Module 9 - Compass
Compass Errors in Practice
What you will learn
Monitor a compass with independent references, diagnose conflicting heading data and choose a safe response when magnetic or electronic heading evidence becomes unreliable.
Compass reliability is an ongoing navigation task. A check should use a known direction and controlled assumptions, then lead to recorded evidence, restricted reliance or competent investigation when results disagree.
Use a reliable known direction whenever possible: a charted transit, a properly calculated celestial bearing at the relevant level, or another approved reference. Record the compass reading, reference, heading, time and conditions. MCA guidance says regulated ships should determine errors after large alterations of course and monitor them regularly; a yacht navigator should apply the same principle proportionately and follow the vessel's procedures.
COG can screen for a problem only when the vessel holds a steady heading and speed and cross-current, leeway, manoeuvring, GNSS smoothing and sensor alignment are negligible or quantified. It is not proof on its own. A persistent discrepancy demands independent checks and an assessment of which source is unreliable.
Heel can alter deviation because the compass and vessel magnetism change relative geometry. In conventional compass analysis, heeling error is most evident on north/south headings. Do not assume an upright deviation card describes every sailing condition; record the condition and seek competent assessment if heel-related error is material.
A fluxgate is an electronic magnetic sensor: it still needs correct installation and calibration and remains vulnerable to magnetic interference, power and data faults. Some equipment stores its own deviation corrections. A conventional gyrocompass derives true north without magnetic variation or deviation but has different alignment, latitude, speed, power and maintenance considerations and is uncommon on small yachts. GNSS heading sensors are a separate technology and must not be called gyrocompasses merely because they output true heading.
| Observation | Likely issue to test | Safe response |
|---|---|---|
| Steering compass changes near a new speaker | Local magnetic interference | Remove the influence; restrict reliance; arrange a check |
| Compass and COG differ in cross-tide | Heading is being compared with ground track | Quantify tide/leeway or use a known bearing reference |
| Fluxgate and magnetic compass disagree | Calibration, alignment, interference, power or data fault | Cross-check independently; do not choose the preferred display |
Manage conflicting heading sources
At night the fluxgate display and steering compass differ by 8°, while COG varies in a strong cross-tide near a headland.
1. Do not vote between displays
Two instruments can share interference, alignment or input faults, and COG is not heading in cross-tide.
2. Protect navigation
Increase margins, slow or hold in safe water if appropriate, and use charted visual, radar, depth and position evidence.
3. Record and isolate
Note headings and conditions, remove portable interference, check power/data/alignment and arrange competent inspection before relying on the unexplained source.
Sense check: Selecting the instrument closest to the desired route is confirmation bias, not a compass check.
Common mistake or limitation
- A steady COG does not prove a correct compass heading when current or leeway can create an equally steady offset.
- Electronic heading is not automatically true or error-free: identify whether the source is magnetic, gyro or GNSS and apply its documented corrections and limits.
Recap
- Use independent known directions and record the conditions of every meaningful compass check.
- When evidence conflicts, increase margins and identify the failure mode before restoring reliance.
- Treat heel, equipment changes and electronic calibration as configuration evidence, not footnotes.
Optional quick check
Section 7 of 7
In a strong cross-tide, why is a difference between compass heading and GNSS COG not enough to prove compass error?
Sources and factual review
Reviewed 20 August 2026 - Compass Revision editorial review.
Continue studying Compass
This public lesson is part of Revision Module 9. Full revision access adds the guided Learn, Practise and Test flow, flashcards and revision progress.