opinions, findings, and conclusions expressed in this

Celik Ozyildirim H. Scientist Research opinions, findings, and conclusions expressed in this (The are those of author and not necessarily those of report Highway & Transportation Research Council Virginia Cooperative Organization Sponsored Jointly by Virginia (A of Highways $ Transportation and Department University of Virginia) 1982 June 82-R58 VHTRC FINAL REPORT USE OF NEOPRENE PADS IN TESTING CONCRETE CYLINDERS by sponsoring agencies.) Charlottesville, Virginia

ESTES, Chairman of Civil Engineering Technology, 01d Dominion E. University CONCRETE RESEARCH ADVISORY COMMITTEE J. E. GALLOWAY, JR., Chairman, Asst. Materials Engineer, VDH&T T. R. BLACKBURN, District Materials Engineer, VDH&T C. L. CHAMBERS, Division Structural Engineer, FHWA W. R. DAVIDSON, Asst. Maintenance Engineer, VDH&T J. G. HALL, Materials Engineer, VDH&T F. C. MCCORMICK, Department of Civil Engineering, U. Va. W. R. MUSTAiN, Assistant District Engineer, VDH&T A. D. NEWMAN, District Materials Engineer, VDH&T H. C. 0ZYILDiRIM, Highway "Research Scientist, VH&TRC W. T. RA EY, District Bridge Engineer, VDH&T J. F0 J. VOLGYI, JR., Bridge Design Engineer, VDH&T W. E. WINFREY, Assistant Construction Engineer, VDH&T ii

study investigated feasibility of using neoprene This confined by steel end caps instead of sulfur-mortar caps pads compressive strength tests on concrete cylinders. The 1/2 in. in mm) thick neoprene pads had a 50 durometer hardness and were (13 to fit in end caps, which had an inside diameter of 6-1/4 in. cut mm) with a tolerance of +0 and -1/16 in. (-2 mm). Compressive (159 linear regression analysis indicated a good correlation between A two methods. SYNOPSIS data were obtained from 438 pairs of cylinders prepared strength field from commercial batches of concrete. One cylinder of in pair was tested with neoprene pads and or with sulfur- each caps. Although results indicate statistically significant mortar in values obtained by two capping methods, differences are considered negligible from a practical standpoint. differences iii

Celik Ozyildirim H. Scientist Research a previous study utilizing laboratory specimens, it was In that compressive strengths of 6 x 12 in. (150 x 300 mm) found tested with neoprene pads confined by steel end caps cylinders both ends were in close agreement with strengths obtained on on tested with standard sulfur-mortar caps. (I) The cylinders in. (13 mm) thick neoprene pads had a 50 durometer hardness, 1/2 were placed in 6-1/4 in. (159 mm) inside diameter steel and caps. In general, compressive strengths of specimens end with neoprene pads were slightly lower than those of tested capped with sulfur-mortar. However, based on specimens limited a conducted by New York Department of Transportation Work DOT)also has shown that neoprene pads are an acceptable sub- (NY. for sulfur-mortar caps$2) In NYDOT tests, neoprene stitute were of same thickness and type as described above but.y pads NY DOT report mistakenly stated as being 6-1/2 in. [165 mm]), tial of 6-1/4 in. (159 mm) diameter. Specimens tested instead NY DOT study gave strength values slightly higher than those in at same time with sulfur-mortar caps. At present, tested study by a consulting firm showed that uncapped specimens A using steel end caps with rubber inserts yielded compressive tested values that, for all practical purposes, were about strength as those obtained on specimens capped with sulfur-mortar.(3) same used steel end caps with a diameter of 6-1/2 in. (165 mm) and They 1/2 in. (13 mm) thick rubber inserts, which had a 50 durometer hardness, were snugly fitted into caps. FINAL REPORT USE OF NEOPRENE PADS IN TESTING CONCRETE CYLINDERS by INTRODUCTION of specimens differences between two methods were number significant at 95% confidence level. not were used in steel end caps having a 6-3/16 in. (157 mm) (in ini- pads in steel end caps are widely used by NY DOT to neoprene compressive strength of 6 x 12 in. (150 determine 300 mm) x cylinders.

