Abstract Efficiency calibration and coincidence summing correction have been performed for two large arrays of NaI(Tl) detectors in two different configurations. They are, a compact array of 32 conical detectors of pentagonal and hexagonal shapes in soccer-ball geometry and an array of 14 straight hexagonal NaI(Tl) detectors in castle geometry. Both of these arrays provide a large solid angle of detection, leading to considerable coincidence summing of gamma rays. The present work aims to understand the effect of coincidence summing of gamma rays while determining the energy dependence of efficiencies of these two arrays. We have carried out extensive GEANT4 simulations with radio-nuclides that decay with a two-step cascade, considering both arrays in their realistic geometries. The absolute efficiencies have been simulated for gamma energies from 700 to 2800 keV using four different double-photon emitters, namely, 60Co, 46Sc, 94Nb and 24Na. The efficiencies so obtained have been corrected for coincidence summing using the method proposed by Vidmar et al. [11] . The simulations have also been carried out for the same energies assuming mono-energetic point sources, for comparison. Experimental measurements have also been carried out using calibrated point sources of 137Cs and 60Co. The simulated and the experimental results are found to be in good agreement. This demonstrates the reliability of the correction method [11] for efficiency calibration of two large arrays in very different configurations.