930 6990 6900 5320 5710 6100 6680 7480 7410 6640 5380 5010 6240 7040 6990 8800 9420 6960 Black Oak Red Oak 02 5-16 3-8 5-16 5-16 1-4 3-16 5-16 5-16 5-16 1-4 1-4 1-4 1-2 5-16 1-4 5-16 1-4 5-16 1-4 1-4 1-4 1-4 1-4 1-4 1-4 1-4 5-16 5-16 5-16 9-16 9-16 5-8 6960 36 ILLINOIS ENGINEERING EXPERIMENT STATION TABLE XIII-Continued 9-16 5-8 9-16 9-16 1-2 Av. 9-16 Av. 1-2 Av. 1-2 Av. 9-16 Av. Resistance in Pounds for Pull of 4080 7210 2510 6540 1980 4850 2850 5840 2530 5760 2790 6040 3920 4700 2840 6300 1660 5100 2100 6340 2630 5610 2960 6820 2910 5710 2890 5610 2830 2900 3360 5450 3360 6610 3770 6780 2870 6930 3540 5110 3150 5770 2850 5840 2530 5760 2250 6210 2630 3940 2790 5220 2610 6300 2610 5550 3030 3080 2620 5760 2850 3840 2830 4230 4720 3360 4380 3220 3510 3840 3860 4070 5300 5150 4590, 3820 4010 3240 2800 2940 3680 3470 4740 5060 3750 3220 3510 4640 3350 4220 3900 3810 2740 3560 3290 3200 3300 3180 4050 2290 2730 3090 3280 3600 4370 4540 3950 3790 3550 2850 2690 2620 3210 2890 4360 4010 3330 2290 2730 3570 2870 3680 3370 3080 2320 2940 2730 2660 Maximum Resistance 8180 8380 8830 6180 5760 7460 6460 6480 8510 8760 7550 7100 7270 5610 6000 5450 6610 8200 6930 7640 6750 6180 5760 7170 4940 6010 6300 6060 4370 5760 5500 5210 5-16 5-16 7-16 5-16 1-4 3-16 5-16 5-16 5-16 .3-16 5-16 1-4 3-16 1-4 1-4 3-8 1-4 3-8 5-16 1-4 5-16 3-16 3-16 1-4 3-16 1-4 3-16 Kind of Tie Ash Beech Sweet Gum .a a 0 0 'a a op -* 1-1 1-1 WEBBER-HOLDING POWER OF RAILROAD SPIKES TABLE XIII-Concluded Kind of Tie Red Oak Beech Resistance in Pounds for Pull of Hole 1-8 in. Smaller than Spike a) 0(J) a) 0 1-2 Av. 7 -16 Av. 1-2 Av. 4190 3950 3440 2100 2190 2500 3060 3620 4100 2640 3360 3250 3810 3630 4150 3420 4360 3630 2320 2540 2240 2370 Maximum Resistance 7270 6860 5850 4410 4410 6030 5800 7080 6920 6000 7280 6940 7830 6150 7950 8230 7660 7200 3980 4750 5200 4640 1800 5710 2340 . 6860 2630 5850 3170 4410 4070 3000 4720 4220 3270 5010 1340 4560 2540 5620 4000 4720 . 3560 7280 3580 5270 3240 6900 2510 6150 2290 4620 2790 6380 1900 3790 2180 5530 2400 ... 2850 3180 2950 3700 2730 3430 Sweet Gum 5000 4490 4010 2570 2600 2550 3540 3530 4720 3000 3800 3800 4020 3710 5410 4630 5010 4160 2710 2920 3300 2980 5-8 9-16 5-8 38 ILLINOIS ENGINEERING EXPERIMENT STATION TABLE XIV AVERAGE RESISTANCE OF SPIKES WITH AND WITHOUT BORED HOLES Kind of Tie Water Oak Black Oak Red-Oak Beech Ash Sweet Gum Av. for all Timbers Red Oak Beech Sweet Gum Av. for all Timbers C) rJ~ 0 C) cj~ VI 0 6 Resistance in Pounds for - ¢o -s Drift 1-16 of an inch Hole 2310 4810 6110 No Hole 2960 5660 6670 Hole 3300 6750 7310 No Hole 2970 5320 6490 Hole 3070 6390 6640 No Hole 3260 5450 6820 Hole 2950 5890 6960 No Hole 2310 4760 7660 Hole 3150 5770 6760 No Hole 2180 4700 9410 Hole 2790 6040 7460 No Hole 4150 4630 6810 Hole 2610 5550 6060 No Hole 2190 3730 4610 Hole 2830 4230 5210 No Hole 3460 4450 5460 Hole 2930 5680 6570 No Hole 2880 4840 6740 Drift 1-8 of an inch Hole 3270 5010 5800 No Hole 2310 4760 7660 Hole 2780 5530 7200 No Hole 2180 4700 9410 Hole 2730 34 30 4640 No Hole 3460 4450 5460 Hole 2930 4660 6550 No Hole 2650 4640 7510 Relative Resistance P.ý 4- 92 100 113 100 97 100 91 100 72 100 110 100 131 100 96 100 98 100 75 100 77 100 85 100 87 100 00 WEBBER-HOLDING POWER OF RAILROAD SPIKES As far as conclusions can be drawn from these experiments, the spike driven into a bored hole is superior to one driven" in the ordinary way. I Effect upon the Holding Power of Be-driving the Spike In practice, when the spike is pulled out of the tie a moderate distance, it is driven back, provided the hole is not greatly en- larged. If the hole is much enlarged the spike is driven at an- other point. This constant re-spiking rapidly ruins the tie. A ser- ies of tests was made to determine the effect upon the holding power of re-driving the spike. The average maximum holding power of the re-driven spikes is shown in Table XV along with the original maximum holding power of the same spike. It will be seen that the holding power of the re-driven spike is very much less than that of the newly-driven spike. The resist- ance is affected so much in some woods as to make the practice of TABLE XV RELATIVE HOLDING POWER OF NEWLY-DRIVEN AND RE-DRIVEN SPIKES Kind of Tie Ash Water Oak Red Oak Elm Poplar Sweet Gum Average Maximum Re- U- sistance, Pounds 0 Per ent nf Original 6 8640 6 8020 6 8030 6 7910 6 4920 6 5040 Original After Re- driving 6490 5760 5230 4840 3980 4150 75 72 65 61 81 82 re-driving the spike a questionable procedure if the holding power alone is considered; but as the practice of re-driving the spike helps to lengthen the life of the tie, the practice can not be justly condemned so long as the holding power is not excessively re- duced. ART. 2 HOLDING POWER OF SCREW SPIKES WITHOUT LININGS A series of tests was made to determine the holding power of screw spikes. The tests were conducted in the same manner as those with the ordinary spikes. 