Tensile Strength of Welded Splices of QST Reinforcing Bars

(ProQuest Information and Learning: ... denotes formula omitted.)

INTRODUCTION

Quench and self tempering (QST), which is also referred to as TEMPCORE or the Thermex process, and micro-alloying of steel are two current methods for achieving the required properties in hot-rolled bars throughout

the world.1-11 In the QST process, the bar is intensively sprayed with high-pressure water for a short time immediately as it exits the last hot-rolling sequence. This transforms only the surface of the bar to a martensitic structure, while the core remains austenitic. After leaving the quench, the quenched surface layers are tempered by the conserved heat from the core. As a result, the bar has a composite metallurgical structure with a strong martensitic surface layer and a relatively soft, ductile ferrite-pearlite core.

QST bars show distinct advantages with respect to ductility both before and after welding, fatigue properties, impact toughness, and economy when compared with micro-alloyed hot-rolled bars.1-3 QST treatment of the steel also improves its corrosion resistance.12 The strength of such reinforcing bars after heating, however, is questionable1,11 because the quenching effect can be reduced or even completely eliminated by hot working of the material as occurs when welding. As an example, despite the fact that Grade 500 (characteristic yield strength f^sub y^ = 500 N/mm^sup 2^ [73 ksi]) QST bars are considered weldable in many countries of Europe,5-7,9 welding such bars is prohibited in New Zealand.11

It should be noted that Grade 500 reinforcing bars are the most modern type of reinforcing bars that have been introduced instead of or along with reinforcement of Grades 400 to 460 during the last two decades in countries of Europe,3,6,7,9 Australia and New Zealand,8 Canada,13 India,10 and others. There are also European and international standards on Grade 500 reinforcing bars.4,5

According to current standards, reinforcing steel is considered weldable if chemical elements content and carbon equivalent (CE) value of the steel do not exceed maximum values prescribed by the standards. CE value is a composite analytical parameter that defines a hardenability of the steel, which is the ability of the steel to form hard, brittle metallurgical structures after a particular heating and cooling cycle. There are various equations for calculating CE.1,4-6,9,14,15 In the present study, Eq. (1), which is adopted in ISO 6935,4 prEN 10080,5 BS 4449,6 STO ASChM 7,9 ASTM A 913,15 and many other standards, was employed for calculating CE.

... (1)

where the symbols of the chemical elements indicate their content in percent by mass (C equals carbon, Mn equals manganese, Cr equals chromium, Mo equals molybdenum, V equals vanadium, Cu equals copper, and Ni equals nickel).

The requirement of a maximum CE value prevents brittle failures and provides a required ductility of welded QST bars. As will be shown in the current study, however, the limitation of the maximum CE value is a necessary but insufficient requirement for weldability of QST bars, because weldability of steel bars is defined not only by ductility of their welded splices, but also by their strength, which can be significantly reduced after welding if a CE value is too low.

In addition to a CE value, self-tempering temperature (STT), which is the maximum temperature of bar surface after quenching, affects the properties of QST bars significantly. For QST bars with the same chemical composition, yield and tensile strengths increase by STT reduction. Consequently, required mechanical properties of QST bars can be achieved by simultaneous reduction of CE and STT. As mentioned previously, however, excessive decreasing of CE and STT can lead to a significant strength reduction of the bars after welding. Thus, specifying a minimum STT that requires a sufficiently high CE value to obtain the required strength of QST bars is the indirect way of providing the required strength of QST bars after welding. It should be noted that the standard specification for QST steel shapes15 requires the minimum STT (that is, 600 ?C) along with the maximum CE values, whereas current standards for reinforcing bars, specifying the maximum CE values, limit neither a minimum STT nor a minimum CE value.

The objectives of this paper are to analyze extensive test data on tensile strength of welded Grade 500 QST bars, to develop equations for calculating tensile strength of welded Grade 500 QST bars, and to determine minimum CE values that ensure the required tensile strength of the welded Grade 500 QST bars.

