We have developed a stereoscopic large LED display with parallax barrier for use by the general public and stereoscopic cameras to show real world images in 3D. This paper aims to analyze stereoscopic camera separation and convergence angle to make the most use of a field of interest and the reproducible space provided by the large stereoscopic LED display. We describe the principle of a stereoscopic LED display with a parallax barrier and its reproducible space that is determined by the allowable range of disparity to fuse stereoscopic images. By using a model of stereoscopic imaging and display process, we introduce the formulas of the reproduced positions on our developed stereoscopic LED display. Furthermore, we analyze relationships between the stereoscopic camera separation, the convergence angle, the area of a field of interest, and the depth range of the reproduced space. The results show there are four categories in camera configurations: there are three kinds of camera configurations that have different characteristics and one configuration that is not recommended. Category A configuration reproduces a wide area of the field of interest in a long range of depth. Category B functions as a reduction of the field of interest. Category C functions as a magnification of the field of interest. In Category D, a narrow area of the field is reproduced in a short range of depth. In particular, for use by stereoscopic LED display with a rather low resolution, Category A and Category C are recommended because they fully use the reproducible positions.