40 ILLINOIS ENGINEERING EXPERIMENT STATION The screw spikes were received from the following compan- ies: No. 1 from the Illinois Central Railroad Company; No. 2 from the American Iron and Steel Manufacturing Company, Scran- ton, Pennsylvania; No. 3 from the South Side Elevated Railroad Company, Chicago, Illinois; No. 4 from the Oliver Steel and Iron Company, Pittsburg, Pennsylvania; and No. 5 from the Pennsyl- vania Railroad Company. A description of the different spikes is given in Table XVI. TABLE XVI DESCRIPTION OF SCREW SPIKES Diameter Projection h of Diameter Z Length, Diameter Pith, Depth of of Bored S enh of Core, of Thread, itch, Insertion, Sinchesinches inchinches Hole, i inches 1 5 21-32 3-16 1-2 4 1-2 11-16 2 5 11-16 1-8 1-2 4 1-2 11-16 3 5 1-4 11-16 1-8 1-2 4 3-4 11-16 4 5 1-2 11-16 1-8 1-2 5 11-16 5 5 21-32 3-16 1-2 4 1-2 11-16 The shank or threaded portion of the spike was usually' 7-8 of an inch in diameter, and approximately one inch of the upper por- tion of the core tapered from the diameter of the core to that of the shank. The hole bored for the spike was not reamed, and the result was a tight fit between the wood and the spike. This tight contact is gained in practice by the head of the spike bear- ing against the base of the rail. The spike was driven by means of a wrench, the thread cutting its own path. The number of screw spikes obtainable was not sufficient to make as long a se- ries of tests as with the ordinary spikes. A study of the results with this spike has been made to de- termine: (A) Relation between the depth of penetration and the holding power; (B) Relation between the holding power of the screw and of the ordinary spikes; and (C) Influence of certain details of the screw spike upon its holding power. The detailed results of the tests with screw spikes are given in Table XVII, and the average results are shown in Plates II and III. WEBBER--HOLDING POWER OF RAILROAD SPIKES TABLE XVII DETAILED RECORD OF TESTS WITH SCREW SPIKES Resistance in Pounds for Maximum Sa Pull of Resistance .2 ^ 2___________---- of Tie- . . . o o o o 0 0 0 0 666 11 0 -. -.4 Blue Ash Sweet Gum 2 2 1 7350 2 5080 3 7520 Av. 6650 1 4 1 3320 2 3740 3 4350 Av. 3800 3 2 1 3810 2 5790 3 4270 Av. 4620 3 4 1 5920 2 4550 3 4780 Av. 5080 34 3 1 4820 2 4670 3 4680 Av. 4720 26 2 1 4110 2 3670 3 5270 Av. 4350 16 3 1 5520 2 4860 3 4260 Av. 4880 23 2 1 5850 2 4910 3 1090 Av. 3950 10900 9930 11650 10830 7480 7570 9200 8080 4940 7100 6030 6060 9000 7400 7120 7840 10230 9170 7030 8810 8010 7420 7790 8290 12370 11410 9870 11220 10290 10780 6370 9150 11650 6270 13470 6010 12300 6220 12470 6160 10840 6520 9410 5940 6800 4870 9010 5780 4870 2420 4900 3280 4620 2820 4790 2840 6000 4000 5600 3410 5090 3290 5560 3560 14530 9630 12140 10000 14360 9660 13680 9800 7190 3490 7850 3970 5190 3540 6740 3660 16930 10720 13100 7390 9760 7690 13260 8600 9460 6600 8590 6000 10400 7100 9380 6560 3370 3190 3030 3190 5000 4560 3260 3940 1900 3770 3450 3040 2900 2300 1800 2330 4600 6260 4490 5100 2150 2790. 2600 2510 6200 4050 3970 4740 4200 2500 6000 4230 Q-o 7-16 1-2 1-2 1-2 1-2 1-2 3-8 1-2 7-16 1-4 3-8 3-8 7-16 3-8 3-8 3-8 1-2 5-8 1-2 7-16 7-16 3-8 7-16 7-16 1-2 7-16 3-8 7-16 3-8 5-8 1-2 Tfl 13360 13470 12300 13040 10840 9410 9700 9980 5980 7100 6590 6560 9720 8100 7870 8560 14530 12640 14360 13840 9620 8900 8060 8860 16930 14350 12160 14480 12500 12570 10400 11820 Water Oak Black Oak 42 ILLINOIS ENGINEERING EXPERIMENT STATION TABLE XVII-Continued Resistance in Pounds for a Pull of C C 7810 12720 11440 11590 9770 11130 2720 6390 4240 4450 Kind of Tie C3 00 Maximum Resistance Beech White Oak Elm 4 2 3 4 3 3 3 3 2 3 1 2 3 Av. 1 2 3 Av. 1 2 3 Av. 1 2 3 Av. 1 2 3 Av. 1 2 3 Av 1 2 3 Av 1 2 11810 12780 14430 14800 12670 8320 10820 11270 10140 10240 17360 14700 14000 8180 9390 7560 8380 5450 6600 5530 16450 11370 Broke 13910 14190 13200 13230 13540 3940 7890 4220 5350 2610 8320 5190 5040 6330 6130 8240 6900 3010 7950 8210 6390 5000 4600 5880 5160 6420 8590 4420 6480 9670 9780 13860 9780 11140 8400 12370 11880 10850 11980 12980 15620 13530 9340 12490 12080 11300 8290 8030 9370 8560 11300 14190 13000 12830 36 14 31 31 32 4310 8290 5040 10920 4200 9130 4520 9450 Av I 7970 7050 9160 8060 9560 7990 2350 9960 4170 6130 6880 5730 4480 9930 8900 7770 6390 5350 4950 5230 2960 3340 3150 9590 5490 7040 6340 7950 7350 7210 4) 'o g ) C1^ p,. °:. Ch . 0 3710 12720 3600 12770 4970 11790 4130 12430 4880 13590 4530 15200 6500 14800 6300 14860 .... 14560 .... 13180 .... 14310 .... 14020 3230 14550 3900 17360 5890 16450 4340 16120 4530 11630 2880 13200 3290 12740 3570 12520 .... 8290 .... 8700 .... 10530 .... 9150 4360 16450 3190 15580 .... 13000 3780 15010 2780 14190 3100 14400 3460 13230 3110 13940 1-2 3-8 3-8 3-8 7-16 3-8 1-2 7-16 5-16 3-8 3-8 3-8 7-16 1-2 7-16 7-16 7-16 5-16 5-16 5-16 1-4 5-16 3-8 5-16 1-2 5-16 1-4 3-8 1-2 7-16 1-2 1-2 13230 --13-0- WEBBER-HOLDING POWER OF RAILROAD SPIKES TABLE XVII-Concluded 1 4 4 1 4 Resistance in Pounds for a Pull of 6090 5220 4570 5290 6830 3270 11560 10400 9890 10280 11280 8650 3700 7840 4570 9260 4130 7980 2960 6200 4760 7970 3950 7380 3450 6300 3300 6550 2640 5260 3130 6040 5200 6950 2750 6210 3240 6260 3730 6480 3070 5460 3960 3270 3940 5630 3660 5450 5260 7610 3840 6270 4830 7780 4640 7270 6180 10220 5350 8260 5820 9240 13 Kind of Tie Maximum Resistance o- 9920 11440 12400 11250 10080 9570 12480 10680 8000 8910 10130 9010 9340 8490 8060 8290 6400. 8250 6160 6940 5680 5310 5580 5520 5510 6210 7360 6390 7590 9060 8200 8280 12 11 11 40 5520 4880 Poplar Chestnut Loblolly Pine 4400 6450 7990 6260 5280 6350 7110 6250 4790 4820 7210 5610 5250 3860 2710 3940 3340 3800 4580 3910 3570 2820 2510 2960 2670 3630 3060 3120 4070 5060 14390 12890 12840 11610 12480 12310 10120 9610 10130 9960 9340 8490 8060 8290 7610 8250 7290 7720 7010 6470 6300 6590 9340 7550 8190 8690 11840 11190 9850 10630 1930 1140 1400 1490 1460 2120 2390 2320 1720 2460 3400 2530 . - 1-4 7-16 7-16 7-16 3-8 7-16 1-2 7-16 7-16 7-16 1-2 7-16 1-2 1-2 1-2 1-2 3-8 1-2 5-16 3-8 7-16 7-16 7-16 7-16 3-8 7-16 3-8 3-8 3-8 3-8 7-16 3-8 44 ILLINOIS ENGINEERING EXPERIMENT STATION A Relation between Depth of Penetration and the Holding Power A series of tests was made to determine the relation between the depth of penetration and the holding power of the screw spikes. The experiments consisted of pulling spikes driven to depths of 1, 2, 3, 4 and 5 inches into a beech tie, three spikes being used for each depth. The numerical results are shown in Table XVIII, and their averages are shown graphically in Plate VI together with some additional matter which is shown for the sake of comparison. TABLE