RESEARCH SIGNIFICANCE

Results of the present study provide important information on tensile strength of welded QST reinforcing bars. The necessity of minimum limitation on a CE value to guarantee a required tensile strength of welded QST bars is shown. The minimum CE values determined based on the developed statistical models of tensile strength of the welded bars, which are functions of CE, bar diameter, and tensile strength of the bars before welding, are proposed.

DESCRIPTION OF TEST DATA

The test data included 1872 tensile strength results of cross-, lap-, and butt-welded Grade 500 QST bars that were tested in the Laboratory of Reinforcement at the Research, Design, and Technological Institute of Concrete and Reinforced Concrete (NIIZhB) over the course of approximately 15 years. The test specimens were welded by electric resistance spot welding (ERSW), manual metal arc welding (MMAW), flash welding (FW), enclosed welding (EW), and enclosed welding with a lap strap (EWLS).

Reinforcing bars

QST bars of Grade 500 that represented 125 heats of steel manufactured at seven steel mills in Russia, Ukraine, Belarus, and Moldova according to the requirements of STO ASChM 79 were used to fabricate welded specimens. Minimum, maximum, and average values of chemical elements content and CE of the tested bars from the mill test certificates are given in Table 1. Table 1 also shows the requirements for the chemistry of Grade 500 bars, from a heat analysis, specified by STO ASChM 79 and prEN 10080.5

Statistical data on yield strength f^sub y^, tensile strength f^sub u^, the ratio of f^sub u^/f^sub y^, elongation at fracture A^sub 5^, and total elongation at maximum force A^sub gt^ of the tested bars, as well as the requirements of STO ASChM 79 and prEN 100805 for mechanical properties of Grade 500 bars are given in Table 2.

As can be seen from Tables 1 and 2, chemistry and mechanical properties of the bars, which were used to make the welded specimens, met not only the requirements of STO ASChM 7,9 according to which they had been made, but also the prEN 100805 requirements specified for Class B500B reinforcement.

Welding processes, types of welded joints, and details of tested specimens

The following welding processes were used to fabricate the tested specimens:

Electric resistance spot welding (ERSW)-ERSW is used for the connection of crossing bars by welding machines primarily in a shop. ERSW results from a combination of pressure and heat generated at the welding interface by the electrical resistance of the joint.

Manual metal arc welding (MMAW)-MMAW was used to make cruciform specimens by tack welding (MMATW) and lap specimens both without additional lap bars (MMALW) and with additional lap bars (MMAWLB). Tack welding is the welding process that is used for the connection of crossing bars by small arc welds primarily at a job site. Tack welding is prohibited or not recommended for the connections of high-strength micro-alloyed reinforcing bars16,17 because such welding may cause embrittlement of the bars. In Russia, however, tack welding is permitted for the connection of QST bars because they possess high ductility even after welding.3 MMALW and MMAWLB are used for the connections of bars by seam welds mainly at a job site.

Flash welding (FW)-FW is used for resistance butt welds that result from a combination of pressure and heat generated at butt-ends of splicing bars by the electrical resistance of the joint. This welding process requires a welding machine and is primarily used in a shop. Two modes of the FW were used to make the specimens-without and with preheating of the bars. FW without preheating is optimal for QST bars and is sometimes referred to as hard mode (FWHM). FW with preheating is sometimes termed soft mode (FWSM) and is characterized by a higher heat input when compared with FWHM. Despite the fact that FWSM is not optimal for QST bars, it was used for fabrication of the test specimens, because this mode of FW can take place in practice.

Enclosed welding (EW)-EW is a butt weld that results when the butt-ends of bars are melted by a heat of molten weld pool. This welding process is performed in copper molds that are set on the joint enclosing a gap between the bars and filled with a metal electrode. EW is primarily used at a job site for the connection of reinforcing bars with diameters of 20 mm (0.79 in.) and greater.

Enclosed welding with lap strap (EWLS)-EWLS is EW when a stay-in-place steel lap strap is used instead of the copper molds to enclose a gap between the bars and to provide butt weld. When the butt weld is completed, the lap strap is fillet-welded to the bars after which it can provide a load transfer from one bar to the other together with the butt weld.

