The battle between analog and digital isn’t over. Sure, many of our modern electronics and communications are based on digital processes but when it comes to sensors the direction isn’t quite as clear. Many sensors offer digital interfaces as well as analog interfaces, this is a requirement so that all systems needs are met. KEMET’s passive infrared/non-dispersive infrared (PIR/NDIR) sensors also come in both analog and digital interfaces, choosing the right one is critical to delivering optimal system performance.
Digital interfaces come in many varieties each one has its pros and cons. Perhaps one of the most popular digital interfaces is the Inter-IC Communication interface (IIC, I2C, or I2C) developed by Philips Semiconductor back in the 90’s. This is a very simple to use 2-wire digital interface that uses a clock and data signal and thanks to an open drain architecture they can use the same nets to communicate from client to host and from host to client. KEMET’s digital variety of environmental sensor use an I2C interface.
I2C isn’t the only type of digital interface. There is also a serial peripheral interface (SPI) that uses a 4-wire interface to connect peripherals to a host processor. Another common type of digital interface is a pulse width modulation (PWM). PWM is a kind of a combination of analog and digital interfaces where the signal levels themselves are discreet but the amount of time the signal is present can be used to indicate different sensor levels.
Creating a digital interface around a sensing element is not easy. It requires packaging sensing elements with digital register transfer level (RTL) and signal conditioning all together. Even after all that is done interfacing between a host processor and a sensor requires knowledge of the register map of the sensor which is well-documented by the manufacturer of the sensor.
Even with all that digital interface are extremely common because it takes signal conditioning and analog to digital conversion out of the hand of the end user.
Analog sensors can be thought of as those sub blocks of digital all decoupled and broken out. The process of signal conditioning and analog-to-digital conversion is not in the hands of the engineering implementing the sensor. While this allows for more control and optimization the task of signal conditioning, filter, and calibration is one that must be carefully done, or the sensor won’t work at all. This is one of the reasons why engineers prefer digital sensors to analog. One of the benefits, though, is that analog implementations bring is that they can be more carefully optimized to a desired signal level or bandwidth in addition to being an overall lower power implementation.
From a software standpoint implementing analog sensor requires a bit less software effort. Analog sensors require setting up analog-to-digital converters than can be within or without the host processor. They also require setting up a polling scheme as they do not have interrupts to let the host processor know that a certain signal level has been reached. Digital sensors require a bit more software effort such as setting up driver logic that reads the sensor values. One of the big benefits of digital sensors is that interrupts can be set up to alert a host processor.
Determining which one to go with is not a trivial matter, different systems can make significant benefit from each and many times it comes down to which variety engineers and circuit designers or more accustomed to implementing. It is for this reason that we offer our environmental sensors in both varieties.
Whether analog or digital, we’ve got you covered with application notes to help you along your design process:
Synchronizing I2C Interface: This application note explains how to synchronize multiple PIR / NDIR sensors to a single I2C interface
Calculating I2C Throughput: This application note explains the process of estimating the digital data throughput of an PIR / NDIR sensors in on an I2C interface.
Determining Linearization Coefficients: This application note explains how to calculate the coefficients to calibrate and linearize the output of PIR / NDIR sensors.
Browse the product offering of our environmental sensors that cover gas, flame, motion, as well as food, General IR Spectroscopy and Oil Monitoring Sensors.
