Laser ultrasonic nondestructive testing technology has become a research hotspot due to its advantages of non-contact, wide frequency, and multi-mode. However, it is also because of its wide-band and multi-mode characteristics that the technology is difficult to identify and extract characteristic signals and determine the receiving position in practical applications. Especially when it comes to the test piece with a complex shape, the generated ultrasonic wave is reflected in the test piece for many times, which increases the difficulty of testing and seriously affects the practical application of this technology. In addition, in order to facilitate the research, the excitation light source is often regarded as an ideal point source, however, the size of the light source is one of the important factors in the actual measurement process. On the basis of previous studies, 1060 aluminum alloy plates with three thicknesses of 5 mm, 10 mm, and 15 mm are taken as the research objects in this paper, and the light source diameters of 1 mm, 2 mm, 3 mm, 4 mm, 5 mm and 6 mm are respectively set to excite the aluminum alloy plates. The Michelson interferometer scans the back of the plates to receive signals. Firstly, the arrival time and amplitude of each echo are obtained from the experimental A-scan data. By analyzing the arrival time of each echo, the echo propagation path under the excitation of light sources with different diameters is derived, and the relationship between the arrival time of each echo of laser ultrasound and the diameter of the excitation light source is given. The directivity formula of ideal point source given by HUTCHINS D A et al. is analyzed, and on this basis, the directivity diagram of laser ultrasonic sound field in polar coordinate system is drawn by using the amplitude of primary longitudinal wave and primary transverse wave excited by light sources with different diameters, from which the law of the directivity of sound field changing with the diameter of the excitation light source is analyzed. The results show that as the detection laser moves away from the center, the amplitude of the longitudinal wave gradually decreases, and then it is submerged in the transverse wave. The arrival time of the longitudinal wave is difficult to extract, and the arrival time of the transverse wave will move forward, deviating from the theoretical arrival time. However, if the influence of waveform aliasing is ignored, the maximum error between the experimental value and the theoretical value of each echo arrival time spread formula given in this paper is not more than 0.1 μs. In addition, by analyzing the amplitude distribution of the primary longitudinal wave and the primary shear wave, it can be seen that an excitation source with a diameter of a≤1 mm can generate a shear wave sound field with good directivity near ±45°, and the longitudinal wave has a considerable amplitude in almost all directions. With the increase of the diameter of the excitation light source, the sound beam of the longitudinal wave sound field becomes narrower and the directivity becomes better, while the shear wave energy gradually moves to the normal direction away from ±45°. However, when the diameter increases to a certain extent, the acoustic fields of both will be affected by the thermoelastic effect. In the longitudinal wave directivity diagram, there are obvious small side lobes around ±30°~±60°, and the energy of the shear wave concentrated in the normal direction will diffuse to both sides. The research results of this paper show that the diameter of the excitation light source is a non negligible factor for the study of the distribution of laser ultrasonic sound field. In practical application, the appropriate diameter of the light source should be selected according to different detection needs.