Details of the welded joints are presented in Fig. 1 to 3. Welding processes, types of welded joints, and details of tested specimens were set in accordance with GOST 14098-91.17

Testing of welded specimens and test results

All welded specimens were tested in tension. Tensile tests were performed on Russian-made R-100, MR-500, and MR-100 hydraulic tensile testing machines under a constant stress rate of 10 N/mm^sup 2^ per second in a laboratory at 15 to 25 ?C. Not less than three specimens of each bar diameter from each heat of steel were made and tested. When the cruciform specimens were tested, tensile load was applied to the ends of the main (larger) bar of the specimen to determine its weakening that resulted from welding.

During the tests, only the tensile strength of the specimens was measured. Along with tensile tests, bend tests were also carried out on the similarly welded specimens. Bend test results are summarized elsewhere.3 Because an analysis of ductility of welded QST bars is beyond the framework of this study, it is not considered herein. It should be noted, however, that all tested specimens withstood the bending angle of 180 degrees around a 3d forming pin without visible cracking on the outside of the bent portion.

Summary statistics on tensile strength f^sub u,w^ and the ratio of f^sub u,w^/f^sub u^, which shows a reduction in tensile strength after welding compared with the original bar material, are presented in Table 3. For each welding process and type of joint, quantities of the tested specimens, heats of the steel from which the bars were selected, and CE values of the steel are also shown in Table 3. All tested specimens failed in a ductile mode.

Failures of approximately 98% of the cruciform specimens, which were fabricated by both ERSW and MMATW, occurred at the sections of the bars away from the heat affected zone (HAZ), and only 2% of the cruciform specimens failed at the sections at the HAZ with insignificant reductions in tensile strength compared with the strength of unwelded companion bars from the same heats. The greatest reductions were 2.8 and 3.3% of the original tensile strength for the bars welded by ERSW and MMATW, respectively. All of the tested cruciform specimens resisted more than the specified tensile strength of the parent bars (that is, 600 N/mm^sub 2^ [87 ksi]). The lap- and butt-welded bars showed a greater strength reduction in comparison with the cruciform specimens (refer to Table 3).

Test results of the bars welded by EW stand separately from the other results, because tensile strength of such specimens was reduced virtually to the tensile strength of the original hot-rolled steel (without QST treatment). In other words, QST effect almost completely disappeared after EW. Such degradation of strength of the bars occurred because butt-ends of the bars were melted when EW was carried out. For this reason, EW of QST bars is prohibited in Russia and will not be considered further.

The addition of the steel lap strap with length of 4d to the joint welded by EW, which results in EWLS, resulted in a significant increase in tensile strength of the welded specimens. Furthermore, the use of EWLS resulted in welded joints having no reduction in tensile strength after welding.

If test results on the bars welded by EW are not taken into account, the greatest reduction in strength of the welded QST bars was observed after FW, especially when FWSM was used. The strength of bars was reduced by FWSM by 11.5% on average. The maximum values of strength reduction (by more than 30% and to 508 N/mm^sup 2^ [74 ksi]) were observed for the bars with diameters from 32 to 40 mm (1.26 to 1.57 in.) with relatively low CE values (from 0.28 to 0.33%).

The specimens welded by FWHM showed a smaller tensile strength reduction with an average of 5.9% compared with the specimens welded by FWSM. Nevertheless, the maximum values of strength reduction, which were also observed for large bars with low CE values, remained considerable (by 29% and to 540 N/mm^sup 2^ [78 ksi]).

QST bars welded by MMALW and MMAWLB showed very close levels of strength reduction. The strengths of those types of joints were reduced on average by 3.6 and 3.9%, respectively. The maximum strength reduction amounted to 20 and 23%, respectively. Overall, the strength reductions of the lap-welded bars were smaller than the strength reductions of the butt-welded bars.

