Shear capacity of indirectly loaded reinforced concrete beams under and after fire exposure

Abstract

To investigate the shear capacity of indirectly loaded reinforced concrete beams under and after fire exposure, load tests were conducted on eight full-scale specimens exposed to fire testing in a furnace chamber, and the effects of additional transverse reinforcement in the junction region between the primary and secondary beams on the shear capacity, fire resistance, failure modes and deflection were analysed. The results indicate that the slopes of the diagonal cracks in post-fire tested reinforced concrete beams without additional transverse reinforcement were shallower than those of a similar reference beam not exposed to fire, and that the ultimate capacities of reinforced concrete beams with additional transverse reinforcement decreased obviously after fire exposure. However, beams with additional transverse reinforcement exhibited increased fire resistance times and reduced strains in their reinforcement, indicating the benefit of conservatively providing such reinforcement. The findings of this study are expected to provide a reference for the improved fire-resistant design of indirectly loaded beams.

Keywords

additional transverse reinforcement deflection fire resistance indirect load post-fire behaviour

Introduction

The shear failure mechanism of reinforced concrete (RC) beams under indirect loading via secondary beams is obviously different from that of RC beams under direct loading, causing the shear capacity of combined primary and secondary beam systems to be considerably lower than that of either beam in isolation (Cao et al., 2017; Chen, 2014; Deng et al., 2003; Lan, 1996; Mao et al., 1990; Shu and Wei, 1985; Taylor, 1960; Wang, 1981; Zhou et al., 2018; Zsutty, 1971). To address this shear capacity problem, the structural design of indirectly loaded RC beams usually requires additional transverse reinforcement to strengthen the junction between the primary and secondary beams. However, exposure to high temperatures is well known to have detrimental effects on the mechanical properties of concrete and steel materials, resulting in the severe redistribution of internal forces and deterioration of the capacity of members fabricated using such materials; accordingly, the occurrence of a fire can endanger the safety of an RC building (Bamonte et al., 2018; Dong, 2001; Fu et al., 2018; Gabriela et al., 2018; Kodur and Naser, 2018; Wu, 2003). Upon exposure to fire (such as a common fire on three or four sides), the mechanical properties of the primary beam in the tension zone can be seriously degraded, making the indirect loading mode even more unfavourable and further increasing the risk of shear failure. Therefore, it is important to study the failure mechanism and load capacity of indirectly loaded RC beams considering the influence of high temperature.

Current research on the shear capacity of indirectly loaded RC beams exposed to high temperatures is insufficient: few previous studies have investigated the shear capacities and failures of indirectly loaded beams exposed to fire, focusing instead on indirectly loaded beams at ambient temperatures. Wang (1981) conducted shear tests on 41 directly and indirectly loaded simply supported RC beams, finding that the shear strengths of indirectly loaded beams were 0%–63.4% lower than those of directly loaded beams. Zhou et al. (2018) conducted tests using three RC beams with 500-MPa high-strength stirrups to analyse their structural behaviour and failure modes under direct and indirect loading, finding that the shear capacity of high-strength concrete beams under indirect load decreased compared with the beams under direct load due to the variation of the shear span ratio and the stirrup ratio. Shear tests conducted by Taylor (1960) found that the shear capacity of indirectly loaded RC beams without shear reinforcement decreased by 0%–70% compared with directly loaded beams. In order to evaluate the ultimate capacities of indirectly loaded RC beams during and after exposure to fire, in this study, eight full-scale RC beams with and without additional transverse reinforcement in the junction region of the primary and secondary beams were designed and tested using the fire simulation system.

