Recently, first neutrino-driven supernova explosions have been obtained in 3D, self-consistent, first-principle simulations, these models are still not always exploding robustly and, in general, the explosions are not sufficiently energetic. To constrain the explosion mechanism, and the related uncertainties, it is thus very helpful to consider observational constraints: pulsar kicks, progenitor association and supernova remnants (SNR). Recent observations of asymmetries in the supernova ejecta of Cas A are very promising, to compare to long-term simulations of the explosion. In addition 3D observations of SN87A are becoming more constraining on the geometry of the ejected material during the explosion. In this talk I will discuss our efforts to model the late time evolution of a 3D supernova explosion, where we include the effects of beta decay, which inflates the structures rich in 56Ni. The structures we find in the simulations depend on the quantities plotted.