Intermittent Fault Capture Worksheet: Root Cause Analysis Toolkit

Section 1 of 5

Fault Definition Worksheet

Complete this section before you do anything else. The goal is to define the fault in precise, observable terms — not theories about the cause. If two technicians read this, they should agree on exactly what the problem is.

Equipment Name / Asset ID

Fault Description — In Observable Terms (what you see, hear, or measure — not a theory)

First Observed Date / Time

Frequency of Occurrence

Conditions When Fault OCCURS (temperature / load level / time of day / shift / operator / production rate)

Conditions When Fault Does NOT Occur

What Was Changed or Serviced Before First Occurrence?

Current Error Codes or Alarm History (list all, with timestamps if available)

Common Mistake

Fault descriptions like "machine acts up" or "runs rough sometimes" are not useful. Write what you would describe to someone who has never seen this machine. Specific examples: "Motor trips overload relay (15A, set at 18A) within 12 minutes of startup on days when ambient temp exceeds 85°F."

Section 2 of 5

Data Capture Plan

Decide in advance what you will monitor and how — before the next fault event. Fill in the Normal Range and Trigger Threshold columns for your specific equipment. The trigger threshold is the value that tells you the fault is imminent or active.

What to Monitor	Sensor / Method	Sample Rate
Supply Voltage	Multimeter / data logger on panel lugs	Continuous or every 1 min
Current Draw	Clamp meter / current logger on motor leads	Continuous or every 30 sec
Temperature	Thermocouple, IR thermometer, or thermal logger	Every 2–5 min
Vibration	Vibration sensor or handheld analyzer on bearing housing	Every 5 min or continuous
Cycle Time	Stopwatch, PLC timer register, or cycle counter	Per cycle
Input Signal States	PLC I/O status screen or I/O force table	At fault event
Output Signal States	PLC I/O status screen or field verification with meter	At fault event
Fluid Pressure	Gauge or pressure transducer with logger	Continuous or every 1 min

Before the Next Fault Window

Have your monitoring equipment installed and recording before the fault is likely to occur. Data logged after a fault event is useful; data captured during the event is essential. If the fault occurs at end of shift or early morning, consider leaving a data logger running overnight.

Section 3 of 5

Fault Event Log

Complete one row for each fault occurrence. The goal is to find the pattern — the condition, time, or operational state that is present every time the fault occurs and absent when it does not. One occurrence is an incident. Three occurrences are a dataset.

Date / Time	Duration	Operating Conditions at Fault	Data Values at Fault	What Changed Before?	Actions Taken	Did Fault Recur?

Pattern Analysis — After 3+ Events

Review the log for any condition that is present in every row. Common intermittent fault patterns: thermal (fault occurs after equipment reaches operating temperature), load-related (fault correlates with production rate or motor current), time-of-day (ambient, utility voltage, or personnel-related), operator-specific (fault frequency changes by shift or operator), or post-maintenance (fault starts after a service event and may indicate an installation issue).

Section 4 of 5

Hypothesis Tracking

Write down each credible theory about the root cause. For each one, document what evidence supports it, what evidence contradicts it, and what test would confirm or deny it. This prevents you from committing to a cause before you have evidence — and keeps the team aligned on what has been ruled out.

Hypothesis	Evidence For	Evidence Against	Test to Confirm / Deny	Result

Key Rule

A hypothesis is not confirmed until the fault stops after corrective action. Correlation is not confirmation — the fact that the fault occurred when voltage was low does not prove voltage is the cause. The test is: did the fault stop after you corrected the voltage? And has it stayed stopped long enough to be confident?

Section 5 of 5

Root Cause Summary

Complete this section only after the fault has been corrected and verified. This is the permanent record — the document that prevents the next technician from starting from scratch six months from now.

Confirmed Root Cause

Evidence That Confirms It (what changed, what was observed, what test result proved it)

Corrective Action Taken (be specific — parts replaced, settings changed, procedures updated)

Verification Method (how did you confirm the fault is resolved — runtime duration, test conditions, load test)

Date Closed

Closed By

File This With the Asset Record

A completed fault capture worksheet is one of the most valuable documents in your maintenance program. It tells the next technician what was tried, what was ruled out, and what actually worked. Store it with the equipment folder or attach it to the closed work order in your CMMS.

Next Step — Build Prevention Into the Schedule

Preventive Maintenance Playbook

This worksheet helps you solve the fault in front of you. The PM Playbook helps you prevent the next one. It includes a complete task-building framework, frequency selection logic, failure mode documentation, and a 90-day implementation tracker — everything you need to turn reactive findings into proactive maintenance tasks.

Learn more about the PM Playbook ($99) →