use of neoprene pads instead of sulfur-mortar caps is The advantageous in that it reduces costs of sample significantly and eliminates hazards from handling hot toxic preparation and air pollution from sulfurous fumes. Consequently, materials study reported here was conducted to compare results of on companion field specimens by alternative methods. The tests in field specimens usually is not as well controlled and concrete top surface is not as uniform as that foundin specimens pre- objective of this study was to investigate feasibility The determining compressive strength of concrete by testing un- of field specimens using neoprene pads confined in steel caps capped an alternate to testing specimens capped with sulfur-mortar. A as of 438 pairs of 6 x 12 in. (150 x 300 mm) cylinders were tested total each procedure, and average and standard deviation of by Richmond, Lynchburg, Salem, and Staunton districts partic- The in this study. Each was asked to provide I00 pairs of test ipated which would give approximately required number of specimens, based on an estimated variability of 300 psi (2.07 MPa) of pairs difference is smallest value considered significant, from an standpoint, for this study. Also, previous labora- engineering experience had indicated that pads could be used at least I00 tory without any significant deterioration of surface texture times hardness. This observation was confirmed in present study; or fact, Richmond District furnished 138 pairs of cylinders in with same pads without any significant apparent damage tested of districts (Lynchburg and Salem) tested cylinders in Two laboratories. The test apparatus in Richmond and Staunton ir labs did not have enough clearance to accommodate height district specimen with steel caps. Therefore, Richmond furnished of pared in laboratory. OBJECTIVE AND SCOPE differences in test results were noted. PROCEDURE Participating Labs differences at 0.01 significance level with a 0.01 prob- paired of detecting a difference of 150 psi (1.03 MPa). (4) This ability to surface touching cylinder ends.

specimens to Department's nearby Central Materials Labora- for testing, and Staunton District sent cylinders to tory Research Council. The remaining four districts did not partic- in study because of limited clearance in ir testing ipate However, clearances in se machines can be adjusted, machines. neoprene pads are adopted for use. if 1980 construction season. From each batch, a pair of cylinders cast in steel molds and cured in a moist room until tested at were sulfur-mortar used was a commercially available material The ASTM requirements. The steel end caps were manufactured by meeting single company and a separate set was furnished each participant. a specified diameter of end caps, as shown in Figure I, was The in. (159 mm) with a tolerance of +0 and -1/16 in. (-2 mm). 6-1/4 base plate of steel end cap was hot-rolled steel cut in a The shape. However, it can be cut in a circular shape for more square centering against machine heads. The neoprene pads convenient all same type and thickness. They were cut to fit were ring and were equal to or slightly larger than 6-1/8 in. inside mm) in diameter. After a few tests, neoprene pad flows (156 to snugly fit inside of ring. sufficiently during loading, but also large enough to permit setting ends conveniently into ring. cylinders addition to compressive strength, each participant In information on type of break. Failure types ranged furnished a shear or cone type to splitting or columnar type, as dis- from later. cussed and Cappin$ Materials Cylinders were cast from batches of concrete furnished during Cylinders days one with a sulfur-mortar cap and or with neoprene 14 All concretes tested met th 5 equirements of Depart- ment;s pads and Bridge Specifications, but had variable strength Road levels. specified inside diameter of ring was selected so as The keep it small enough to prevent flow of neoprene around to Data Obtained

? 1/4" Continuous Weld- 5/16' i. Sketch of a steel end cap. Figure in. = 25.4 mm) (I

14-day compressive strength values for tests with sulfur- The caps ranged from 2,700 psi (18.6 MPa) to 6,930 psi (47.8 MPa), mortar for neoprene pads from 2,360 psi (16.3 MPa) to 6,960 psi and MPa). The average compressive strength test data with both (48.0 caps and neoprene pads and associated standard sulfur-mortar for each district as well as combined data deviations given are Table i. The results indicate that average strength values in standard deviations for two test methods were very and largest difference in average strength values between The methods was found to be 108 psi (745 kpa). For all data, two of individual values in paired tests, and in all cases y ences smaller than 300 psi (2.07 MPa) used in determining were statistical paired t test was applied to data for each A and to total data to determine wher average comdistrict from two districts had statistically significant differences. data se data as well as those from or two districts However, 138 Richmond i00 Lynchburg Std. Dev. 676 601 657 572 Sulfur- Std. Dev. of Std. Dev. Neoprene. Differences Avg. 705 108 236 4,396 637 29 240 3,957 654 5 176 4,198 591 58 206 4,057 4,173 673 55 221 RESULTS Compressive Strength close to each or. difference was 55 psi (379 kpa). The standard deviations average differences shown in Table 1 were obtained from differ- of number of samples for statistical analysis. strength values were significantly different. (4) Based pressive on 0.01 significance level, as shown in Table 2, it was found that a had differences so small as to have no practical significance. Table i. 14-Day Compressive Strengths in psi Sulfur Neoprene Avg. Diff. of No. Tests District 4,504 3,986 Salem i00 4,203 4,115 i00 Staunton All Districts 438 4,228 661 i psi 6.89 kpa

