KMP Automotive's beginner-friendly guide explains how CAN-Bus networks let a motorsport vehicle's electronics communicate, covering the hardware, how messages are structured and decoded, and what it takes to build a reliable bus. It uses simple analogies (like a train with wagons) and closes with a pitch for KMP's CAN engineering services.
At KMP Automotive, CAN-Bus sits at the heart of almost every electronics system we design, build and integrate. If you have heard the term but never quite understood what it means or what it does, this guide is for you. There is no shortage of highly technical documentation online — our aim here is different: to explain the fundamentals of CAN-Bus in clear, approachable language. Throughout this article we use a High-speed CAN 2.0A network (1 Mbit/s [megabits per second], 11-bit identifiers) as our reference, as this is the specification most commonly found in motorsport applications.
CAN stands for Controller Area Network. In simple terms, it is a method that allows microcontrollers (small, dedicated computers) to communicate with one another across a shared network, known as a bus. A helpful comparison is Ethernet (the LAN standard that lets office computers talk to each other) — CAN does much the same job, but it is purpose-built for the demanding environment of a vehicle.
|
The bus is a twisted pair of wires (two wires wound around each other to reject electrical noise) with a 120 Ohm termination resistor (a component fitted at each end to prevent signal reflections) on both sides. One wire is called CAN High and the other CAN Low. Both wires are required for reliable communication. A device connected to the bus is called a node. A CAN network always needs at least two nodes to communicate. In motorsport you will typically find between 3 and 10 nodes sharing a single bus. |
Every node contains at least a microcontroller and a CAN controller. The CAN controller is the translator: it converts digital information into messages that travel across the bus, then decodes incoming messages back into usable data. Think of it as Morse code — information goes in, the CAN controller encodes it into a defined language and sends it out, and the CAN controller at the receiving node translates it back again.
The microcontroller handles the actual work: reading sensor inputs, switching on lights or actuators (devices that create movement, such as a valve or motor), driving a dashboard display, or managing other forms of communication.
A CAN message can carry all kinds of data, but at its core it consists of two parts: an ID and a message frame of up to 8 bytes (a byte is a unit of 8 bits; a bit is a single 0 or 1).
The easiest way to picture this is a train. The train runs at a fixed speed (1 Mbit/s) along the rail (the bus) and is made up of several wagons (the IDs). At every station (node) you can add wagons to the train, or look inside a wagon to see what it is carrying. The wagon with the lowest number has the highest priority and is always handled first. A single node can send multiple IDs onto the bus.
Each wagon (ID) contains 8 private rooms (bytes), and each room has 8 seats (bits). Every seat holds a value of either 0 or 1. A full room of 8 seats can represent 256 different combinations — the maximum value a single byte can hold. In practical terms, one ID can carry several signals at once, such as RPM (engine revolutions per minute), throttle position and temperature. If each signal uses one byte, a single ID can carry up to 8 signals.
This is why it matters whether a signal is transmitted using 1 bit, 1 byte or 2 bytes. Suppose you want to send an engine speed with a range up to 8000 RPM using a single byte. Because one byte offers only 256 steps, the finest resolution you can achieve is 8000 / 256 ≈ 31 RPM — the value will only ever update in jumps of roughly 31 RPM.
For greater accuracy, you can assign 2 bytes to a single signal. Two bytes together provide up to 65,536 steps, so the same 8000 RPM range becomes 8000 / 65,536 ≈ 0.12 RPM per step — a dramatic improvement in precision. Knowing when to use 1 bit, 1 byte or 2 bytes for each signal is fundamental to designing an efficient CAN network.
To read the exact signal you are looking for, you need four things: the ID, the High Byte, the Low Byte and a quantification factor (Q-factor). The Q-factor converts the raw, unitless message value into a meaningful real-world figure with proper units — RPM, Volts, Bar, and so on.
How many IDs (wagons) a bus can handle depends on several factors. As a rule of thumb, once the bus load rises above 60–70%, you risk data loss or a complete communication failure — the equivalent of a crashed train. Poor physical conditions cause the same problems: bad wiring, untwisted cable or missing termination will all corrupt your data.
It is equally important that every node is compatible with the bus it is connected to. A node designed for a 500 kbit/s network cannot simply be placed on a 1 Mbit/s network — the speeds must match.
The comparisons in this guide are deliberately simplified and, we will happily admit, not always technically perfect. But in our experience of explaining CAN-Bus to newcomers, visualising it this way makes the fundamentals genuinely click.
At KMP Automotive we live and breathe this technology from designing individual nodes to engineering complete, dependable CAN networks for motorsport. If you would like to discuss a project, or you need CAN-Bus hardware and integration you can rely on at 1 Mbit/s and beyond, our engineering team is ready to help.
Contact our sales team