The data given in Table 3 show that the average tensile strength of QST bars welded by welding processes considered in this study, except for EW, exceeded the specified tensile strength of the original bars. Considering that the lowest strengths were observed for the bars with relatively low CE values, minimum CE values, which will provide a required strength of welded joints, can be determined. In other words, tensile strength reduction might be significantly decreased or even completely eliminated by selecting proper CE values.

ANALYSIS OF TEST DATA

Developed statistical models

A multiple nonlinear regression analysis of the test data was conducted to determine relationships (statistical models) between strength of the bars welded by the welding processes considered in this study and d, CE, and f^sub u^. The regression analysis of the strength of the cruciform joints welded by ERSW and MMATW was not conducted, because within the range of the data (refer to Tables 2 and 3) strength of the bars welded by ERSW and MMATW was practically equal to the strength of the original bars.

As was mentioned previously, the most important factors affecting mechanical properties of both original unspliced and welded QST bars are chemical composition characterized by CE value and QST level defined by STT. In addition, strength of welded QST bars is affected by their size d. Unfortunately, STT temperatures for the tested bars were not reported. Despite that, QST intensity can be estimated indirectly by using mechanical properties of the original bars, when CE and d are also known. Conducted analysis showed that, compared with other mechanical properties, tensile strength of the original bars in combination with their CE and d provides the best correlation with strength of the welded bars. Thus, the equations of regression for each type of joints were chosen in the form of full quadratic model with three factors. Such a form of the model allows for interaction between the three variables and the quadratic effect of variables on the response f^sub u,w^

f^sub u,w^ = b^sub 0^ + b^sub 1^d + b^sub 2^CE + b^sub 3^f^sub u^ + b^sub 4^d^sup 2^ + b^sub 5^CE^sup 2^ + b^sub 6^f^sup 2^^sub u^ + b^sub 7^d ? CE + b^sub 8^d ? f^sub u^ + b^sub 9^CE ? f^sub u^ + b^sub 10^d^sup 3^ + b^sub 11^CE^sup 3^ + b^sub 12^f^sup 3^^sub u^ + b^sub 13^d^sup 2^ ? CE + b^sub 14^d^sup 2^ ? f^sub u^ + b^sub 15^CE^sup 2^ ? d + b^sub 16^CE^sup 2^ ? f^sub u^ + b^sub 17^f^sup 2^^sub u^ ? d + b^sub 18^f^sup 2^^sub u^ ? CE + b^sub 19^d ? CE ? f^sub u^ (2)

where b^sub 0^ to b^sub 19^ are coefficients to be determined from regression analysis.

The coefficients b^sub 0^ to b^sub 19^ were determined from a multiple nonlinear regression analysis, which was performed for each welding process considered in this study. Subsequently, unimportant terms were dropped out from the model. As a result, Eq. (3) to (7), which establish relations between tensile strength of welded Grade 500 QST bars and chemistry, mechanical properties, and size of the original bars, were developed.

For the bars welded by MMALW

f^sub u,w^ = -107.29d + 15,837.7CE - 6.388f^sub u^ + 2.473d^sup 2^ - 43,134.1CE^sup 2^ + 0.01064f^sup 2^^sub u^ + 0.193d ? f^sub u^ - 0.0402d^sup 3^ + 60.41f^sub u^ ? CE^sup 2^ - 0.000151d ? f^sup 2^^sub u^ - 0.03057CE ? f^sup 2^^sub u^, N/mm^sup 2^ (3)

in which 8 = d (mm) = 32; 0.25 = CE (%) = 0.47; 650 = f^sub u^ (N/mm^sup 2^) = 750; and f^sub u,w^ = f^sub u^.

For the bars welded by MMAWLB

f^sub u,w^ = -2666.5CE + 1.867f^sub u^ - 1.485d^sup 2^ - 0.00251f^sup 2^^sub u^ + 0.0818d ? f^sub u^ + 5.9814f^sub u^ ? CE + 3.613d^sup 2^ ? CE - 0.2008d ? f^sub u^ ? CE, N/mm^sup 2^ (4)

in which 12 = d (mm) = 40; 0.27 = CE (%) = 0.47; 650 = f^sub u^ (N/mm^sup 2^) = 750; and f^sub u,w^ = f^sub u^.