Experimental programme

Test specimens

In order to study the influence of additional transverse reinforcement in the junction region of the primary and secondary beams on the shear capacity of indirectly loaded primary beams during and after exposure to fire, eight full-scale RC beam specimens were designed for indirect loading following the Chinese design code (GB 50010-2010, 2015). Two specimens were not exposed to fire and employed as references (CWJJ1 and CWJJ2), two specimens were subjected to full fire resistance performance tests under simultaneous indirect loading and high temperature (JJL1 and JJL2) and the remaining four specimens were exposed to fire then subjected to indirect load shear tests afterwards (JJLH1–4). The dimensions of the specimens are shown in Figure 1. The length and net span of the primary RC beams were 4000 and 3600 mm, respectively. The cross-sectional size was 250 mm × 450 mm for the primary beam and 200 mm × 300 mm for the secondary beam, and the concrete cover was 25 mm for both. For the longitudinal reinforcement of the primary beams, 2ϕ16 bars were placed in the top face, and 2ϕ18 and 1φ20 bars were placed in the bottom face, resulting in a reinforcement ratio ρ of 0.823%. The shear reinforcement of the primary beams consisted of 2ϕ6 stirrup legs spaced at 150 and 200 mm to provide a stirrup ratio ρ_sv of 0.151%. To provide indirect shear loading of the simply supported primary RC beam, two pairs of cantilevered secondary beams projecting 300 mm from the side faces of the primary beam were provided to a shear span-to-depth ratio λ of 2.1 for all primary beams.

Figure 1.

Specimen dimensions and reinforcement details (dimension in mm): (a) specimens CWJJL1, JJL1, JJLH1 and JJLH2; (b) specimens CWJJL2, JJL2, JJLH3 and JJLH4, showing only additional transverse reinforcement; (c) 1-1: primary beam; (d) 2-2: secondary beam; and (e) 3-3: loading of secondary beams.

Specimens CWJJL1, JJL1, JJLH1 and JJLH2 were designed following the ‘strong bending and weak shear’ design principle to ensure shear failure, whereas specimens CWJJL2, JJL2, JJLH3 and JJLH4 were provided with additional transverse reinforcement following the ‘strong shear and weak bending’ principle to study the effect on shear bearing capacity. Designed according to Chinese Standard Code (GB 50010-2010, 2015), the additional transverse reinforcement consisted of 6ϕ6 stirrups and 2ϕ12 steel hangers. The additional stirrups were placed within the junction region of the primary and secondary beams, defined as a primary beam length of 2h₁+ 3b centred on the secondary beam connection, where h₁ is the difference between the effective depth of the primary beam and the height of the secondary beam, and b is the width of the secondary beam. The main parameters of all specimens are presented in Table 1, and their dimensions and reinforcement details are shown in Figure 1.

Table 1.

Beam specimen parameters and test conditions.

Specimen	Cross section (mm)	ρ_sv (%)	ρ (%)	λ	Additional shear reinforcement	Pre-load (kN)^a	Fire time (min)
CWJJL1	250 × 400	0.151	0.823	2.1	/	0	0
JJL1	250 × 400	0.151	0.823	2.1	/	0.4P_u	Failure time
JJLH1	250 × 400	0.151	0.823	2.1	/	0	90
JJLH2	250 × 400	0.151	0.823	2.1	/	0.4P_u	90
CWJJL2	250 × 400	0.151	0.823	2.1	6ϕ6 + 2ϕ12	0	0
JJL2	250 × 400	0.151	0.823	2.1	6ϕ6 + 2ϕ12	0.4P_u	Failure time
JJLH3	250 × 400	0.151	0.823	2.1	6ϕ6 + 2ϕ12	0	90
JJLH4	250 × 400	0.151	0.823	2.1	6ϕ6 + 2ϕ12	0.4P_u	90

P_u is the ultimate capacity of the reference beam at ambient temperature.

All specimens were fabricated using commercial concrete consisting of P.O. 42.5 (Chinese cement grading system) ordinary Portland cement from Jinfeng City, Grade I fly ash (Chinese fly ash grading system) from Jiangsu Huawang and aggregates in the mix proportions shown in Table 2. The average 150-mm concrete cube compressive strength was 29.4 MPa. The material properties of the steel reinforcing bars are provided in Table 3.

Table 2.

Concrete mix proportions.

Material	Specification	Unit weight (kg/m³)	Mix proportions (by weight)
Cement	P.O. 42.5 Portland cement	206	1.00
Fine aggregate	Medium sands	752	2.46
Coarse aggregate	5–25 mm	1042	3.41
Water	Purified water	180	0.59
Admixture	JX-1	8.08	0.03
Fly ash	Grade I	97.77	0.32

Table 3.

Mechanical properties of reinforcement.