F-ratio test was applied to test data to compare A.n of two test methods, and no statistical difference variability additi6n, a linear regression analysis was made on data In determine degree of association between two test methods. to results for neoprene pads were taken as independent variable Test those for sulfur-mortar caps as dependent variable. The and measurement is explainable by or. The slopes obtained one linear regression analyses were compared to line of from using a statistical test, and it was found that at 0.01 equality level slopes were statistically different from significance 45 a for total data and also for data for each district except line This is shown in Table 5. one. plot of linear regression analysis on total data is A in Figure 2. The standard error of estimate was 216 psi (1.49 shown which is low and indicates a good relationship between two MPa), methods. The regression line indicates that strength values test test methods are equal at 4,920 psi (33.9 MPa). Below this for cylinders tested with neoprene pads yielded slightly value values, and above it y yielded slightly higher results. lower differences in values are small and can be neglected However, strength ranges tested. The slightly lower values at low for levels could result from stretching or flowing of strength pad. At higher loads flow of pad is restricted neoprene values would be closer to those for specimens, with and caps. It is also possible that at high strength levels sulfur-mortar ultimate strength of sulfur-mortar is reached, local stresses a 3,000 psi (20.7 MPa) strength value for neoprene pads, At regression line predicts a 3,138 psi (21.6 MPa) value for caps. The difference is not judged to be significant sulfur-mortar an engineering standpoint. from was found at a 0.01 significance level, as shown in Table 3. and intercept of best fitting line and correlation slope are given in Table 4. The correlation coefficients coefficients from 0.927 to 0.964, indicating a high correlation between ranged two capping methods. Thus, more than 86% of variation in are created, and result is lower strength values.

2. Results of Paired t-test Table 4, p. 3.31) (Ref. Std. Dev., No. of Sulfur-Neoprene, psi Tests psi 108 236 138 53 Yes Richmond 29 240 100 63 No Lynchburg 5 176 100 46 No Salem 58 206 I00 54 Yes Staun=on Richmond Lynchburg Salem S=aunton 676 601 657 572 3. Results of F-Ratio Test Table 4, p. 4.8) (Ref. S Dev. s Std. 0.92 1.61 0.6Z No 705 0.89 1.68 0.59 No 637 1.01 1.68 0.59 No 654 0.94 1.68 0.59 No 591 Error of In=ercept, Std. Estimate psi psi Slope % rrelat ion Co 0.903 227 533 94.2 &ichmond 0.874 226 527 92.7 Lynchburg 0.968 176 139 96.4 Salem 0.907 200 436 93.8 Staun on Richmond Lynchburg Salem Staun on slope i Significantly Slope, to.01 s s of 3.527 2.62 Yes 0.0275 3.530 2.62 Yes 0.0357 1.185 2.62 No 0.0270 2.743 2.62 Yes 0.0339 Significant Avg. Sd Diff..995 cn District All 4 Districts 55 221 438 27 Yes 2 Std. Dev. 2 Neoprene, SN, S F S I/F ignificant.995.995 Sulfur, SS, psi si District N All 4 Districts 661 673 0.96 1.29 0.78 No Table 4. Results of Linear Regression Analysis District All 4 Districts 0.928 216 354 94.6 Table 5. Comparison of Slopes to Line of Equality S=d. Dev. District Different All 4 Districts 0.0153 4.706 2.58 Yes *i psi 6.89

-':-.::: :: "L. ":: "d."..'".: i' "'"" :':"".1:::;...'11 :..." Y \ : " \ // 2 3 S 6 7 ///// with sulfur-mortar caps to neoprene pads tests error of estimate was 216 psi (1.49 MPa). Standard 7 Y = X / ".. " ".y.o:...': = 0.928 x + 0.354 :'::.::'z4":: ::: ; : :'. "..: 3. ' / Tested with Neoprene, ksi 2. Simple linear regression analysis of 14-day Figure strength test data correlating compressive NOTE: I ksi = 6.89 MPa.

Type of Failure on occurrence of different types of failure, as Data in Figure 3, are tabulated in Table 6. Figure 4 displays depicted types of breaks in a histogram for total test data. The indicate that most of breaks for both test methods results a combination of shear and splitting types of failure. How- are re is a tendency for more shear, or conical, type failure ever, cylinders tested with sulfur-mortar caps than for those with for pads, and more splitting type failures in latter. neoprene would anticipate that at higher strength levels re would One more restraint to flow of neoprene pads causing tangential be at surface as found in specimens with sulfur-mortar stresses Therefore, types of failure for specimens with strengths caps. 4,500 psi (31.0 MPa) were plotted in a histogram as shown above Figure 5. It appears that on a percentage basis, less splitting in of specimens attaining high strength when tested Some neoprene pads shattered when y failed, as did some tested with sulfur-mortar caps. There was no dangerous scattering of with pieces beyond bounds of machine in any of concrete However, for safety reasons, it is advisable to use a tests. net, such as a cage, around specimen during testing restraining was observed for se specimens as compared to all speci- failure However, percentage of conical type failures was not mens. as much as that with sulfur-mortar caps shown in Figure 4. with both test methods, especially at high strength levels.

