Definition: Sensors are devices that measure physical quantities such as temperature, pressure, vibration, position, force or light and convert them into electrical signals. Control systems, computers or IoT networks process these signals. They are the data basis for monitoring, analysis and automation.

In B2B mechanical engineering, sensors are essential for condition monitoring, quality assurance, predictive maintenance and digital twins. They supply real-time data about the machine condition, production quality and process parameters. This is the basis for data-driven decisions.

English: Sensors German: Sensoren, Messwertaufnehmer

Sensor Types in Mechanical Engineering

Temperature sensors

Measure the thermal energy of objects, components or environments. The most important types: - Thermocouples (type K, J, T): Robust industrial sensors for high temperatures - RTDs (Pt100, Pt1000): Precise resistance temperature detectors for high-accuracy applications - Infrared temperature sensors: Non-contact measurement for rotating parts

Use: Monitoring of motors, bearings, hydraulics, drives. Early detection of overheating, wear, cooling problems.

Pressure sensors

Measure the pressure of liquids and gases in hydraulic, pneumatic and process systems. Types: - Piezoelectric pressure sensors: Dynamic pressure changes - Piezoresistive sensors: Static and dynamic pressure - Capacitive pressure sensors: High-precision measurements

Use: Hydraulic presses, pneumatic controls, process monitoring, quality testing. Monitoring of system pressure, leak detection, diagnosis of pressure drops.

Vibration sensors (accelerometers)

Measure vibrations, accelerations and oscillations on machines. Technologies: - Piezoelectric accelerometers: High-frequency vibrations - MEMS sensors: Compact and low-cost - Eddy current sensors: Non-contact vibration measurement

Use: Predictive maintenance (bearing monitoring), imbalance detection, wear diagnosis. Vibration patterns show bearing damage, imbalance, loose parts and assembly problems.

Position sensors and displacement transducers

Measure position, displacement, distance or proximity. Types: - Inductive sensors: Metal objects up to ~50 mm - Capacitive sensors: All materials, short range - Ultrasonic sensors: Precise distance measurement up to several meters - Laser distance sensors: High-precision positioning (µm range) - LVDT (linear variable differential transformer): Precise displacement measurement

Use: Position monitoring of cylinders, tool changers, robots. Quality control (dimensional inspection), collision avoidance, precision positioning.

Force sensors (strain gauges, load cells)

Measure mechanical forces and loads. Applications: - Process monitoring (press force, screw force) - Quality assurance (joining inspection, assembly inspection) - Overload protection

Optical sensors and camera systems

Visual capture for quality control, object recognition, positioning: - 2D vision systems: Part inspection, OCR, barcodes - 3D camera systems: Dimensional inspection, robot guidance - LiDAR sensors: Spatial capture for navigation and mapping

Use: Automatic quality control, robot vision, part identification, creation of digital twins.

Sensors in Industrial Use

Machine monitoring (condition monitoring)

Sensors continuously monitor the machine condition: - Temperature at critical components (motors, bearings, gearboxes) - Vibrations for wear detection - Pressure in hydraulic and pneumatic systems - Energy consumption of individual units

Goal: Find problems early, before they cause failures. The data goes into IoT platforms, where it is visualized.

Predictive maintenance

AI algorithms analyze sensor data to predict failures: - Vibration patterns: Bearing failures can be detected 4-8 weeks in advance - Temperature trends: Detect overheating early - Pressure drops: Identify leaks and wear

Advantage: Maintenance is done when it is necessary, not at fixed intervals. This reduces downtime by up to 50% and extends the machine service life.

Quality assurance in real time

Sensors check product quality during manufacturing: - Force sensors: Joining inspection, press force control - Vision systems: Surface inspection, dimensional inspection - Position sensors: Positional accuracy of components - Temperature sensors: Process temperature monitoring

Result: 100% inspection instead of random samples. Immediate alarm when there are deviations. Less scrap. Documentation for traceability.

Process automation

Sensors control automated processes: - Position sensors for robots and handling systems - Pressure sensors for pneumatic controls - Vision systems for pick-and-place applications - Proximity sensors for safety functions

Digital twin integration

Sensor data synchronizes digital twins with physical machines: 1. Sensors capture real-time data (temperature, vibration, position, pressure) 2. The data goes through the IoT into the digital twin 3. 3D visualization shows the current condition 4. Analysis dashboards show trends and anomalies

Example: Data from temperature sensors is shown as a heatmap on the 3D model of the machine. The service technician sees critical hotspots immediately.