XVIII RESULTS OBTAINED FROM EXPERIMENTS ON DEPTH OF PENETRATION Resistance in Pounds for a Penetration of Test 1 inch 2 inches 3 inches 4 inches 5 inches 1 2770 4560 9610 13100 17360 2 2760 6000 10000 14330 17500 3 2790 4940 8490 13330 16840 Av. 2770 5170 9360 13590 17230 The results in Plate VI can be quite closely represented by two intersecting straight lines. The probabilities are that the actual resistances would be more nearly represented if the two straight lines were joined by a short curve near their intersection. Only the upper portion of the diagram is of interest, since penetra- tions of less than four inches should never be used, at least on heavy traffic railroads, the only roads likely to use screw spikes. The diagram shows that the resistance varies directly with the depth of penetration. B Relative Holding Power of Screw Spikes and Ordinary Spikes Table XIX has been prepared from Table XVII and from Table III, to determine the relation between the holding power of the screw spike and that of the ordinary spike. As previously stated, the ordinary spikes were driven into the tie to a uniform depth of 5 inches, while the screw spikes, being of different lengths, necessarily were inserted to unequal depths. On account of the relation existing between the depth of penetration and the holding power, the resistance for the screw spikes, shown in Table XIX, is based upon a penetration of 5 inches. WEBBER-HOLDING POWER OF RAILROAD SPIKES PLATE VI LepfA of Penerio/on , /nch-es. / Z 3 4 Is /6000 /4000 /0000 6000 S0ooo S4000 2000 0 - -- - -~ i i/ - ----- I--=^z 7/ o/ _ _ _ - _ _ _ _ 5._ _ / -------_----__ /_ - /_ _ _ / ^_ ===x=^- =/=: - " 7- / ,/ '-zz zz - T z _ - _ _ _ _ - _ _ _ < L - _ _ _ _ C1'rves ///u3srab:/"7y Rest'.scrce /o W/drw'rawya/ o/ /Ae c-rew- a * Oz' nardy Sy'eo/ e 7S l,'/oV/s De/o/,/ of /Penyetrv/b/oi 46 ILLINOIS ENGINEERING EXPERIMENT STATION From Table XIX it will be seen that the holding power of the screw spike is always greater than that of the ordinary spike, and that the relation between the two varies in the several timbers. For a pull of 1-4 of an inch in the hard woods the holding power of the screw spike is from 167 to 221 per cent of that of the ordinary spike, and in the soft woods the range is from 117 to 258 per cent; or the average gain in the hard woods is 76 per cent, and in the soft woods 98 per cent. It is interesting to note that the resist- ances in the several timbers for the 1-8-inch pull with the screw spike are in eight out of eleven instances nearly the same as, or greater than, the resistances for the 1-4-inch pull with the ordinary spike. This signifies that the screw spike is about twice as effi- cient as the ordinary spike for a pull of 1-4 of an inch or less. The curve in Plates II and III show graphically the relative efficiency of the two forms of spikes with some information to be referred to later. C Effect of Certain Details of the Screw Spike upon Its Holding Power In countries where the screw spike is extensively used it has been perfected in detail until it nearly fulfills the requirements of practice. In North America the screw spike will probably be the successor to the ordinary spike, and it may again be necessary to adjust the details to suit local conditions. Therefore a few obser- vations on the relation of some of the details of this spike to its holding power come within the scope of this paper. The details to be discussed are the diameter of the core, the projection and pitch of the thread and the length of the thread. These details being interdependent will be discussed collectively. The soft steel from which the screw spike is made has an ulti- mate strength of about 66,000 pounds per square inch, so that the tensile strength of a spike 11-16 of an inch in diameter is approxi- mately 24,000 pounds. The ultimate compressive resistance across the grain of well-seasoned white oak is about 4,000 pounds per square inch, and experiments demonstrate that the thread of the spike in compacting the wood fibers increases the resistance about 40 per cent.* Therefore, taking 5,600 pounds as the ultimate com- pressive strength of compacted white oak, and taking 17 3-4 inches and 1-8 of an inch respectively as the length and projection of the *Bulletin No. 50, U. S. Dept. of Agriculture. WEBBER-HOLDING POWER OF RAILROAD SPIKES TABLE XIX RELATIVE HOLDING POWER OF THE SCREW SPIKE AND OF THE ORDINARY SPIKE IN SEVERAL TIMBERS Kind of Tie Water Oak Black Oak Red Oak White Oak Ash Beech Elm Poplar Chestnut Sweet Gum Loblolly Pine ance in Pounds for Resist Kind of Spike pike 1-8-in. Pull Ordinary 2870 Screw 4888 Ordinary 2910 Screw 4760 Ordinary 2950 Screw 4900 Ordinary 3510 Screw 6250 Ordinary 3570 Screw 5700 Ordinary 2600 Screw 6450 Ordinary 2380 Screw 5120 Ordinary 2830 Screw 3880 Ordinary 2850 Screw 3690 Ordinary 3230 Screw 5430 Ordinary 2920 Screw 5750 1-4-in. Pull 5730 9180 5890 10420 5350 10400 5950 11900 5200 10470 5490 13140 5580 10090 5290 6210 4070 6340 4120 7710 3500 9050 Relative Resistances 1-8-in. Pull 1-4-in. Max. Pull Resist. thread on the 5-inch spike, and making no allowance for frictional resistance between the core of the spike and the wood, the theo- retical resistance would be 5,600 x 17 3-4 inches x 1-8 inches==12,430 pounds. The average actual resistance obtained in white oak ties as shown in Table XIX is 12,630 pounds which agrees closely with the theoretical resistance. The tensile strength of the screw spike is Max. Resist. 