Bar	Diameter d (mm)	Yield strength f_y (MPa)	Ultimate strength f_u (MPa)
ϕ6	6	410	508
ϕ10	10	440	620
ϕ12	12	482	634
ϕ16	16	455	610
ϕ18	18	447	593
ϕ20	20	430	575

Test set-up and instrumentation

The fire exposure tests were conducted in a 9000-mm long, 4500-mm wide and 1500-mm high horizontal furnace chamber at Shandong Jianzhu University, shown in Figure 2(a). During the fire exposure test, rock wool sheets were wrapped around the cantilevered secondary beams to serve as fire protection, as shown in Figure 2(b). The static shear loading devices for the reference and post-fire exposure specimens were consistent with those used for loading during the fire exposure test, as shown in Figure 2(c).

Figure 2.

Fire exposure and indirect shear loading test set-up (dimensions in mm): (a) furnace chamber, (b) specimens arrangement in furnace chamber, (c) dimensions of shear loading apparatus and (d) loading apparatus during fire exposure tests.

The six fire test specimens were exposed to fire temperatures on three sides. In the fire tests, the heating system adopted the standard ISO 834 (1999) fire curve to control the heating, while the temperature in the furnace was measured by thermocouple. These specimens were tested in three batches. The first batch consisted of specimens JJLH1 and JJLH3, which were exposed to fire temperatures under only self-weight. The second (JJLH2 and JJH4) and third batches (JJL1 and JJL2) were pre-loaded to 0.4 times the ultimate capacity of the reference specimen at ambient temperature by the loading device prior to and throughout the application of high temperature, as shown in Figure 2(d). After fire exposure, all specimens were cooled to ambient temperature by natural cooling.

The furnace temperature, temperature distribution, applied load and deflection for each specimen were measured during the fire exposure tests in the locations shown in Figure 3. Five linear variable differential transducers (LVDTs) were used to measure the vertical displacement, and K-type thermocouples were installed to measure the temperature distribution in the beam section during the fire exposure test. The strains in the stirrups and longitudinal reinforcing bars, deflection and ultimate bearing capacity were measured during the ambient temperature static load testing of the reference and post-fire specimens in the locations shown in Figure 4.

Figure 3.

Locations of LVDTs and thermocouples (dimensions in mm).

Figure 4.

Locations of static load test instrumentation (dimensions in mm).

Failure criterion

During the fire resistance tests, a deformation failure criterion and a load resistance failure criterion were applied. The deformation failure criterion was defined in terms of deformation D (mm) and deformation rate dD/dt (mm/min) according to the Chinese Standard (GB/T9978.1-2008, 2009) as

D = \frac{L^{2}}{400 d}

(1)

\frac{dD}{dt} = \frac{L^{2}}{9000 d}

(2)

where L is the net span of the primary beam (mm) and d is the height between the most extreme compression fibre and the tensile force in the cross section of the primary beam (mm). Considering that the indirectly loaded primary beams evaluated in this study had an L of 3600 mm and a d of 320 mm, they were considered to have failed in deformation when either the deformation magnitude exceeded 101.25 mm or the deformation rate exceeded 4.50 mm/min. The load resistance failure criterion was defined as the brittle failure of the primary beam or when it could otherwise no longer sustain applied loading.

Results and discussion

Fire test application and analysis

Due to the limitations of the furnace equipment, strict temperature control could not be provided in accordance with the ISO 834 fire curve. The measured furnace temperatures were lower than those dictated by the standard ISO 834 temperature curve, but otherwise generally matched the geometry of the standard curve, as shown in Figure 5. The furnace temperatures for the first batch are consistently about 50 °C lower than the standard ISO 834 fire curve, those of the second batch are eventually about 100 °C lower than those of the standard curve and those of the third batch increase more slowly in the beginning of the fire test (t = 30 min), ending at a final temperature about 200 °C lower than the standard curve. These inconsistent differences with ongoing batches may be due to reasons such as the reuse of the test furnace system or the tightness of the cover slabs. For instance, the use of fireproof cotton between the cover slab would produce gaps, so that its heat insulation function was weakened. The difference in the performance of burners and the exhaust system also caused fluctuation in the furnace temperature.

Figure 5.

Measured time–temperature curves in the furnace.

