Event Logger & Data Logger in Railway Signalling — Purpose, Architecture, and Implementation
Event Logger vs Data Logger explained — architecture, implementation approaches (standalone, integrated, IoT-based), point machine current signature analysis, and best practices for railway signalling forensics and predictive maintenance.
What Are Event Loggers and Data Loggers?
In railway signalling, two types of recording systems capture critical operational data:
Event Logger (ELD — Event Logging Device)
Records discrete events — state changes in signalling equipment:
- Signal aspect changes (Red → Green → Yellow → Red)
- Point machine movements (Normal → Reverse)
- Track circuit state changes (Clear → Occupied → Clear)
- Route setting and cancellation
- Level crossing barrier movements
- Power supply alarms
- Communication link failures
Data Logger
Records continuous or sampled data — analog and digital measurements over time:
- Point machine current waveforms during operation
- Track circuit voltage levels
- Signal lamp current readings
- Power supply voltage and frequency
- Temperature and environmental readings
- Communication link quality metrics
Key Difference
| Feature | Event Logger | Data Logger | |---------|-------------|-------------| | Data Type | Discrete events (state changes) | Continuous/sampled measurements | | Storage | Timestamped event records | Time-series data | | Trigger | State change occurs | Periodic sampling or threshold | | Volume | Moderate (events per day) | High (samples per second) | | Primary Use | Incident investigation, audit | Predictive maintenance, analysis |
Why Are They Important?
1. Incident Investigation (Forensics)
When an incident occurs, event logs provide a precise timeline:
Timestamp Event Equipment State
─────────────────────────────────────────────────────────────────
2026-02-18 14:23:01.234 Track Circuit TC-05 Clear → Occupied
2026-02-18 14:23:01.456 Signal S3 Green → Red
2026-02-18 14:23:02.100 Point P2 Normal (locked)
2026-02-18 14:23:05.789 Track Circuit TC-06 Clear → Occupied
2026-02-18 14:23:15.234 Track Circuit TC-05 Occupied → Clear
This precise timestamped data is critical for understanding what happened and why.
2. Preventive & Predictive Maintenance
Data loggers capture equipment health trends:
Point Machine P7 — Current Waveform Trend:
Jan: Peak 2.1A, Duration 3.2s ✓ Normal
Feb: Peak 2.4A, Duration 3.5s ⚠ Increasing
Mar: Peak 2.8A, Duration 4.1s ⚠ Warning
Apr: Peak 3.5A, Duration 5.2s ✗ Maintenance required
By tracking trends, maintenance teams can schedule repairs before equipment fails in service.
3. Performance Analysis
- Train punctuality correlation with signalling equipment performance
- Track circuit failure frequency and duration analysis
- Point machine operation statistics (cycles, failures, maintenance history)
4. Safety Compliance
Regulatory bodies require signalling equipment state to be recorded for:
- Safety audit trails
- Compliance with EN 50129, EN 50126 standards
- Evidence for safety case documentation
Event Logger Architecture
┌─────────────────────────────────────────────────────┐
│ FIELD EQUIPMENT │
│ Signals, Points, Track Circuits, Level Crossings │
└───────────────────────┬─────────────────────────────┘
│ (Relay contacts / Digital I/O)
▼
┌─────────────────────────────────────────────────────┐
│ INPUT ACQUISITION MODULE │
│ - Scans digital inputs at high frequency (10ms) │
│ - Detects state changes │
│ - Timestamps each event (GPS-synced clock) │
└───────────────────────┬─────────────────────────────┘
│
▼
┌─────────────────────────────────────────────────────┐
│ EVENT PROCESSING ENGINE │
│ - Filters duplicate/bouncing events │
│ - Enriches with equipment descriptions │
│ - Sequences events in chronological order │
│ - Applies alarm rules │
└───────────────────────┬─────────────────────────────┘
│
┌───────┼───────┐
▼ ▼ ▼
[Local Store] [Remote] [Display]
(SD/HDD) (Server) (HMI)
Key Requirements
- Time accuracy — GPS-synchronized clock, microsecond resolution
- No data loss — Circular buffer with overflow protection
- High scan rate — 10ms or better for detecting fast events
- Contact bounce filtering — Debounce logic to prevent false events
- Storage capacity — Minimum 30-90 days of local storage
- Tamper-proof — Logged data cannot be modified after recording
Data Logger Architecture
┌─────────────────────────────────────────────────────┐
│ FIELD EQUIPMENT │
│ Point machines, Track circuits, Power supplies │
└───────────────────────┬─────────────────────────────┘
│ (Analog signals: current, voltage)
▼
┌─────────────────────────────────────────────────────┐
│ ANALOG-TO-DIGITAL CONVERSION │
│ - ADC channels (12-16 bit resolution) │
│ - Sampling rate: 100 Hz - 10 kHz per channel │
│ - Anti-aliasing filters │
│ - Current transformers / voltage dividers │
└───────────────────────┬─────────────────────────────┘
│
▼
┌─────────────────────────────────────────────────────┐
│ DATA PROCESSING & STORAGE │
│ - Trigger-based recording (event-driven) │
│ - Continuous recording (time-based) │
│ - Data compression and aggregation │
│ - Local storage (weeks to months) │
└───────────────────────┬─────────────────────────────┘
│
▼
┌─────────────────────────────────────────────────────┐
│ COMMUNICATION & UPLOAD │
│ - Ethernet / 4G LTE to central server │
│ - MQTT / HTTPS data transfer │
│ - Scheduled bulk upload or real-time streaming │
└─────────────────────────────────────────────────────┘
Point Machine Current Signature Logging
One of the most valuable data logger applications:
Current (A)
4 │ ╭──╮
│ ╱ ╲
3 │ ╱ ╲
│ ╱ ╲──────────╮
2 │ ╱ ╲
│ ╱ ╲
1 │ ╱ ╲
│───╱ ╲───
0 └─────────────────────────────────── Time (s)
0 1 2 3 4 5 6 7
└──┘ └────┘ └──────────┘ └────┘ └──┘
Start Unlock Traverse Lock Complete
The shape of this waveform reveals the health of the point machine. Changes in peak current, duration, or waveform shape indicate developing faults.