6780 12190 7230 14110 7730 13560 7870 12630 7730 12760 8840 16230 7500 13690 5670 7490 5200 8700 5300 8280 4300 10620 48 ILLINOIS ENGINEERING EXPERIMENT STATION about 12,000 pounds greater than the maximum resistance of white oak, which difference is greater than necessary and indicates an uneconomical use of metal in the spike. Since the ties tested are representative of American practice, there is no apparent reason for not having the ultimate strength of the two materials in con- tact more nearly equal than at present, and by some slight change in the detail of the spike this could readily be accomplished. Three ways in which the ultimate strength of the materials may be made more nearly equal are: (1) increase in length of threaded portion; (2) increase in projection of thread, the length and the diameter of the core remaining the same; (3) increase in projec- tion of thread at the expense of the core, the length remaining the same. The pitch is assumed to be 1-2 inch in all cases, since it has been found in practice that this pitch gives better results than either a greater or smaller pitch. * (1) The length of the thread on the 5-inch spike is 17 3-4 inches and the width is 1-8 of an inch; therefore, the bearing area is 2.22 square inches. If the spike is made 6 inches long two convolu- tions of the thread will be added, the bearing area will become 2.71 square inches, and the holding power will be increased from 12,630 pounds to 15,180 pounds. This leaves a difference of only 8,900 pounds between the ultimate strength of the wood and that of the spike. (2) If the length of the spike and the diameter of the core are not changed, and if the projection of the thread is increased 1-32 of an inch, the total resistance would amount to 15,510 pounds, leaving the ultimate strength of the -spike only 8,500 pounds greater than that of the wood. (3) If the length of the threaded portion of the spike remains unchanged and if the projection of the thread is increased 1-32 of an inch at the expense of the core, the maximum resistance would amount to 15,510 pounds, while the ultimate strength of the spike would be reduced to 20,200 pounds. The diameter of the shank of the spike would have to be in- creased with some of the changes in the detail of the lower por- tion, and when the resistance to lateral displacement is taken into account, we see that this change also would be beneficial. The conclusion is that the screw spike in its present form is *Bulletin No. 50, U. S. Dept. of Agriculture. PLATE VII SCREW SPIKES AND TOOLS FOR INSERTING THEM WEBBER-HOLDING POWER OF RAILROAD SPIKES about twice as efficient as the ordinary spike; and that this effi- ciency could be increased by some slight change in the detail of the screw spike. ART. 3 HOLDING POWER OF SCREW SPIKES WITH HELICAL LININGS A few experiments were made with screw spikes having hel- ical linings. On account of the small number of linings obtainable the tests were limited; as this lining, being a foreign invention, is not yet used by the railroads of this country except for experi- mental purposes. The tests were still further limited since the linings could not be used a second time; and further since all of the linings could not be driven successfully, as the friction be- tween the metal and the wood sometimes caused the driver to loosen its hold, which could not be regained even after carefully following printed instructions. This accounts for the use of only two linings in some of the timber. The linings together with a set of special tools for inserting them in the tie were furnished by Mr. Robert Trimble, Chief Engineer Maintenance of Way, Pennsylvania Lines, (see Plate VII). The linings were made by Mr. J. Thiollier of Paris, France, and are described by him as being 0.33 inch by 0.17 inch in section, and also as being of the class which he calls P. M. or small sized linings. They were 4 inches long with a 1-2-inch pitch. The total diameter was 1 5-16 inches, the diameter inside of the spiral band slightly over 11-16 of an inch, and the thickness and width of the metal band 1-8 and 1-4 of an inch, respectively. The linings were evidently designed to be used with the screw spike of the French Eastern Railway, No. 1, Table XVI, and hence they were tested with this spike only. The method of fixing the lining in place was as follows: A hole having the same diameter as the core of the spike was bored in the tie; the hole was tapped, and the lining inserted by means of special tools designed for the purpose; the spike was inserted in the usual manner. The detailed results of these tests are shown in Table XX, and the average results are shown graphically in Plates II and III. The ielative holding power of the several kinds of spikes in dif- ferent timbers is shown in Table XXI. The results of this table and the diagrams in Plates II and III show that in hard woods 50 ILLINOIS ENGINEERING EXPERIMENT STATION the resistances for a 1-8-inch pull are usually greater for the spike and lining than for the naked screw spike, but for pulls greater than 1-8 of an inch the reverse is true. In soft woods the spike and lining gave greater resistances than the naked screw spike except in sweet gum. The lower resistance in the hard woods is account- ed for by the fact that the spike begins to move before the lining, and the fibers, being hard, are bent slightly upward so that the bearing surfaces of the wood and the spike are only partially in contact. Moreover, the fibers probably slip over the rounded edge of the lining, which tends to lower the resistance. In the soft woods more than in the hard woods, the fibers mash together as the spike is pulled out, consequently the bearing surfaces of the wood and the spike have full contact and the resistance is greater than with the naked screw spike. In justice to Mr. Thiollier it is only right to say that he claims no more for the P. M. lining than is set forth in these ex- periments. He says that the P. M. lining will offer no more resist- ance than a naked screw spike. The principal