In this study, 16 reinforced concrete (RC) simple-span beams were test
ed statically up to failure to quantitatively investigate the arch act
ion. Variables included four shear-span ratios (2, 2.5, 3, and 4), two
longitudinal steel ratios (1 and 2%), and the presence or absence of
stirrups. The measured longitudinal reinforcement tensions in the shea
r span were significantly higher than the calculated values by beam th
eory. This may be attributed to the reduction of the internal moment a
rm length by the development of the arch action. The measured steel te
nsion in a shear span was resolved into two components-one for beam ac
tion and one for arch action. After reaching the cracking load, the ra
tio of steel tension from arch action to that from beam action was con
stant, ranging from 0.2 to 1.2 at the center of the shear span. At ult
imate load, over the entire length, the reinforcement force in the bea
ms without web reinforcement and with shear span-to-depth ratio smalle
r than 3 was almost constant because the beams acted as tied arches. O
n the basis of the experimental findings, an equation is proposed to p
redict the internal moment arm length, which then leads to a new metho
d to calculate the steel tension.