Implementation Approaches
Standalone Event Logger
Dedicated hardware installed at each signalling location:
- Self-contained unit with I/O cards
- Local storage and optional remote upload
- GPS clock input
- Typical: 64-256 digital input channels
- Examples: Amsler ELD, Westermo data loggers
Integrated with Electronic Interlocking
Modern EI systems have built-in event logging:
- Events recorded directly by the vital computer
- Higher accuracy (events logged at source)
- No additional hardware needed
- Data extracted via maintenance interface
IoT-Based Data Logger
Using modern IoT platforms for data logging:
[Current Sensor] → [ADC Module] → [IoT Gateway] → [MQTT Broker] → [ThingsBoard/InfluxDB]
(Raspberry Pi / │
Industrial PC) [Grafana Dashboard]
Benefits:
- Flexible and extensible
- Cloud or on-premise deployment
- Rich visualization and analytics
- Lower cost than proprietary solutions
Data Analysis and Visualization
Event Log Analysis
-- Find all point failures in the last 30 days
SELECT timestamp, equipment_id, event_type, state
FROM event_log
WHERE equipment_type = 'POINT_MACHINE'
AND event_type = 'FAILURE'
AND timestamp > NOW() - INTERVAL '30 days'
ORDER BY timestamp DESC;Trend Analysis with Data Logger
Using time-series databases (InfluxDB, TimescaleDB) and visualization tools (Grafana):
- Plot point machine current waveforms over months
- Detect gradual degradation trends
- Set threshold alerts for abnormal readings
- Compare current readings against known-good baselines
Best Practices
- GPS time synchronization — All loggers must be time-synced for event correlation
- Redundant storage — Local + remote storage to prevent data loss
- Regular data backup — Archive old data for long-term trend analysis
- Access control — Logged data must be tamper-proof for forensic validity
- Alerting — Configure real-time alerts for critical events
- Data retention policy — Define how long data is kept (regulatory requirements vary)
Frequently Asked Questions
What is the difference between an event logger and a data logger in railway signalling?
An event logger (ELD) records discrete state changes — such as signal aspect transitions, point machine movements, and track circuit occupancy changes — with precise timestamps. A data logger records continuous or sampled analog measurements over time, such as point machine current waveforms, track circuit voltage levels, and power supply readings. Event loggers are primarily used for incident investigation and audit trails, while data loggers are used for predictive maintenance and trend analysis. Many modern systems combine both functions.
Why is GPS time synchronization important for event loggers?
GPS time synchronization ensures all event loggers across a railway network record events with the same precise clock reference, typically with microsecond accuracy. This is critical during incident investigations where the exact sequence and timing of events across multiple signalling locations determines what happened. Without GPS sync, comparing event logs from different stations would be unreliable due to clock drift, making forensic analysis impossible. Regulatory standards like EN 50129 require accurate time-stamped records.
What is a point machine current signature and why is it logged?
A point machine current signature is the waveform of electrical current drawn by a point machine (railway switch motor) during operation. It shows distinct phases — start, unlock, traverse, lock, and completion — each with characteristic current levels and durations. By logging this waveform over months, maintenance teams can detect gradual degradation: increasing peak current, longer traversal times, or abnormal waveform shapes indicate developing mechanical faults. This enables predictive maintenance — scheduling repairs before the point machine fails in service and causes train delays.
How long must railway signalling event logs be retained?
Retention periods vary by railway authority and regulatory framework, but typical requirements range from 30 days to 5 years. Most railways require a minimum of 90 days of local storage at the signalling location, with longer-term archival on central servers. Safety-critical event data related to incidents may need to be retained indefinitely for legal and regulatory purposes. EN 50126 and EN 50129 standards require that safety-relevant data be maintained for the operational lifetime of the system plus a defined period.
Can IoT-based data loggers replace traditional railway event loggers?
IoT-based data loggers can complement traditional event loggers but do not fully replace them in safety-critical applications. Traditional ELDs are designed to meet stringent railway safety standards (SIL 2/SIL 4) with tamper-proof storage, redundant recording, and certified hardware. IoT solutions excel at adding monitoring capabilities — tracking environmental conditions, equipment health trends, and enabling cloud-based analytics — but they typically operate as non-vital overlay systems. The trend is hybrid architectures: certified ELDs for safety-critical event recording, plus IoT sensors for expanded condition monitoring and predictive maintenance.
Conclusion
Event Loggers and Data Loggers are essential tools for modern railway signalling. They provide the visibility needed for incident investigation, predictive maintenance, and safety compliance. As IoT technology matures, data logging is becoming more accessible and powerful, enabling railways to transition from reactive to proactive maintenance.
Related: IoT-Based Predictive Maintenance for Railway Signalling and Point Machine Working Principle.