Vibration‑based condition monitoring generates large volumes of high‑frequency data. Managing that data efficiently – without sacrificing diagnostic value – is a challenge, especially in distributed or bandwidth‑limited systems.
CAN‑MD diagnostics solve this by processing CAN‑bus vibration data directly at the sensor and transmitting only meaningful results over a CAN network.
As digital monitoring sensors, CAN‑MD devices reduce data volume while enabling scalable, reliable condition monitoring across complex systems.
Traditional vibration monitoring systems rely on centralised data acquisition and continuous streaming of raw signals. As system size increases, this approach quickly becomes constrained by bandwidth, wiring complexity, and post‑processing effort.
With CAN‑MD accelerometers, data acquisition, digital signal processing, and vibration CI analytics are embedded inside each sensor. Instead of raw time‑domain data, sensors output compact condition indicators (CIs) – such as RMS levels, spectral band energy, or order‑tracked features – over a standard CAN 2.0B network.
This on‑sensor processing dramatically reduces data volume while preserving the information required for reliable machine diagnostics.
CAN‑MD enables distributed vibration diagnostics by combining data acquisition, embedded signal processing, and condition indicator extraction directly within each sensor.
By processing signals locally and transmitting only relevant results over a deterministic CAN network, it reduces data load, simplifies system architecture, and enables scalable, real‑time condition monitoring.
Reduce vibration data at the source
High‑rate vibration signals are processed inside each sensor using time‑domain analysis, FFT‑based spectral processing, and speed‑synchronous techniques.
Only application‑relevant results are transmitted, making continuous monitoring practical on a CAN bus.
Scale Your System Effortlessly
Each CAN‑MD sensor operates as an intelligent, addressable node.
Additional sensors can be added without changing the overall architecture, supporting scalable bus‑based machinery diagnostics from small test setups to full systems.
Cut Cabling Complexity and Costs
A shared CAN bus replaces point‑to‑point analog wiring, reducing cabling, weight, and installation effort.
Embedded configuration and unique sensor IDs improve traceability and reduce commissioning time.
Get actionable diagnostics faster
By delivering edge‑processed vibration data, CAN‑MD provides condition indicators that are ready for trending, alarms, and decision‑making – without extensive post‑processing.
Designed for demanding applications
CAN‑MD smart sensor diagnostics are proven in applications where reliability, scalability, and deterministic communication are essential, including:
CAN‑MD is also well suited for converting legacy analog systems to all‑digital, CAN‑based condition monitoring, including Industrial Internet of Things (IIoT) architectures.
Edge intelligence inside every sensor
Each CAN‑MD sensor functions as an autonomous diagnostic node, combining:
Raw time histories or spectra can be accessed on demand for setup or detailed analysis, but continuous raw data streaming is not required for effective condition monitoring.
CAN-MD is an advanced distributed sensing network that uses a CAN bus architecture to connect up to 31 intelligent vibration sensors, each with its own processing capability and unique Node ID.
Instead of sending large volumes of raw vibration data, each sensor analyzes data locally and transmits only meaningful condition indicators, dramatically reducing bandwidth and system complexity. This decentralized approach eliminates the need for a central processor, simplifies wiring, and enables scalable, intelligent installations.
With built-in configurability, real-time diagnostics, and seamless integration with existing CAN-based systems, CAN-MD delivers a smarter, lighter, and more efficient solution for modern machinery monitoring.
CAN‑MD smart sensors combine vibration sensing, signal processing and CAN bus communication in a single device. By delivering actionable condition data directly at the sensor, they reduce system complexity and support more efficient, reliable machine health monitoring.
Download our white paper to explore the technology and learn how to build a smarter condition-based maintenance strategy.
The Réseau express métropolitain (REM) is the world’s longest automated light rail network, serving the Greater Montréal area. To meet strict safety requirements in tunnels and stations, REM needed a reliable way to monitor smoke evacuation fans. This case study shows how CAN‑MD smart sensors enable proactive vibration monitoring to improve safety, reliability and operational efficiency.
In this video, Brian Johnson, Applications Development Supervisor at Dytran by HBK, introduces CAN‑MD smart sensor technology. He outlines the background and capabilities of CAN‑MD, explains key applications and benefits, and provides an overview of the available developer kits, software and product line developments.
Watch the video to see how CAN‑MD enables a smarter approach to machine health monitoring.