The temperatures measured by the thermocouples in the specimens during the fire exposure tests are shown in Figure 6. Note that during the fire exposure tests, unloaded specimens JJLH1 and JJLH3 were only subjected to their self-weight, so their deflections did not change appreciably and their surfaces exhibited mostly irregular short cracks. However, the deflections of pre-loaded specimens JJLH2 and JJLH4 increased considerably during the fire exposure test, and slight diagonal cracks appeared in their shear-bending zones. It can be observed in Figure 6 that the temperatures near the bottom faces of the loaded specimens are higher than those of the unloaded specimens, and that their temperature peaks are relatively high. This may be the result of the pre-load, which induces a certain degree of cracking in the primary beams prior to the fire exposure test, enabling the elevated furnace temperature to more easily migrate into the interior of the beam. In addition, the temperatures of the concrete in the junction region of the primary and secondary beams were lower than those in the primary beam itself; this is due to the reduction in exposed surfaces in the junction region.

Figure 6.

Measured time–temperature curves of specimen thermocouples: (a) JJL1, (b) JJL2, (c) JJLH1, (d) JJLH2, (e) JJLH3 and (f) JJLH4.

The failure modes of fire-resistance-tested specimens JJL1 and JJL2 are shown in Figure 7 after natural cooling. Specimen JJL1 exhibited a shear failure at the junction of the primary and secondary beams at a fire resistance limit of 131 min and fell into the furnace. The longitudinal bars of the primary beam were all observed to have broken, possibly because it exhibited obvious diagonal cracks under the thermodynamic effect, allowing the heat to pass through the diagonal cracks, reducing the strength of the longitudinal reinforcement and the stirrups and leading to the failure under the coupled action of applied temperature and load. In the primary beam of specimen JJL2, vertical cracks up to 6-mm wide appeared, the tensile steel bars yielded and the compressive concrete was crushed, indicating flexural failure when the rate of deformation reached 4.50 mm/min; the fire resistance limit of specimen JJL2 was accordingly determined to be 160 min.

Figure 7.

Failure modes of specimens loaded to failure during fire resistance performance tests: (a) JJL1 and (b) JJL2.

Static load test phenomena and discussion

The failure modes of the reference specimens and post-fire exposure specimens under indirect static loading are shown in Figure 8, and their ultimate capacities are provided in Table 4. It can be seen from Figure 8 and Table 4 that specimens CWJJL1, JJLH1 and JJLH2 exhibited shear failure caused by diagonal cracks at the junction of the primary and secondary beams. The slope of the diagonal cracks in fire-damaged specimens JJLH1 and JJLH2 was shallower than those in reference specimen CWJJL1. In addition, the shear capacity of specimen JJLH2 (pre-loaded during the fire test) is lower than that of JJLH1 (unloaded during the fire test). The ultimate load capacity of specimen JJLH1 is slightly higher than that of specimen CWJJL1, which is contrary to the expected result of material deterioration under fire exposure. This higher ultimate load may be due to the large dispersion of shear failure loads. The shear failure of specimens CWJJL1, JJLH1 and JJLH2 belongs to the brittle failure, while the bending failure of specimens CWJJL2, JJLH3 and JJLH4 belongs to ductile failure. However, the shear failure has great dispersion. Specimens CWJJL2, JJLH3 and JJLH4 with additional transverse reinforcement all exhibited flexural failure, according to the ‘strong shear and weak bending’ design principle. As a result, the failure modes of the specimens subjected to the fire exposure test were the same as that of the specimen not subjected to the fire exposure test. However, the ultimate load capacities of the post-fire specimens with additional transverse reinforcement were considerably smaller than that of the reference specimen with additional transverse reinforcement, decreasing by 14.2% and 19.1% for specimens JJLH3 and JJLH4, respectively.

Table 4.

Ultimate capacities of specimens.

Specimen	Total ultimate load F (four-point total) (kN)	Ultimate load P_u (one load point) (kN)	Failure mode
CWJJL1	318	79.5	Shear
CWJJL2	408	102.0	Bending
JJLH1	335	83.75	Shear
JJLH2	315	78.75	Shear
JJLH3	350	87.5	Bending
JJLH4	330	82.5	Bending

Figure 8.

Failure modes of specimens loaded to failure: (a) CWJJL1 and (d) CWJJL2 with no fire exposure; (b) JJLH1, (c) JJLH2, (e) JJLH3 and (f) JJLH4 after fire exposure tests.

