Spectroscopy has shown the presence of the CN band dicothomy and the Na-O anticorrelations for 50--70% of the investigated samples in the cluster 47 Tuc, otherwise considered a normal prototype of high metallicity clusters from the photometric analysis. Very recently, the re-analysis of a large number of archival HST data of the cluster core has been able to put into evidence the presence of structures in the Sub Giant Branch: it has a brighter component with a spread in magnitude by $sim$0.06 mag and a second one, made of about 10% of stars, a little fainter (by $sim$0.05 mag). These data also show that the Main Sequence of the cluster has an intrinsic spread in color which, if interpreted as due to a small spread in helium abundance, suggests $Delta$Y$sim$0.027. In this work we examine in detail whether the Horizontal Branch morphology and the Sub Giant structure provide further independent indications that a real --although very small-helium spread is present in the cluster. We re--analyze the HST archival data for the Horizontal Branch of 47 Tuc, obtaining a sample of $sim$500 stars with very small photometric errors, and build population synthesis based on new models to show that its particular morphology can be better explained by taking into account a spread in helium abundance of 2% in mass. The same variation in helium is able to explain the spread in luminosity of the Sub Giant Branch, while a small part of the second generation is characterized by a small C+N+O increase and provides an explanation for the fainter Sub Giant Branch. We conclude that three photometric features concur to form the paradigm that a small but real helium spread is present in a cluster that has no spectacular evidence for multiple populations like those shown by other massive clusters.