The development of software simulators for robotic applications involves many advantages such as faster trajectory planning in virtual worlds, savings in energy consumption, reduction in the purchase of real robots, feasibility in collecting datasets for AI model training, etc. Commercial, free-access, open-source simulators for any type of robot can be found on the Internet and in the literature. Some of them allow the implementation of specific or new control models. For a behavior closer to the real world, the following must be taken into account: kinematics and dynamics, gravity, masses, etc. This paper proposes a new simulator for a neuromorphic robotic arm (modified Scorbot ER-VII) whose hardware controllers are based on Field Programmable Logic Arrays (FPGA) to allow the implementation of neuro-inspired motor controllers, such as Spiking Proportional-Integral-Derivative (SPID), to manage each joint of the robot. This simulator is developed in Matlab and Simulink and is available to the public. The simulator allows for the evaluation of different trajectories with two controllers, an Independent Joint Control (IJC) and a Computerized Torque Control (CTC). It has been tested for square and lemniscate trajectories with a small number of internally interpolated points to create reference trajectories with more points. RSME improvements of more than 10x have been measured for CTC over IJC, requiring less than 13 ms per point evaluated for a 3.6 GHz i7 computer and 32 GB RAM.