SPI – Serial Peripheral Interface

SPI (Serial Peripheral Interface) is a full duplex synchronous serial communication interface used for short distance communications. It is usually used for communication between different modules in a same device or PCB. SPI devices communicates each other using a master slave architecture with a single master. Multiple slaves are supported through individual slave select lines.SPI is called as a 4-wire bus as it requires four wires for its communication as shown below. In the case of single slave communications we need only 3 wires, as slave select (SS) is not required. So SPI requires more communication lines in contrast to UART, I2C, USB etc.Hope you got the basic idea of an SPI bus. Now let’s see in little more detailed level like how it is working in hardware level. SPI having relatively very simple hardware compared to other serial communication protocols like UART, I2C etc. Its operation is based on shift registers (usually 8 bit). Each devices (master and slave) uses shift registers to shift data in and out of the system.For the sake of explanation, I omitted Buffer Register. It acts as an interface between user (processor, programmer) and SPI. Usually shift register won’t be directly accessible. So if we need to transmit data, we will write it to the buffer register. So it will automatically written to shift register when it is free and transmission will start. Similarly data is received in the shift register is automatically transferred to buffer register once the reception is complete. We can easily read from it. Thus buffer register will avoid all glitches that can happen if we try to read or write to shift register directly while transmission is taking place.