For the bars welded by FWHM

f^sub u,w^ = -1535.8 + 99.1d + 4467.3CE + 3.363f^sub u^ - 2.19d^sup 2^ - 15,002.8CE^sup 2^ - 0.00124f^sup 2^^sub u^ - 0.1464d ? f^sub u^ + 0.004d^sup 3^ + 11,323.2CE^sup 2^ + 0.0027f^sub u^ ? d^sup 2^ + 50.4d ? CE^sup 2^ + 3.1f^sub u^ ? CE^sup 2^, N/mm^sup 2^ (5)

in which 10 = d (mm) = 40; 0.25 = CE (%) = 0.47; 650 = f^sub u^ (N/mm^sup 2^) = 750; and f^sub u,w^ = f^sub u^.

For the bars welded by FWSM

f^sub u,w^ = 5290.54 - 41.18d - 19.795f^sub u^ + 0.659d^sup 2^ - 38,661.57CE^sup 2^ + 0.019654f^sup 2^^sub u^ + 131.012d ? CE + 36.7853CE ? f^sub u^ - 1.435CE ? d^sup 2^ + 57.08f^sub u^ ? CE^sup 2^ - 0.0554CE ? f^sup 2^^sub u^ - 0.0602d ? CE ? f^sub u^, N/mm^sup 2^ (6)

in which 10 = d (mm) = 40; 0.25 = CE (%) = 0.47; 650 = f^sub u^ (N/mm^sup 2^) = 750; and f^sub u,w^ = f^sub u^.

For the bars welded by EWLS

f^sub u,w^ = 38,620.8 + 1041.34d - 198.418f^sub u^ - 0.155d^sup 2^ + 33,803.3CE^sup 2^ + 0.378644f^sup 2^^sub u^ + 1698.73d ? CE - 3.8643d ? f^sub u^ - 104.84CE ? f^sub u^ - 50,463.4CE^sup 3^ - 0.0002466f^sup 3^^sub u^ + 723.2CE^sup 2^ ? d + 0.00355f^sup 2^^sub u^ ? d + 0.137944f^sup 2^^sub u^ ? CE - 3.0777f^sub u^ ? d ? CE, N/mm^sup 2^ (7)

in which 28 = d (mm) = 40; 0.27 = CE (%) = 0.47; 650 = f^sub u^ (N/mm^sup 2^) = 750; and f^sub u,w^ = f^sub u^.

For each type of welding joints, statistical characteristics of the estimations of tensile strength of the welded bars (coefficients of determination [R^sup 2^], averages and coefficients of variation [COV] of the ratio of f^sup test^^sub u,w^/f^sup calc^^sub u,w^, relative errors at 95% confidence limit, and standard errors for the f^sub u,w^ estimate [S]) are given in Table 4. The statistical data demonstrate that the developed statistical models appear to be reasonable in representing the factors affecting tensile strength of welded Grade 500 QST bars.

Figures 4 to 8 show contour plots of f^sub u,w^ and f^sub u,w^/f^sub u^ drawn according to Eq. (3) to (7) as functions of CE and f^sub u^ for three values of d (minimum, maximum, and interim values from the tested ranges).

As can be seen from both Fig. 4 to 8 and Eq. (3) to (7), the relationships between f^sub u,w^, f^sub u,w^/f^sub u^ and d, CE, and f^sub u^ are significantly nonlinear. Despite some exceptions, there is a tendency of increasing absolute strength of the welded bars f^sub u,w^ with increasing CE and f^sub u^ when all other variables are equal. Relative strength of the welded bars f^sub u,w^/f^sub u^ also increases with an increase of CE, but decreases with an increase of f^sub u^. In other words, increasing f^sub u^, which is achieved by reducing STT when CE is constant, results in increasing strength reduction of QST bars after welding.