Specimen stiffnesses

The deflection curves of the indirectly loaded specimens evaluated in this study are shown in Figure 9. Through analysis of these curves, the following conclusions can be obtained:

The deflection of specimens JJL1 and JJL2 loaded to failure during the fire exposure test changed only slightly up to a fire exposure time t of 90 min, after which the deflection rate of JJL1 (without additional transverse reinforcement) began to increase. At a fire exposure time t of 131 min, the deflection of JJL1 severely increased as the connection between the primary and secondary beams was severed, indicating the fire resistance limit. However, the deflection rate of specimen JJL2 (with additional transverse reinforcement) reached 4.50 mm/min at a fire time t of 160 min, indicating the fire resistance limit. Clearly, additional transverse reinforcement can effectively delay the failure time of indirectly loaded beams exposed to high temperatures.

The vertical deflections of shear failure specimens CWJJL1, JJLH1 and JJLH2 were 20.774, 23.954 and 18.829 mm, respectively. Thus, the application of high temperatures had no obvious effect on the stiffness of indirectly loaded beams failing in shear.

The mid-span deflections of flexural failure specimens CWJJL2, JJLH3 and JJLH4 reached 65.650, 28.530 and 32.060 mm, respectively, demonstrating that the deflection of specimens failing in flexure (yielding of tensile reinforcing and crushing of compressive concrete) decreases considerably after fire exposure.

Figure 9.

Specimen deflection curves: (a) specimens loaded to failure during fire exposure test and (b) other specimens loaded to failure.

Strains in specimen reinforcement

Considering the large volume of test data, only strain data from the tensile reinforcement at mid-span and a typical stirrup in the failure zone of the primary beam were used to draw the strain–load curves shown in Figure 10, from which the following conclusions can be obtained:

The strains in the primary tensile reinforcement at mid-span in the fire-damaged specimens were less than those in the undamaged reference specimens.

The strains in the primary tensile reinforcement of shear failure specimen JJLH1 (exposed to fire testing without applied load during the fire test) were greater than those in the primary tensile reinforcement of specimen JJLH2 (with an applied pre-load of 0.4P_u during the fire test). However, the strains in the primary tensile reinforcement of flexural failure specimen JJLH4 (with an applied pre-load of 0.4P_u during the fire test) were greater than those of flexural failure specimen JJLH3 (without applied load during the fire test). This is because specimens subjected to an applied pre-load of 0.4P_u exhibited a large number of fine cracks in their primary beams that allowed for the increased penetration of heat, reducing their stiffness. This applied pre-load had a more detrimental influence on the flexural failure specimens than on the shear failure specimens due to the reliance of the flexural failure specimens on the contribution of the reinforcement tensile and concrete compressive strengths.

The strains in the stirrups of flexural failure specimens JJLH3 and JJLH4 (with additional transverse reinforcement) exposed to the fire test were greater than those in the stirrups of reference specimen CWJJL2, which was not exposed to the fire test. However, the strains in the stirrups of shear failure specimens JJLH1 and JJLH2 exposed to the fire test were less than those in the stirrups of reference specimen CWJJL1, which was not exposed to the fire test. Clearly, additional transverse reinforcement has a considerable influence on the strains in the reinforcement of fire-damaged beams under indirect shear loading.

Figure 10.

Reinforcement strain–load curves: (a) strains in the tensile reinforcement at mid-span and (b) strains in typical stirrups in the failure zone.

Conclusion

This article presented the results of an experimental investigation of the shear behaviour of indirectly loaded RC beams during and after fire exposure. The main conclusions of this research are as follows:

The presence of additional transverse reinforcement was observed to effectively increase the failure time of indirectly loaded beams during fire exposure tests. In order to prevent the brittle shear failure of indirectly loaded beams exposed to fire, additional transverse reinforcement in the junction region of the secondary and primary beams should be conservatively designed.

Two failure modes were observed during the static load tests according to beam design: (a) brittle shear failure of the indirectly loaded primary beams without additional transverse reinforcement, caused by diagonal cracks at the intersection between the secondary and primary beams; and (b) flexural failure of the indirectly loaded primary beams with additional transverse reinforcement. Therefore, although fire damage was not observed to affect the failure mode of the indirectly loaded primary beams, the slope of the diagonal cracks in the primary beams without additional transverse reinforcement after fire exposure was observed to be shallower than those in primary beams not exposed to fire, and the ultimate load capacity of indirectly loaded primary beams with additional transverse reinforcement was observed to be considerably reduced after fire exposure.