claims for the P. M. lining are that it can be placed on the track without re- moving either the rail or the tie, and that it forms an advantag- eous substitute for the square wooden dowel used on some rail- ways. As a repair measure this lining is of doubtful value, for it ex- tends only about 1-8 of an inch beyond the thread of the spike; and when the spike has been pulled even a small distance the adjacent wood is badly damaged, so that the wood which remains after the hole is tapped for the lining can offer but slight resistance. More- over, it is not certain that the extreme fibers reached by the lin- ing are not somewhat affected, hence it would be better to ream the hole, cutting out all damaged wood and to introduce a thread- ed hard wood dowel, or to use a lining of larger size. The writer claims that the use of the small lining is imprac- ticable for the following reasons: (1) It is designed to be put in place with the tie in the track; (2) The lining cannot always be inserted into the wood to its full length by means of hand tools, even with utmost precaution; (3) At best the holding power is not increased to any marked degree over that of the naked screw spike; and (4) The labor involved is more than double that re- quired to drive the naked screw spike, and the cost is increased. WEBBER-HOLDING POWER OF RAILROAD SPIKES TABLE XX RESISTANCE OF SCREW SPIKES WITH HELICAL LININGS Resistance in Pounds for Maximum SPull of01 esistance 1-8 in. 1-4 in. 1-2 in. 3-4 in. 1 in. Pounds Pull, 1 1 8410 11380 10150 2 5830 8670 9410 3 5670 8070 7930 Av. 6640 9370 9160 3 1 6010 9100 7750 2 4830 6440 7650 3 4270 6250 8600 Av. 5030 7260 8000 26 1 3420 7100 11080 2 2970 6460 12080 Av. 3190 6780 11580 32 1 5810 10740 8420 2 7070 11020 6650 Av. 6440 10880 7530 23 1 5960 11130 9810 2 5420 9710 10770 Av. 5690 10420 10290 1 10830 10120 8070 2 8610 I 11600 11850 Av. 9720 10860 9960 11 1 3970 8860 9900 2 4080 9470 10550 3 3670 8260 9910 Av. 3910 8860 10120 1 7020 9600 8230 2 5750 7010 8890 3 6300 7240 9280 Av. 6390 7950 8810 7570 6590 4690 6280 5380 6270 6130 5930 8290 9250 8780 6890 6170 6530 8560 8470 8510 7320 10350 8830 5880 5940 6030 5950 6920 8180 7660 7590 6480 12160 5630 10500 4200 8750 5440 10470 5150 9510 4380 7970 4410 8600 4650 8690 8740 11080 9170 12080 8960 11580 7120 12900 6340 11020 6750 11960 7520 12550 7960 12460 7740 12500 5390 10830 6280 13480 5830 12150 5300 9920 5110 11140 5250 9910 3220 10320 6120 9770 6730 8890 6860 9280 6900 9150 1-4 3-8 3-8 3-8 1-4 3-8 1-2 3-8 1-2 1-2 1-2 3-8 14 3-8 3-8 3-8 3-8 1-8 3-8 1-4 3-8 3-8 1-2 3-8 3-8 1-2 1-2 1-2 Sweet Gum Water Oak White Oak Black Oak Beech Poplar Chestnut 52 ILLINOIS ENGINEERING EXPERIMENT STATION TABLE XXI RELATIVE HOLDING POWER OF THE ORDINARY SPIKE, THE SCREW SPIKE, AND THE SCREW SPIKE WITH HELICAL LINING IN SEVERAL TIMBERS Resista Tie Spike 1-8-in. Pull White Oak Ordinary 3510 Screw* 6250 Lining 6440 Water Oak Ordinary 2870 Screw* 4880 Lining 3190 Black Oak Ordinary 2910 Screw* 4760 Lining 5690 Ash Ordinary 3570 Screw* 5700 Lining 6640 Beech Ordinary 2600 Screw* 6450 Lining 9720 Poplar Ordinary 2830 Screw* 3850 Lining 3910 Chestnut Ordinary 2850 Screw* 3690 Lining 6390 Sweet Gum Ordinary 3230 Screw* 5430 Lining 5030 * Screw spike with helical lining. The * belongs after "Lining." ice in Pounds for Relative Resistance 1-4-in. Max. Pull Resist. 5950 7870 11900 12630 10880 11960 5730 6780 9180 12190 6780 11580 5890 7230 10420 14110 10420 12500 5200 7730 10470 12760 9370 10470 5490 8840 13140 16230 10860 12150 5290 5670 6210 7490 8860 10320 4070 5200 6340 8700 7950 9150 4120 5300 7710 8280 7260 8690 1-8-in. 1-4-in. Max. Pull Pull Resist. Resistar PLATE VIII Kqx" IMPACT APPARATUS ! !l * * i l f WEBBER-HOLDING POWER OF RAILROAD SPIKES PART II RESISTANCE TO LATERAL DISPLACEMENT The railroad spike is subjected not only to a direct pull by the undulation of the rail, but also to a horizontal thrust due to the lateral movement of the rail. On roads having a large amount of curvature the lateral resistance is of more importance than that of direct pull. 6 To determine the amount of the resistance to lateral displace- ment which is developed by various forms of spikes the writer made a series of tests in which the lateral thrust was produced by the blows of a heavy hammer. The hammer consisted of a cast- iron weight suspended by a wooden rod from the joists of the floor above. The place in which the apparatus was used was such that a good photograph could not be taken. Plate VIII is a view of the apparatus set up in a light suitable for photographing. All essential features are correctly represented. Fastened to the joists were metal strips upon which the knife edges of the rock- ing arm rested. These strips were 6 feet long, and were notched along the entire upper edge to permit the placing of the rocking arm in different positions. The length of the suspending rod was 9 feet. The weight of the hammer was 100 lb. and the distance through which it was allowed to fall was 1 1-2 feet, so that the amount of the impact for each blow was 150 ft.-lb. The hammer delivered its blow on the end of a tool-steel bar which projected beyond the end of the tie, the other end of the bar being shaped to fit under the head of the spike. The spikes used in this series of tests were 9-16 inch and 5-8 inch ordinary spikes and screw spikes. Each spike was subjected to five blows and the displacement produced by each blow was care- fully measured. Usually four or five spikes of each kind were tested, but when there was much lack of uniformity in the results a larger number were tested. All of the spikes were bent to a curve, the central point of which was about 1 1-2 inches below the surface of the tie. The ordi- nary spikes were pulled from the tie a short distance, but the thread of the screw spikes gripped the wood so as to prevent the spike from being pulled out even a perceptible amount. 