We model the visual light curve of CAL 87 based on the assumption that an accreting steadily burning white dwarf irradiates the accretion disk and the secondary star, as suggested by van den Heuvel et al. (1992). We use constraints on the geometry derived from the known orbital period. As sources of visual light we include the secondary star and an accretion disk with an optically thick, cold, clumpy spray at its riml presumably caused by an accretion stream of high mass flow rate impinging on the disk at the hot spot. This spray moving around the disk can account for the asymmetry in the light curve and the depth of the secondary minimum. It also might be the cause of the observed low X-ray luminosity if the white dwarf is permanently hidden by this disk.