For QST bars welded by MMAWLB, FWHM, FWSM, and EWLS, there is a tendency of reduction of f^sub u,w^ and f^sub u,w^/f^sub u^ with an increase of bar diameter. For the bars welded by MMALW, effect of bar diameter on strength of the bars is not uniform. By increasing the diameter from 8 to 14 or 16 mm (0.31 to 0.55 or 0.63 in.) (depending on values of CE and f^sub u^), the strengths of the welded bars f^sub u,w^ and f^sub u,w^/f^sub u^ decrease. Further increase of bar diameter to 25 or 28 mm (0.98 or 1.10 in.) (also depending on values of CE and f^sub u^) results in an increase of the strengths of the welded bars f^sub u,w^ and f^sub u,w^/f^sub u^. Finally, increase of the diameter to 32 mm (1.26 in.) again causes a decrease of the strengths of the welded bars f^sub u,w^ and f^sub u,w^/f^sub u^. Apparently, such results were caused by the bending that occurred in the specimens welded by MMALW because of the eccentricity of load transfer through them that depends on diameter of the welded bars.

Minimum CE values that provide required strength of welded Grade 500 QST bars

The developed statistical models were employed to determine minimum CE values that provide the required tensile strength of welded QST bars. The required strength of welded Grade 500 QST bars was chosen to be at least equal to the specified tensile strength of the parent bars (that is, 600 N/mm^sup 2^ [87 ksi]). In other words, strength of the welded bars was chosen to be at least equal to 1.2f^sub y^, where f^sub y^ is the specified yield strength of the original bars. ACI 318-0216 prescribes more stringent requirements, at first glance, for tensile strength of full welded joints; namely, it should be equal to or greater than 1.25f^sub y^. It should be noted, however, that the weldable reinforcement that is used for design of reinforced concrete according to ACI 318 should comply with the requirements of ASTM A 706,14 according to which the ratio of f^sub u^/f^sub y^ is equal to 1.33. In other words, ACI 318 allows tensile strength of welded bars to be lower than the specified tensile strength of unspliced bars by approximately 7%. Thus, the requirements used in the present work are actually more stringent in comparison with ACI 318 requirements, because they do not allow tensile strength of welded bars to be less than the specified tensile strength of the parent bars.

For each welding process and bar diameter, CE values that provide tensile strength of welded joints averaging 600 N/mm^sup 2^ (87 ksi) were calculated by substituting the average strengths of the parent bars, which ranged from 707 to 721 N/mm^sup 2^ (103 to 105 ksi) for different types of welding processes, in Eq. (3) to (7), equating them to the required strength of welded bars (that is, 600 N/mm^sup 2^ [87 ksi]), and solving them for CE. If the calculated CE values were less than the lower limit values of the CE data ranges, they were taken equal to the lower limit values (Table 5).

According to many standards, characteristic tensile strength of reinforcement is based on 5% fractile (90% confidence that there is a 95% probability that the actual strength will exceed required strength). It is evident that safety of structures with both unspliced and spliced bars should be equivalent. Therefore, the tensile strength of welded bars should be considered as a characteristic value, which should be guaranteed with the same level of probability as the tensile strength of parent bars.

A similar procedure as that for calculation of the CE values that provided tensile strength of welded joints averaged 600 N/mm^sup 2^ (87 ksi) was used for determination of the minimum CE values, which provided the required strength of welded joints as a characteristic value; however, the values f^sub u,w^ in Eq. (3) to (7) were equated to 600 + k ? S (a normal distribution was assumed), where k is a coefficient corresponding to a reliable failure rate of 5%, and S is the standard error for f^sub u,w^ (refer to Table 4). The coefficient k was taken in accordance with Table 5 of prEN 100805 as a function of the number of test results. It was equal to 1.78, 1.79, 1.77, 1.77, and 1.84 for MMALW, MMAWLB, FWHM, FWSM, and EWLS, respectively. The calculated minimum CE values are given in Table 5. They were also taken equal to the lower limit values of the CE data ranges if they were less than the lower limit values.

As can be seen from Table 5, the different minimum CE values, which also depend on bar diameter, correspond to each type of welding joints. These differences can be attributed to different conditions of heating and cooling of the bars at the splice zone caused by distinctions between the welding processes.