54 ILLINOIS ENGINEERING EXPERIMENT STATION ART. 3 LATERAL RESISTANCE OF ORDINARY SPIKES The detailed results of the experiments with ordinary spikes are given in Table XXII and the average movement of the spike for each of the several blows is shown in Table XXIII. The aver- age total movement of the 5-8 inch spikes in the first seven tim- bers was 0.65 inch, and that of the 9-16 inch spikes was 0.75 inch. In the last four timbers the average total movement of the 5-8 inch spikes was 0.74 inch, and that of the 9-16 inch spikes was 0.94 inch. The total deflection of the 9-16 inch spikes was usually suffi- cient to allow a rail to clear the head of the spike if it were over- turned. The corresponding movement of the 5-8 inch spikes was not usually sufficient to allow a like clearance, although it was considerably more than would be allowed in practice. The first blow is of more importance than the succeeding blows in testing the efficiency of a spike. While the distances through which the different sized spikes were deflected by the first blow differ but a small amount, this difference is sufficient to show that the deflection is less for the 5-8 inch spikes than for the 9-16 inch. These results, together with the fact that the 5-8 inch spikes were bent less by the impact than the 9-16 inch spikes, indicate that the 5-8 inch spike is more efficient in resisting lateral displace- ment than the 9-16 inch spike. ART. 4 LATERAL RESISTANCE OF SCREW SPIKES The method of determining the lateral resistance of screw spikes was the same as that used for ordinary spikes. The results for this set of tests are given in Table XXIV. The screw spikes used were all practically alike except that they were of various lengths. In making the tests the spikes were used indiscriminate- ly, but since they were not all of the same length some tests were made to determine the effect of impact upon spikes which were driven into the tie to different depths. The spikes used for the latter tests were all of the same make, and were cut to lengths of 3, 3 1-2, 4, 4 1-2 and 5 inches, and were all driven into a single kind of timber. The results of these tests are shown in Table XXV. While the results for the 4- and 4 1-2-inch spikes are the same, the WEBBER-HOLDING POWER OF RAILROAD SPIKES 55 TABLE XXII DETAILED RESULTS OF IMPACT TESTS OF ORDINARY SPIKES Total Lateral Movement of Spikes in Inches Size of Spike, Number of Blows in sq _______________ 1 2 3 4 White Oak Water Oak Black Oak 0.35 0.48 0.65 .35 .56 .67 .22 .33 .45 .35 .50 .52 .35 .60 .74 5 0.81 .73 .54 .60 .93 9-16 0.27 .18 .10 .30 .21 1. 0.21 5-8 0.11 .15 .19 .21 .20 r. 0.17 9-16 0.23 .20 .14 .20 .19 v. 0.19 5-8 0.12 .20 .15 .19 .20 T. 0.17 9-16 0.25 .13 .16 .24 .23 .26 v. 0.21 5-8 0.23 .17 .17 .15 .11 .22 T. 0.17 Kind of Tie 0.32 0.49 0.61 0.70 0.20 0.26 0.30 0.39 .30 .41 .50 .57 .36 .50 .60 .68 .36 .49 .65 .74 .34 .42 .50 .57 0.31 0.42 0.51 0.59 0.34 0.52 0.60 0.75 .33 .56 .73 .88 .42 .53 .68 .75 .35 .48 .54 .65 .39 .63 .72 .78 0.37 0.54 0.65 0.76 0.25 0.36 0.48 0.55 .37 .54 .63 .69 .25 .31 .39 .50 .28 .43 51 .65 .37 .53 .65 .69 .30 0.43 0.53 0.61 0.40 0.56 0.70 0.75 .30 .41 .58 .72 .32 .49 .58 .70 .44 .62 .71 .80 .35 .56 .65 .69 .39 .59 .67 .78 0.37 0.54 0.65 0.71 0.38 0.50 0.58 0.65 .30 .42 .53 .64 .35 .50 .61 .77 .32 .40 .49 .55 .26 .37 .41 .45 .35 .50 .59 .65 0.33 0.45 0.53 0.62 A - Black Oak Av Av Av Av Av 56 ILLINOIS ENGINEERING EXPERIMENT STATION TABLE XXII-Continued Total Lateral Movement of Spikes in Inches Number of Blows 1 2 3 4 5 1.21 0.35 0.51 0.61 0.73 .19 .30 .46 .57 .75 .20 .37 .55 .64 .77 .22 .41 .49 .61 .72 1.21 0.36 0.50 0.61 0.74 1.12 0.21 0.32 0.42 0.49 .15 .24 .34 .43 .50 .12 .25 .35 .49 .53 .18 .42 .55 .72 .85 Av. 0.14 0.28-- 0.39 0.52 0.60 Ash 9-16 0.24 0.45 .057 0.68 0.80 .24 .43 .53 .65 -74 .20 .33 .52 .65 .75 .25 .41 .60 .72 .83 Av. 0.23 0.41 0.56 0.68 0.78 5-8 0.19 0.37 0.55 0.73 0.84 .9 I .3a .48 .6 .Ulf .18 .31 .44 .60 .69 .15 .30 .39 .54 .63 Av. 0.18 0.33 0.47 0.63 0.73 9-16 0.22 0.33 0.50 0.67 0.78 .21 .30 .39 .56 .70 .25 .37 .49 .58 .66 .18 .30 .43 .54 .67 Av. 0.22 0.33 0.45 0.59 0.70 5-8 0.20 0.38 0.50 0.61 0.71 .21 .35 .48 .60 .72 .20 .35 .49 .61 .70 .21 .32 .44 .55 .66 Av. 0.21 0.35 0.48 0.59 0.70 9-16 0.28 0.30 0.58 0.72 0.87 .26 .46 .57 .75 .86 .21 .32 .53 .65 .75 .30 .54 .63 .71 .89 .19 .37 .55 .70 .80 .27 .46 .61 .72 .86 0.25 0.41 0.58 0.71 0.84 Elm Beech WEBBER-HOLDING POWER OF RAILROAD SPIKES 57 TABLE XXII-Continued Total Lateral Movement of Spikes in Inches . . . .. , S i z p n f Spike, Number of Blows in. sq. 1 2 3 Kind of Tie Poplar Chestnut Av. 5-8 Av. Sweet Gum 9-16 Av. .35 3 r .31 .29 .30 0.32 0.17 .10 .27 .25 .24 .28 0.22 5-8 0.15 .13 .16 .12 .12 .14 Av. 0.14 9-16 0.27 .22 .30 .27 .27 Av. 0.27 5-8 0.10 .16 .20 .17 Av. 0.16 9-16 0.35 0.29 0.51 .23 .40 .30 .51 .31 .54 0.28 0.49 0.23. .20 .27 .26 .30 .25 0.25 0.41 .40 .45 .41 .40 0.41 0.29 .28 .39 .39 0.34 0.65 .60 .60 .62 .52 .50 0.58 0.40 .30 .45 .48 .40 .42 0.41 0.33 .29 .36 .43 .37 .31 0.35 0.59 .54 .60 .54 .52 0.56 0.41 .41 .50 .39 0.43 0.90 .80 .90 .91 .75 .73 0.83 0.60 .67 .63 .70 .57 .53 0.61 0.60 .66 .67 .72 0.66 4 0.46 .41 .49 .50 .46 .39 0.45 0.75 .67 .68 .75 .61 0.69 0.50 .51 .66 .46 0.53 1.06 .97 1.12 1.01 .93 .93 1.00 0.78 .88 .80 .91 .75 .65 0.79 0.78 .75 .75 .97 0.81 0.96 5 0.53 .49 .58 .57 .62 .50 0.55 0.88 .74 .84 .76 0.81 0.63 .60 .75 .57 0.64 1.40 1.10 1.35 1.19 1.18 1.19 1.23 0.85 1.05 .92 1.03 .90 .84 .93 0.95 .88 .92 1.10 58 ILLINOIS ENGINEERING EXPERIMENT STATION TABLE XXII- Concluded I Movemen Lt of Spikes in Inches Number of Blows 3 4 5 Kind of Tie Sweet Gum Size of Spike, in. sq. 5-8 0.14 .18 .16 .14 