Sensor Properties and Requirements

Accuracy

The necessary accuracy depends on the application: - Quality inspection: High accuracy (µm range for dimensional inspection) - Process monitoring: Medium accuracy (temperature ±1°C is sufficient) - Presence detection: Low accuracy (simple binary sensors)

Measuring range and resolution

Sensors must be designed for the expected value ranges: - Temperature sensors: -40°C to +800°C (depending on the use) - Pressure sensors: 0-400 bar (hydraulics) - Accelerometers: ±2g to ±100g

Robustness and degree of protection

Industrial sensors must withstand harsh environments: - IP ratings: IP67/IP69K for dust, moisture, cleaning - Temperature resistance: Correct function at the ambient temperatures - Vibration resistance: Resistance to machine vibration - EMC immunity: Immunity to electromagnetic interference

Interfaces and connectivity

You must be able to integrate sensors into control systems and IoT networks: - Analog signals: 0-10V, 4-20mA (classic PLC connection) - Digital interfaces: IO-Link, EtherCAT, PROFINET, Modbus - Wireless connections: Bluetooth, Wi-Fi, LoRaWAN (for IoT)

Challenges

Calibration and drift

Sensors are subject to aging and drift. Regular calibration is necessary, especially for safety-critical and quality-relevant measurements.

Environmental effects

Temperature, moisture, vibration and electromagnetic fields can affect sensor signals. Sensors must be suitable for the operating environment, and interference must be compensated.

Data volume and data processing

High-frequency sensors generate large volumes of data. Edge computing (preprocessing at the sensor) and intelligent data reduction (send only relevant changes) are necessary.

Cost-benefit analysis

Not every machine needs a full set of sensors. The decision depends on: machine value, cost of failures, maintenance effort, quality requirements.

Frequently Asked Questions (FAQ)

Which sensors are the most important for predictive maintenance?

The "big 3" for predictive maintenance are: 1) vibration sensors (detect bearing and imbalance problems), 2) temperature sensors (detect overheating and wear), 3) current sensors (show higher energy consumption caused by problems). These three capture most wear patterns and make failure predictions possible 4-8 weeks in advance.

How often must you calibrate industrial sensors?

This depends on the sensor type, the application and the regulatory requirements. Temperature sensors: Every year for critical processes. Pressure sensors: Every 6-12 months for quality applications. Vibration sensors: Rarely, usually only a zero adjustment. Vision systems: After a changeover or when the quality drifts. Critical safety and quality sensors: As specified by the manufacturer and regulatory standards.

Can you add sensors to old machines later?

Yes, a retrofit is possible and often makes economic sense. You can install clamp-on sensors (for example vibration and temperature sensors) without changes to the machine. Wireless sensors reduce the cabling effort. Retrofit kits are available for common machine types. Costs: €2,000-10,000 per machine. For expensive systems, the ROI is often less than 12 months because failures are prevented.

How is sensor data visualized?

On several levels: 1) dashboards with KPIs and trends (web/mobile), 2) alarm systems for critical values, 3) 3D visualization on machine models (heatmaps, status indicators), 4) AR overlays directly on the real machine (the service technician sees sensor values on a tablet/AR headset), 5) digital twin with live synchronization of the physical machine.

How much do the sensors for a production machine cost?

This depends strongly on the scope. Basic monitoring (temperature, vibration, operating hours): €1,500-4,000. Full condition monitoring (10-20 sensors, gateway, cloud connection): €5,000-15,000. High-end with vision and precision sensors: €20,000-50,000. Running costs: cloud platform €50-200/month/machine. For machines with a value of more than €100,000, the investment usually pays for itself within 1-2 years.

Industry Standards and Technologies

IO-Link (IEC 61131-9) Standardized interface for intelligent sensors and actuators. It makes parameter setting and diagnostics possible. Source: https://io-link.com/

OPC UA Platform-independent communication standard for Industry 4.0 and IoT integration. Source: https://opcfoundation.org/

IEEE 1451 Standard for smart sensor interfaces. It defines Transducer Electronic Data Sheets (TEDS). Source: https://standards.ieee.org/

NAMUR NE 107 Guideline for self-monitoring and diagnostics of field devices in the process industry. Source: https://www.namur.net/

ISO 10816 Standard for vibration monitoring on machines. It defines limit values for vibration sensors. Source: https://www.iso.org/standard/50528.html

Sensor technologies: - Temperature measurement – thermal capture - Pressure measurement – hydraulics/pneumatics - Vibration analysis – vibration measurement - Vision systems – optical inspection - RFID/NFC – identification and tracking

Applications: - IoT – sensor networking and data capture - Digital twin – sensor data for digital twins - Predictive maintenance – maintenance before failures - Condition monitoring – monitoring of the machine condition - Quality assurance – quality control

Visualization: - 3D visualization – sensor data on 3D models - Augmented reality – AR overlay with sensor data - Dashboards – KPI visualization - Heatmaps – display of temperature and condition - Trend analyses – changes over time

Systems: - SCADA – Supervisory Control and Data Acquisition - MES – Manufacturing Execution System - Edge computing – local data processing - Cloud platforms – IoT data analysis - Machine learning – pattern recognition in sensor data