A greater minimum CE value corresponds to FW, especially when preheating is used (FWSM). The reason for this is a higher heat input in FW, and particularly in FWSM, which results in a greater strength reduction of QST bars after welding when compared with the other considered welding processes. Consequently, greater CE values are necessary to decrease reduction in strength of QST bars after FW.

MMALW, MMAWLB, and EWLS (that is, the welding processes used mainly at a job site) require slightly lower CE values to provide the required strength of welded QST bars in comparison with FW. To avoid any possible confusion and to retain unification of reinforcing steel, however, it appears that the minimum CE values should be the same for any type of welding, because weldable bars should have a chemistry that ensures the required strength of the bars welded by whatever type of welding process. Thus, for each bar diameter, the generalized minimum CE values (which guarantee the required strength of Grade 500 QST bars welded by welding processes considered in this study) were calculated as a maximum of the CE values obtained for each type of welding process considered herein (refer to Table 5).

The conducted analysis has shown that the minimum CE values increase with an increase of bar diameter. For comparison, the minimum CE value is equal to 0.25% for Grade 500 QST bars with diameters of 8 and 10 mm (0.31 and 0.39 in.), whereas it is equal to 0.43% for the bars with diameters from 28 to 40 mm (1.10 and 1.57 in.).

It should be also noted that the obtained minimum CE values for the bars with diameters from 8 to 40 mm (0.31 to 1.57 in.) are lower than the maximum CE values (that is, 0.50%) required by ISO 6935, prEN 10080, STO ASChM 7, and other standards for weldable reinforcement of Grade 500. Thus, CE values ranging from the minimum given in Table 5 to the maximum of 0.50% appear to guarantee both the required ductility of welded QST bars of Grade 500 and their required strength.

SUMMARY AND CONCLUSIONS

The conducted analysis of test results of Grade 500 QST bars welded by six different welding processes has shown the need of limitation for minimum CE values to guarantee the required tensile strength of welded QST bars. The minimum CE values of Grade 500 QST reinforcement with diameters from 8 to 40 mm (0.31 to 1.57 in.) were determined based on the developed statistical models of tensile strength of the welded bars, which are functions of CE, bar diameter, and tensile strength of the bars before welding. Study results have demonstrated that bars with a greater diameter require a higher minimum CE value.

ACKNOWLEDGMENTS

The author wishes to express his gratitude to S. A. Madatian, head of the Laboratory of Reinforcement at the Research, Design, and Technological Institute of Concrete and Reinforced Concrete (NIIZhB), Moscow, Russia, and to L. A. Zborovskiy, leading engineer of the Laboratory of Reinforcement at NIIZhB, for their assistance in gathering data.

NOTATION

A^sub 5^ = elongation of reinforcing bar at fracture, percent

A^sub gt^ = total elongation of reinforcing bar at maximum force, percent

COV = coefficients of variation, percent

d = diameter of reinforcing bar, mm

ERSW = electric resistance spot welding

EW = enclosed welding

EWLS = enclosed welding with lap strap

FW = flash welding

FWHM = flash welding without preheating (hard mode of flash welding)

FWSM = flash welding with preheating (soft mode of flash welding)

f^sub u^ = tensile strength of reinforcing bar, N/mm^sup 2^

f^sub u,w^ = tensile strength of welded splice of bars, N/mm^sup 2^

f^sup calc^^sub u,w^= calculated tensile strength of welded splice of bars, N/mm^sup 2^

f^sup test^^sub u,w^= tensile strength of welded splice of bars obtained from tests, N/mm^sup 2^

f^sub y^ = yield strength of reinforcing bar, N/mm^sup 2^

k = coefficient corresponding to reliable failure rate of 5%

MMALW = manual metal arc lap welding

MMATW = manual metal arc tack welding

MMAW = manual metal arc welding

MMAWLB = manual metal arc welding with additional lap bars

S = standard errors for f^sub u,w^ estimate, N/mm^sup 2^