0.16 0.22 .23 .12 .24 .26 .23 0.22 0.16 .17 .17 .15 .23 .12 .23 0.18 averages in the last column of the table show that the amount of the lateral movement decreases as the depth of penetration in- creases. Also, the difference between the deflections of the 4-, 41-2-, and 5-inch spikes is practically negligible, but for shorter lengths the difference in the deflections becomes greater. Table XXVI gives the lateral movement of the screw spikes for each of the several blows for which the total movements were given in Table XXIV. The number of spikes used in each kind of timber was usually three; but in case there was consider- able variation in the results, more spikes were tested. By a study of this table the effect of impact upon screw spikes in different kinds of timber may be determined. Total Latera 0.45 0.62 0.78 .54 .62 .75 .46 .62 .70 .42 .50 .61 0.47 0.59 0.17 0.50 0.61 0.70 .65 .76 .81 .35 .42 .50 -58 .71 .88 .53 .70 .75 .64 .72 .77 0.54 0.65 0.74 0.40 0.50 0.65 .63 .72 .85 .30 .51 .55 .40 .52 .59 .46 .61 .71 .29 .36 .41 .53 .68 .78 Loblolly Pine 0.56 0.65 Av. 9-16 AV. 5-8 Av. 1 0.43 2 0.28 .35 .33 .38 0.34 0.33 .38 .23 .37 .42 .45 0.36 0.30 .42 .22 .223 .38 .19 .39 0.30 WEBBER-HOLDING POWER OF RAILROAD SPIKES TABLE XXIII LATERAL MOVEMENT OF ORDINARY SPIKES FOR E AOH BLOW 0M Kind ofTie 0, 0). White Oak 9-16 5-8 Water Oak 9-16 5-8 Black Oak 9-16 5-8 Red Oak 9-16 5-8 Ash 9-16 5-8 Elm 9-16 5-8 Beech 9-16 5-8 Poplar 9-16 5-8 Chestnut 9-16 5-8 Sweet Gum 9-16 5-8 Loblolly Pine 9-16 5-8 Movement for Each of the Sev Blows, inches 1 0.21 0.17 0.19 0.17 0.21 0.17 0.21 0.14 0.23 0.18 0.22 0.21 0.25 0.14 0.27 0.16 0.32 0.22 0.28 0.16 0.22 0.18 2 3 4 0.12 0.09 0.11 0.10 0.11 0.08 0.11 0.14 0.12 0 16 0.13 0.11 0.13 0.10 0.14 0.10 0.17 0.18 0.15 0.12 0.11 0.13 eral aoa 5 0.09 0.136 0.08 0.118 0.11 0.152 0.08 0.122 0.06 0.142 0.09 0.124 0.13 0.148 0.08 0.122 0.10 0.156 0.10 0.146 0.11 0.138 0.11 0.140 0.13 0.168 0.10 0.110 0.12 0.164 0.11 0.128 0.23 0.246 0.14 0.186 0.15 0.192 0.12 0.142 0.04 0.148 0.09 0.128 60 ILLINOIS ENGINEERING EXPERIMENT STATION TABLE XXIV DETAILED RESULTS OF IMPACT TESTS OF SCREW SPIKES Total Lateral Movement of S ike in Inches Kind of Tie White Oak Black Oak Water Oak Red Oak Ash Elm Beech Av. Av. Av. Av. Av. Av. Number of Blows 1 2 0.09 .10 .07 0.09 0.11 .10 .11 0.11 0.09 .11 .08 0.09 0.12 .11 .17 0.13 0.17 .18 .12 0.16 0.11 .12 .21 .25 0.17 0.10 .11 .12 .16 .17 .20 0.16 .20 .14 0.17 0.21 .19 .18 0.19 0.13 .17 .18 0.16 0.21 .20 .23 0.21 0.23 .27 .25 0.25 0.30 .22 .40 .40 0.33 0.18 .18 .19 .28 .31 .40 3 0.23 .24 .21 0.23 0.26 .25 .24 0.25 0.22 .23 .26 0.24 0.35 .34 .33 0.34 0.34 .35 .33 0.34 0.38 .37 .58 .52 0.46 0.23 .26 .25 .38 .52 .52 4 0.30 .32 .28 0.30 0.36 .33 .31 0.33 0.33 .34 .35 0.34 0.45 .44 .46 0.45 0.47 .46 .45 0.46 0.48 .49 .85 .63 0.61 0.28 .31 .32 .49 .58 .60 0.26 5 0.38 .41 .40 0.40 0.40 .44 .42 0.42 0.42 .45 .41 0.43 0.54 .52 .52 0.53 0.54 .55 .53 0.54 0.56 .53 .96" .75 0.70 0.36 .37 .42 .58 .65 .68 1. 14 0.14 0.26 0.36 0.43 WEBBER-HOLDING POWER OF RAILROAD SPIKES TABLE XXIV-Concluded Total Lateral Movement of Spike, in Inches Kind of Number of Blows Tie123 1 2 3 4 5 0.09 0.16 0.32 0.60 0.78 .10 .16 .27 .40 .61 .09 .15 .34 .39 .49 .19 .35 .44 .61 .78 .18 .40 .53 .62 .75 .17 .27 .40 .63 .71 .16 .30 .39 .51 .62 Av. 0.17 0.24 0.38 0.54 0.67 Chestnut 0.16 0.23 0. 0 .13 .22 .37 .52 .56 .12 .24 .33 .42 .51 .20 .31 .39 .51 .59 .19 .28 .39 .48 .65 Av. 0.16 0.26 0.37 0.47 0.56 0.20 0.38 0.52 0.68 0.78 .26 .46 .60 .71 .79 .30 .48 .51 .74 .8g .18 .32 .40 .49 .61 .25 .38 .47 .59 .68 Av. 0.24 0.40 0.50 0.64 0.74 Table XXVII is given to facilitate the comparison of the rel- ative lateral resistance of ordinary and screw spikes. The data were collected from Tables XXIII and XXVI. The average total deflection of the screw spike in the first seven timbers is 0.50 inch which is 0.15 inch less than that of the 5-8-inch ordinary spike and 0.25 inch less than that of the 9-16-inch ordinary spike. In the Poplar Sweet Gum T - -1^ -11. - - ^ 62 ILLINOIS ENGINEERING EXPERIMENT STATION TABLE XXV RELATION BETWEEN THE DEPTH OF PENETRATION AND THE RESISTANCE TO LATERAL DISPLACEMENT Deflection in Inches Depth of Insertion 3 in. 3 1-2 ii 4 in. 1-2 ii Number of Blows 0.24 .22 .24 Av. 0.23 n. 0.24 .24 .19 Av. 0.22 .20 .21 .23 Av. 0.21 n. 0.24 .20 .22 0.22 0.22 .23 .15 0.20 2 3 4 0.46 .41 .43 0.43 0.46 .39 .34 0.40 0.39 .40 .33 0.37 0.30 .34 .36 0.33 0.38 .40 .34 0.34 0.64 .55 .67 0.62 0.62 .53 .49 0.55 0.49 .57 .57 0.54 0.50 .53 .54 0.52 0.49 .55 .48 0.51 0.78 .69 .76 0.73 0.77 .69 .63 0.70 0.60 .63 .62 0.62 0.65 .68 .62 0.65 0.61 .67 .57 0.87 .84 .98 0.90 0.80 .80 .74 0.78 0.71 .77 .72 0.73 0.74 .73 .79 0.75 W b a 8fl Cl 0 0.582 0.530 0.494 0.494 0.62 0.72 0.478 last four kinds of timber the average total deflection of the screw spike was 0.70 inch, which is practically the same as that of the 5-8-inch ordinary spike, but which is 0.24 inch less than that of 9-16-inch common spike. The results in the last two columns of Table XXVII show that the screw spike is superior to the 9-16-inch ordinary spike in all but two kinds of timber, and that the screw spike has a higher efficiency than the 5-8-inch ordinary spike in all but three kinds of timber. WEBBER-HOLDING POWER OF RAILROAD SPIKES TABLE XXVI LATERAL MOVEMENT OF THE SCREW SPIKE FOR EACH BLOW Movement for Each of the a Kind Several Blows Z')- of Tie W 0 t) 1 2 3 4 5 White Oak 0.09 0.08 0.05 0.07 0.10 0.078 Black Oak 0.11 0.08 0.06 0.07 0.09 0.082 Water Oak 0.09 0.07 0.08 0.10 0.09 0.086 Red Oak 0.13 0.08 0.13 0.12 0.08 0.108 Ash 0.16 0.09 0.09 0-12 0.08 0.108 Elm 0.17 0.16 Beech .0.14 0.12 Poplar 0.17 0.07 Chestnut 0.16 0.10 Sweet Gum 0.24 0.16 0.13 0.10 0.12 0.11 0.10 Loblolly Pine 0.21 0.13 0.19 0.15 0.09 0.140 0.07 0.08 0.102 0.16 