The AER protocol is widely used in neuromorphic vision sensors to encode data in time or frequency and transfer it from the pixel to the receiver using a shared channel. Data throughput, power consumption and latency is reduced; however, they present several timing limitations to scale the size of the array. This paper analyzes the effect of increasing the number of pixels in terms of transmission delay and proposes several design guidelines to overcome scaling limitations in high-resolution pixel arrays, achieving a reduction of more than 70% in the time delay of the request line.