A: CAN-MD reduces data at the source by converting high-rate time-domain vibration signals into compact diagnostic features inside each sensor. Instead of continuously transmitting raw waveform data, which is impractical over a CAN bus due to bandwidth limitations, each sensor performs onboard digital signal processing to generate condition indicators (CIs) such as RMS, spectral energy, or order-based features. Only these condensed, application-relevant results are broadcast on the CAN network. Raw time histories or spectra are available on demand for setup or deeper analysis but are not required for continuous monitoring. This approach transforms “terabytes of raw data” into “kilobytes of actionable data,” enabling efficient trending, storage, and real-time diagnostics.
A: Yes. CAN-MD is inherently designed as a scalable, distributed architecture. Each sensor operates as an independent node on the CAN bus, allowing additional sensors to be added without redesigning the overall system. A single CAN network supports up to 31 CAN-MD nodes, each with a unique Node ID for identification and data attribution. Multiple CAN networks can be deployed enabling expansion from small test setups to full vehicle or plant-wide monitoring systems.
A: Yes. CAN-MD communicates over standard CAN 2.0B, using conventional CAN frame structures and broadcast messaging. Each sensor transmits data tagged with a unique Node ID, and all devices on the network can receive and filter messages based on relevance. This allows CAN-MD to integrate with common CAN infrastructure such as vehicle networks, industrial controllers, data loggers, and CAN-to-USB or CAN-to-ethernet gateways.
A: CAN-MD is best suited for applications where distributed, scalable vibration diagnostics are required and where continuous raw data streaming is impractical. It is commonly applied in aerospace health and usage monitoring systems (HUMS) and onboard condition monitoring systems, as well as in rotating machinery diagnostics for components such as gearboxes, motors, and bearings. The platform is also well aligned with mobile and autonomous systems including off-highway equipment, electric vehicles and rail, along with industrial machinery and factory automation environments. It is particularly effective in scenarios where wiring complexity, system weight, and data bandwidth are constrained, and where there is a need for machines to provide self-diagnosing capabilities through embedded, sensor-level analytics.
A: A conventional analog sensor paired with a centralised high-speed data acquisition (DAQ) system is more appropriate in applications where full, continuous access to raw vibration waveforms is required. This is particularly true for high-resolution modal analysis or research testing, where complete time histories are necessary, as well as for detecting rare, non-repetitive transient events that may be missed with reduced data approaches. It is also preferred in advanced signal processing applications that rely on uninterrupted waveform capture, or in scenarios where system constraints make bus latency or bandwidth limitations unacceptable.
CAN-MD, by contrast, is optimised for condition monitoring and diagnostics rather than continuous raw data streaming. While it can capture and transmit raw time-domain or spectral data when needed, its primary strength lies in reducing, processing, and structuring vibration data at the source into meaningful condition indicators, enabling efficient monitoring without the complexity and data burden of traditional high-end laboratory DAQ systems.
A: Yes. CAN-MD supports synchronous measurements through integration with a tachometer input, enabling tachometer speed-correlated analysis directly at the sensor. Using tachometer signals, sensors can compute order-domain condition indicators, perform time synchronous averaging (TSA), and extract amplitude and phase information associated with rotating components. This allows advanced diagnostics such as imbalance detection, gear mesh analysis, and fault tracking relative to shaft speed to be performed at the edge without requiring centralised processing.
A: Yes. CAN-MD supports integration of external sensors through adapter modules, including the 4760A CAN-MD® IEPE Adapter and the 4785A CAN-MD Single-Ended Inline Charge Amplifier. The 4760A enables traditional IEPE sensors or charge-mode accelerometers, when used with external inline charge amplifiers, to interface with the CAN-MD network, effectively digitising and processing their signals within the distributed architecture. The 4785A extends this capability by combining charge conversion and CAN-MD signal processing into a single integrated device, allowing direct connection of charge-mode accelerometers without the need for separate conditioning hardware. Together, these solutions allow legacy and high-temperature sensors to participate fully in the CAN-MD ecosystem, producing the same condition indicators and diagnostic outputs as native CAN-MD sensors. This provides a seamless migration path from conventional analog sensing approaches to distributed digital diagnostics, while preserving the ability to deploy specialised sensors in extreme or high-temperature environments.
This will bring together HBM, Brüel & Kjær, nCode, ReliaSoft, MicroStrain and Discom brands, helping you innovate faster for a cleaner, healthier, and more productive world.
