In this paper, the global identification of spring balancer, dynamic parameters and joint drive gains of a 6 Degrees Of Freedom (DOF) robot is performed. Off-line identification method is based on the use of the Inverse Dynamic Identification Model (IDIM) which takes into account a spring balancer for gravity compensation and linear Least Squares (LS) technique to estimate the parameters from the positions and joint torques. It is key to get accurate values of joint drive gains to get accurate identification because the joint torques are calculated as the product of the current reference by the joint drive gains. Recently a new method validated on small payload robots (less than 10 Kg) allows to identify simultaneously all joint drive gains and dynamic parameters. This method is based on the Total Least Squares (TLS) solution of an over-determined linear system obtained with the inverse dynamic model calculated while the robot is tracking reference trajectories without load and trajectories with a known payload fixed on the robot. This method is used to identify accurately the heavy industrial robot Kuka KR270 (270Kg payload) with its spring balancer. This is a new step to promote a practical and easy to use method for global dynamic identification of any small or heavy gravity compensated industrial robots that does not need any a priori data, which are too often missing from manufacturer's data sheet.