Longitudinal painting has been presented as an elegant technique to fill the longitudinal phase space at injection to the CERN PSB once it is connected with the new Linac4 [1]. Painting brings several advantages related to a more controlled longitudinal filamentation, lower peak line density and beating reduction, resulting in a smaller space-charge tune spread. This could be an advantage especially for high intensity beams (> 6 × 1012 protons per bunch) to limit losses on the transverse acceptance of the machine. This paper presents an overview on the possible advantages of the technique for operational and test beams, taking care of the hardware limitations and possible failure scenarios. INTRODUCTION In the frame of the upgrades foreseen for the LHC injectors (LIU project [2]), Linac4 [3] will send Hions on to a stripping foil at 160 MeV. The fully stripped H+ ions will be injected in the PSB, while the partially stripped H0 and unstripped Hwill be dumped. The increase in βγ2 by a factor two, with respect to the present 50MeV injection from Linac2, will double the brightness in the PSB assuming the same transverse space charge (SC) tune spread of today. An optimisation of the longitudinal parameters [4] could further enhance the brightness achievement obtained by the increase in the injection energy. Two different schemes are foreseen: “un-modulated” and “modulated” injection. The “un-modulated” injection scheme is the baseline for HL-LHC beams. It consists in a multi-turn injection of chopped trains of Linac4 bunches (at 352.2 MHz) with a fixed bunch length and energy spread. In this configuration the energy offset ∆E0 is equal to 0 and the number of injection turns to integrate the desired number of charges in the PSB depends on the current delivered by the linac. Latest estimations for the future LHC standard intensity (3.4×1012 p) predict 23 turns, considering 40 mA before chopping, 63% chopping factor (CF), 403 keV rms energy spread (δE) and 100% transmission [4]. A disadvantage of the “un-modulated” injection scheme is the limited possibility to uniformly fill the buckets and thus minimise the longitudinal filamentation. This effect is due to pure geometrical reasons, as the almost rectangular shape of the injected bunches do not match with the longitudinal iso-Hamiltonian contour (target area, see Fig. 1). Almost all the bunches are injected into the PSB in a double harmonic (h) radiofrequency (RF) bucket with voltages (V) in anti-phase, in ∗ vincenzo.forte@cern.ch order to smear the peak line density and, thus, reduce the transverse SC tune spread. LIU beams will be produced by injecting the train of Linac4 bunches in an accelerating bucket. The result of dedicated simulations is presented in Fig. 1 and shows the longitudinal phase space of the first injected bunch in the PSB for Vh=1=8 kV, Vh=2=6 kV and Û Bρ = 10 Tm s , for a target matched area of 1.5 eVs at 160MeV.