E F G H Lynchburg S al em S t aunt Richmond on N S N S N S N S 22 2 3 7 i I 3 A 8 2 6 B 38 9 2 4 15 26 20 i0 C 15 27 32 25 22 26 61 40 D 6 25 3 12 6 15 4 4 E 9 20 24 31 40 12 9 14 F 2 14. 38 25 8 15 2 22 G i i 2 5 3 i H 2 9 2 2 12 21 29 31 15 5 19 20 19 ii 2 i 3. Types of failure of 6 x 12 in. (150 Figure 300 nun) x (From Ref. 2) cylinders. Table 6 of Break Types Shown in Figure 3 as a Occurrence of Totals Percentage (S Sulfur, N Neoprene) TOTAL Type of N S Breaks I0

of Break Type Sulful-Mortar.,4 80 4. Types of breaks for total test data Figure accordance with Figure 3. in Type of Break B C D E F G H A of Break Type Breaks 50 ps (3 0 NP ) and tested above 140 140 120 120 100 I00 80 o 0 60 o WO 2O 20 / Neoprene 5O B C D E F G H neoprene

From standpoint of engineering evaluations of strength i. and quality assurance testing, negligible differences levels difference between two procedures was statistically at 0.01 significance level, average of all significant with neoprene pads was only 55 psi (379 kpa) lower tests average for cylinders capped with sulfur-mortar than on 438 pairs of test data. The linear regression based indicated a good correlation between two test analysis At strengths below 4,920 psi (33.9 MPa) results methods. cylinders tested with neoprene pads were slightly for than results for cylinders capped with sulfur- lower but at strengths above 4,920 psi (33.9 MPa) opposite mortar, true. The difference in strength values is considered was from an engineering standpoint. negligible The comparison of overall variability of two test 2. at a 0.01 significance level revealed no statistical methods Cylinders with high strengths shattered when y failed, 3. for reasons of safety, care should be exercised. A re- thus The inside diameter of ring used in this study was 6-1/4 in. 4. mm) with +0 nd -I/16 in. (-2 mm) tolerance. Diameters (159 than 6-1/4 in. (159 mm) are not recommended because of larger possibility that neoprene could flow around ends affect test results. The size of inside diameter and restricted by convenience of locating specimen in is The neoprene pads with a 50 durometer hardness can be used 5. physical damage is observed. In this study, all until pads were still in usable condition after i00 tests. testing of concrete cylinders with neoprene pads confined The steel end caps is recommended as an acceptable alternate to test- in more information is available on effects of hardness on until strength. CONCLUSIONS found between results of tests made on uncapped were x 12 in. (150 x 300 mm) concrete cylinders tested with 6 pads confined in steel end caps and cylinders neoprene in standard manner with sulfur-mortar. Even though capped differences. straint around cylinder is desirable. steel cap and normal variability of specimen diameter. RECOMMENDATION with sulfur-mortar caps. Neoprene pads may be used until y ing damage. A neoprene hardness of 50 durometer should be used show 12

thanks are given to J. E. Galloway, Jr., T. R. Special A. D. Barnhart, R. H. Canody, A. D. Newman, and Blackburn, E. Steele for ir generous assistance in collection R. data. Appreciation is extended to employees of of whose assistance made this study possible. The Department by S. N. Runkle and C. S. Hughes in statistical assistance and by P. S. Harris in computer applications is analysis much appreciated. very ACKNOWLEDGEMENT 13

Ozyildirim, H. C., "Neoprene Pads for Capping Concrete i. VHTRC 79-R39, Virginia Highway and Transporta- Cylinders," Research Coh cil, Charlottesville, Virginia, March tion 1979. Grygiel, J. S., and D. E. Amsler, "Capping Concrete Cylinders 2. Neoprene Pads," Research Rep.ort 46, New York State Depart- with Bowery, F. J., and F. T. Higgins, "The Evaluation of Alternate 3. Procedure for Concrete 28-Day Quality Assurance Speci- Capping Natrella, M. G., Experimental Statistics, National Bureau of 4. Handbook 91, August 19'8'i. Standards Road and Bridge >peci ications, Virginia Department of Highways 5. Transportation, January 1978. and REFERENCES ment of Transportation, ew Yo 'k' " 1977. Woodward Clyde, Consultants, Rockville, Maryland, January mens," 1978. 15