The present study explores the impact of two different microphysical parameterization schemes (i.e. Zhao and Carr, Mon Wea Rev 125:1931-1953, 1997 Ferrier, Amer Meteor Soc 280-283, 2002: called as BF) of National Centers for Environmental Prediction (NCEP) Climate Forecast System version 2 (CFSv2) on Indian summer monsoon (ISM). Critical relative humidity (RHcrit) plays a crucial role for the realistic cloud formation in a general circulation model (GCM). Hence, impact of RHcrit along with microphysical scheme on ISM is evaluated in the study. Model performance is evaluated in terms of simulation of rainfall, lower and upper tropospheric circulations, cloud fraction, cloud condensate and outgoing longwave radiation (OLR). Climatological mean features of rainfall are better represented by all the sensitivity experiments. Overall, ZC schemes show relatively better rainfall patterns as compared to BF schemes. BF schemes along with 95 % RHcrit (called as BF95) show excess precipitable water over Indian Ocean basin region, which seems to be unrealistic. Lower and upper tropospheric features are well simulated in all the sensitivity experiments; however, upper tropospheric wind patterns are underestimated as compared to observation. Spatial pattern and vertical profile of cloud condensate is relatively better represented by ZC schemes as compared to BF schemes. Relatively more (less) cloud condensate at upper level has lead to relatively better (low) high cloud fraction in ZC (BF) simulation. It is seen that OLR in ZC simulation have great proximity with observation. ZC (BF) simulations depict low (high) OLR which indicates stronger (weaker) convection during ISM period. It implies strong (weak) convection having stronger (weaker) updrafts in ZC (BF). Relatively more (less) cloud condensate at upper level of ZC (BF) may produce strong (weak) latent heating which may lead to relatively strong (weak) convection during ISM. The interaction among microphysics, thermodynamics, and dynamics works in tandem through a closed feedback loop.