0.13 0.130 0.10 0.09 0.132 0.14 0.10 0.148 0.12 0.14 0.154 The last two columns in Table XXVII show that the ordinary spike was usually displaced more than the screw spike by each blow. This should be expected since the common spike was smaller in cross section than the screw spike, and also since the latter had better bond with the wood. While the use of the screw spike is recommended to the American railroads, it is thought that the practice of Bavarian railroads could be followed to ad- vantage. These roads have adopted the use of the screw spike on the gage side of the rail to resist overturning, but use two square spikes on the outside to resist lateral movement. This practice has been found to give very beneficial results. The figures in the last two columns of Table XXVII show that the lateral resist- ance of two ordinary spikes is considerably more than that of one screw spike, and therefore if two spikes are considered as resist- ing the impact instead of one, the results will be in favor of the ordinary spikes. Not only is this true, but the first cost for spikes would be reduced, since the screw spike costs about four cents at 64 ILLINOIS ENGINEERING EXPERIMENT STATION TABLE XXVII RELATIVE LATERAL DISPLACEMENT OF ORDINARY AND SCREW SPIKES ent of Ordi- y Spikes 5-8 in. 0.118 0.122 0.124 0.122 0.146 0.140 0.110 0.128 0.186 0.142 0.128 Average Movement of Screw Spike, inches 0.078 0.082 0.086 0.108 0.108 0.140 0.102 0.130 0.132 0.148 0.154 Average Movement of Ordinary Spikes in Terms of per cent of Move- ment of Screw Spike 9"-16 in. 5-8 in. 186 141 129 96 the present time, whereas the ordinary spike costs much less. The maintenance cost of either form of spike is almost negligible. An item of interest -which is properly beyond the limits of this article is that of the ninety screw spikes used in making these tests only two were broken. One was broken under a tension of 14,000 pounds, the break being caused by an incipient crack just under the head of the spike. The other spike broke under the fourth blow of the hammer, this break being due to uncombined graphite in the metal. As the spikes were obtained from differ- ent sources, and were of different manufacture, it is thought that the test was sufficiently severe to show that the screw spike, as manufactured at present, will successfully withstand the shocks of passing trains. As the spikes were used several times during the tests, the percentage of spikes broken is very low. Movem nar 9-16 in. 0.136 0.152 0.142 0.148 0.156 0.138 0.168 0.164 0.246 0.192 0.148 Kind of Tie White Oak Black Oak Water Oak Red Oak Ash Elm Beech Poplar Chestnut Sweet Gum Loblolly Pine WEBBER-HOLDING POWER OF RAILROAD SPIKES SUMMARY OF RESULTS (1) The maximum resistance to direct pull varies from 6,000 to 14,000 pounds for screw spikes, from 3,000 to 8,000 pounds for ordinary spikes when driven into untreated timbers, and from 4,000 to 9,000 pounds for ordinary spikes when driven into treat- ed timbers. (2) The direct pull required to withdraw ordinary spikes 1-8- inch varies from 2,000 to 3,500 pounds for untreated timbers, and from 2,500 to 3,500 pounds for treated timbers. (3) The direct pull required to withdraw ordinary spikes 1-4- inch varies from 3,000 to 5,400 pounds for untreated timbers and from 3,800 to 5,900 pounds for treated timbers. (4) Timbers having loose fiber structures have lower resis- tances to direct pull than timbers having compact fiber structures. (5) The amount of withdrawal which must occur for ordinary spikes to develop the maximum resistance is less for soft woods than for hard woods. (6) Spikes driven into treated timber offer a greater resis- tance to direct pull than spikes in untreated timbers, and the difference between this resistance for treated and untreated tim- bers is greater for soft woods than for hard woods. (7) The difference in the resistance to direct pull for the different sized spikes in use (9-16 inch, 19-32 inch, and 5-8-inch) is very small. (8) The resistance of ordinary spikes to direct pull varies directly as the depth of penetration, neglecting the tapering point. (9) Blunt-pointed and bevel-pointed spikes have a slightly greater resistance to direct pull than chisel-pointed spikes. (10) For withdrawals less than 1-4 inch, ordinary spikes which are driven into bored holes have a little greater resistance to direct pull than spikes driven in the ordinary way. (11) The resistance to direct pull for re-driven spikes is from 60 to 80 per cent of the resistance of newly driven spikes. (12) The efficiency of screw spikes to resist withdrawal is nearly twice as great as that of common spikes. (13) The resistance of 5-8-inch spikes to lateral displacement is slightly greater than that of 9-16-inch spikes. (14) The resistance to lateral displacement increases with 66 ILLINOIS ENGINEERING EXPERIMENT STATION the depth of penetration, but the increase is negligible for depths of penetration greater than 4 inches. (15) Screw spikes are more efficient than ordinary spikes in resisting lateral displacement. PUBLICATIONS OF THE ENGINEERING EXPERIMENT STATION Bulletin No. 1. Tests of Reinforced Concrete Beams, by A. N. Talbot. 1904. Circular No. 1. High-Speed Tool Steels, by L. P. Brecken- ridge. 1905. Bulletin No. 2. Tests of High-Speed Tool Steels on Cast Iron, by L. P. Breckenridge and Henry B. Dirks. 1905. Circular No. 2. Drainage of Earth Roads, by Ira 0. Baker. 1906. Bulletin No. 3. The Engineering Experiment Station of the University of Illinois, by L. P. Breckenridge. 1906. Bulletin No. 4. Tests of Reinforced Concrete Beams, series of 1905, by A. N. Talbot. 1906. Bulletin No. 5. Resistance of Tubes to Collapse, by A. P. Carman. 1906. Bulletin No. 6. Holding Power of Railroad Spikes, by R. I. Webber. 1906.