The shear failure specimens without additional transverse reinforcement exhibited brittle failure with no obvious deflection. However, the flexural failure specimens with additional transverse reinforcement exhibited considerable deflection as the tensile reinforcement yielded and the compressive concrete crushed. Thus, the deterioration of the steel and concrete properties due to fire exposure resulted in the observed decrease in the stiffness of the indirectly loaded beams.

The provision of additional transverse reinforcement had a considerable influence on the strains in the reinforcement of indirectly loaded fire-damaged beams.

The findings of this study are expected to be useful to researchers and designers looking to improve the performance of indirectly loaded beams when exposed to fire. However, the numerical study on shear capacity of RC beams during and after fire exposure was not conducted in this study, and should be further investigated in future research studies.

Footnotes

Declaration of Conflicting Interests

The author(s) declared no potential conflicts of interest with respect to the research, authorship and/or publication of this article.

Funding

The author(s) disclosed receipt of the following financial support for the research, authorship and/or publication of this article: This research was supported by the National Natural Science Foundation of China (Project No. 51478254), which is gratefully acknowledged.

ORCID iDs

Yamin Song

Shuting Liang

Longji Dang

References

Bamonte

Gambarova

Kalaba

, et al. (2018) Some considerations on shear and torsion in R/C structural members in fire. Journal of Structural Fire Engineering 9(2): 94–107.

Cao

Chen

Song

, et al. (2017) Trial research on shear capacity of RPC beams under indirect loading. Journal of Huaqiao University (Natural Science) 38(3): 330–335.

Chen

(2014) Experimental study on shear behavior and finite element analysis of concrete beam with high-strength stirrup under indirect load. PhD Thesis, Tongji University, Shanghai, China.

Deng

Sun

(2003) Additional crosswise reinforcing steel in reinforced concrete beams of indirect loading. Journal of Shijiazhuang Railway Institute 16(3): 45–47.

Dong

(2001) Fire Safety Design of Concrete Structure. Beijing, China: Science Press.

Song

Yin

, et al. (2018) Experimental study on shear bearing capacity of reinforced concrete beams under thermodynamic coupling. Journal of Shandong Jianzhu University 33(5): 1–10.

Gabriela

Augusto

Joao

PCR

, et al. (2018) Behavior of thermally restrained RC beams in case of fire. Engineering Structures 174: 407–417.

GB 50010-2010 (2015) Code for Design of Concrete Structures. Beijing, China: China Architecture and Building Press.

GB/T9978.1-2008 (2009) National Standard of the People’s Republic of China, Fire-Resistance Tests-Element of Building Construction-Part 1: General Requirements. Beijing, China: Standards Press of China.

10.

ISO 834 (1999) Fire-Resistance Tests: Elements of Building Construction, Part 1.1: General Requirements for Fire Resistance Testing. Geneva: International Organization for Standardization.

11.

Kodur

VKR

Naser

(2018) Approach for shear capacity evaluation of fire exposed steel and composite beams. Journal of Constructional Steel Research 141: 91–103.

12.

Lan

(1996) Additional crosswise reinforcing steel in reinforced concrete beams of indirect loading. Journal of Suzhou Institute of Urban Construction and Environmental Protection 9(1): 27–32.

13.

Mao

Zhu

(1990) Experimental investigation on the reinforced concrete deep beam under indirect point-loads. Journal of Wuhan University of Hydraulic and Electric Engineering 23(4): 23–30.

14.

Shu

Wei

(1985) An exploration of shear strength of indirect load reinforced concrete cantilever beams. Journal of Guizhou University of Technology 3: 74–85.

15.

Taylor

(1960) Some shear tests on reinforced concrete beams without shear reinforcement. Magazine of Concrete Research 12(36): 145–154.

16.

Wang

(1981) Shear strength of RC beams under indirect load. Journal of Chongqing Institute of Architecture Engineering (1): 54–78.

17.

(2003) Mechanical Properties of Reinforced Concrete Structures after Fire. Beijing, China: Science Press.

18.

Zhou

Cai

Liu

, et al. (2018) Shear behavior analysis of concrete beams reinforced with 500 MPa stirrups under indirect load. Journal of Tongji University (Natural Science) 46(2): 156–161.

19.

Zsutty

(1971) Shear strength predication for separate categories of simple beam tests. ACI Journal Proceedings 68(2):138–143.