Coplanar shadowgrams provide an affordable mean to retrieve the 3d shape of an object especially when classical stereopsis fails (eg:textureless objects). Its principles are similar to the concepts used for Shape-From-Silhouettes with the only exception that here, light sources and cameras are interchanged. However, it is well known that any attempt to use the shadowgram to retrieve light sources positions is subject to a 4-parameter ambiguity. In this paper, we show how using the light spot visible in the camera reduces this ambiguity to a single parameter. We also suggest some practical solutions to gain a supplemental constraint on light sources positions and break the ambiguity. We demonstrate the effectiveness of our method using synthetic and real images.