Inferensys

Glossary

Disturbance Recorder

A function within an Intelligent Electronic Device (IED) that captures high-resolution analog and binary signal waveforms during a power system fault, storing them in the COMTRADE format for post-mortem analysis and fault location.
Modern WeWork hardware lab area with product team collaborating around AI device prototypes, 3D printer in background, dramatic industrial lighting with product sketches on glass walls.
FAULT WAVEFORM CAPTURE

What is a Disturbance Recorder?

A disturbance recorder is a specialized function within an Intelligent Electronic Device (IED) that captures high-resolution, time-synchronized analog and binary signal waveforms during power system anomalies for post-mortem analysis.

A disturbance recorder is an automated data capture function embedded in a substation Intelligent Electronic Device (IED) that continuously samples voltage and current waveforms into a circular buffer. When a trigger condition—such as a protection trip, undervoltage threshold, or binary status change—is detected, the recorder freezes the buffer, storing pre-fault, fault, and post-fault data in the standardized COMTRADE (IEEE C37.111) format for offline analysis.

The recorded files contain time-stamped instantaneous values from instrument transformers and status changes from circuit breakers and relays, enabling protection engineers to perform precise fault location, verify protection scheme operation, and analyze transient stability events. Modern disturbance recorders leverage Precision Time Protocol (PTP) or GPS synchronization to align waveforms across multiple IEDs, creating a system-wide, microsecond-accurate view of a cascading grid disturbance.

FAULT ANALYSIS FUNDAMENTALS

Core Characteristics of a Disturbance Recorder

A disturbance recorder is not merely a data logger; it is a high-resolution forensic tool embedded within an Intelligent Electronic Device (IED) that captures the transient electrical signatures of power system faults for precise post-mortem analysis.

01

High-Resolution Waveform Capture

The primary function is the synchronous recording of analog current and voltage waveforms at high sampling rates, typically 1 kHz to 15 kHz. This granularity captures the high-frequency transient phenomena and DC offset decay that occur during the first few cycles of a fault, which are invisible to standard SCADA polling. The recorder stores pre-fault, fault, and post-fault data to provide a complete picture of the event evolution.

02

Binary Signal & Sequence of Events

Alongside analog data, the recorder captures the state changes of binary inputs and outputs with microsecond timestamp resolution. This includes:

  • Protection element pickups and trips
  • Circuit breaker auxiliary contacts (52a/52b)
  • Communication-assisted trip signals (POTT, DTT) This creates a precise Sequence of Events (SOE) log, allowing engineers to verify that the protection scheme operated correctly and within its designed clearing time.
03

COMTRADE File Standardization

Disturbance recorders store data in the IEEE C37.111 (COMTRADE) format, a universal standard ensuring interoperability between different manufacturers' IEDs and analysis software. A COMTRADE record consists of:

  • .CFG: Configuration file defining channel scaling and sampling rate
  • .DAT: Binary or ASCII data file containing the sample values This standardization is critical for fault location calculations and protection coordination studies across a heterogeneous substation asset base.
04

Precision Time Synchronization

Accurate fault analysis requires all recordings across a substation or wide-area network to be aligned to a common time reference. Disturbance recorders utilize IRIG-B or IEEE 1588 (PTP) to timestamp samples with absolute time accuracy down to 1 microsecond. This allows engineers to overlay recordings from different bays to analyze traveling wave phenomena and precisely locate faults on transmission lines using double-ended methods.

05

Triggering & Auto-Recloser Integration

Recording is initiated by sophisticated triggers, not just a simple threshold crossing. Triggers include:

  • Protection element start/trip signals
  • Rate-of-change-of-frequency (ROCOF)
  • Analog magnitude violation The recorder is tightly integrated with the auto-recloser logic, capturing the complete sequence of a transient fault, the dead time, and the subsequent successful or unsuccessful reclose attempt, which is vital for distinguishing transient from permanent faults.
06

Fault Location Calculation

Modern disturbance recorders perform embedded impedance-based fault location using the captured voltage and current phasors. By applying the Takagi algorithm or similar methods, the IED calculates the reactance to the fault point, compensating for load current and arc resistance. The result is a distance-to-fault estimate (e.g., 12.3 miles from the substation) that is appended to the event record, drastically reducing patrol time for line crews.

DISTURBANCE RECORDER INSIGHTS

Frequently Asked Questions

Clear, technical answers to the most common questions about disturbance recorder functionality, COMTRADE file structure, and fault analysis workflows.

A disturbance recorder is a dedicated function within an Intelligent Electronic Device (IED) that continuously captures high-resolution analog and binary signal waveforms during power system anomalies. It operates as a circular buffer, constantly sampling voltage and current channels at rates typically between 1 kHz and 14.4 kHz. When a trigger condition is met—such as a protection element pickup, a binary input change, or an under/over voltage threshold—the recorder freezes a configurable window of pre-fault, fault, and post-fault data. This captured record is then time-stamped using a common time source like GPS or PTP and stored in non-volatile memory in the COMTRADE (Common Format for Transient Data Exchange) format, preserving the precise sequence of events for post-mortem analysis.

Prasad Kumkar

About the author

Prasad Kumkar

CEO & MD, Inference Systems

Prasad Kumkar is the CEO & MD of Inference Systems and writes about AI systems architecture, LLM infrastructure, model serving, evaluation, and production deployment. Over 5+ years, he has worked across computer vision models, L5 autonomous vehicle systems, and LLM research, with a focus on taking complex AI ideas into real-world engineering systems.

His work and writing cover AI systems, large language models, AI agents, multimodal systems, autonomous systems, inference optimization, RAG, evaluation, and production AI engineering.