Most Design Codes encourage the staggering of lapped splices in a section, such that a portion of the bars in the cross section remain continuous throughout the lap zone. A penalty on lap length may be imposed if all bars are lapped at one section. In the literature, however, most of test specimens have had all reinforcement lapped at the same section. Therefore, this work aims to present experimental results of several tests to study the behaviour of bond strength of lapped joints of bundles where only one or two bars are lapped at a section. For this purpose, four point bending test on 24 full-scale beams with two different bar diameter and different percentage of bars lapped at the mid-span have been carried out. All the beams have been designed with the same concrete cover and with the minimum amount of transversal reinforcements, equal to 0.5 the area of lapped bars. The splitting and flexural cracking behaviour, the resistance and the ductility of the joint have been measured and compared with the results of the beam with continuous bars. The results are in good agreement with the semi-empirical expression included in the new proposal of fib TG 4.5 (Bond Models) for the new Model Code.