📊 DAY 14: PATCH TEST (PART 2) – PROCESSING & APPLYING CORRECTIONS

From Raw Data to Calibrated MBES – Roll, Pitch, Yaw, and Latency Processing
Instructor: Engr. Rokib Hossain | River Engineering Solutions
🏠 Course Homepage

📋 Processing Overview

Now that you have collected the patch test lines (Day 13), it's time to process them. The goal is to compute the residual angular offsets (roll, pitch, yaw) and latency. Most MBES acquisition software includes a dedicated Patch Test / Calibration module (e.g., Hypack Patch Test, QINSy Calibration, Teledyne PDS Calibration). The general workflow is:

  1. Load the raw data files (lines collected for roll, pitch, yaw, latency).
  2. Select the appropriate algorithm for each parameter.
  3. Compute the correction values.
  4. Apply the corrections to the system configuration.
  5. Re‑process a small dataset to verify improvement.

We will cover each parameter individually, followed by a cross‑line validation check.

📌 Key concept: Processing order matters. Typically, you process roll first, then pitch, then yaw, then latency – because yaw and latency corrections depend on accurate roll and pitch.

📏 Roll Offset Processing

Data used: The roll calibration line (sailed over a slope perpendicular to the line).

Processing steps:

  1. Open the roll line in the patch test module.
  2. Select the "Roll" calibration type.
  3. The software will compare the port and starboard beam depths across the slope.
  4. It will compute the angular offset that makes the two sides match.
  5. The result is a roll correction value (e.g., +0.15°).

Interpretation: A positive roll means the starboard side is deeper than the port side; you need to apply a correction to level the swath.

Before and after roll correction – swath becomes level
Figure 1: Roll correction – left (raw) shows asymmetric swath; right (corrected) shows level swath.
✅ Pro tip: If the computed roll offset is >0.5°, re‑check your mechanical alignment. Large values may indicate a physical misalignment that should be corrected mechanically rather than purely in software.

📐 Pitch Offset Processing

Data used: Two lines in opposite directions over a slope parallel to the lines.

Processing steps:

  1. Load both pitch lines (Line A and Line B, opposite directions).
  2. Select the "Pitch" calibration type.
  3. The software overlays the two profiles.
  4. If pitch offset exists, the two profiles will have a systematic depth difference.
  5. The software computes the pitch correction that makes the profiles coincide.

Interpretation: A positive pitch means the bow is deeper than the stern; you need to apply a correction to align the fore‑aft axis.

Pitch patch test overlay – before and after correction
Figure 2: Pitch correction – the two opposite profiles are overlaid; the offset is computed from the depth difference.

🧭 Yaw Offset Processing

Data used: Two lines in opposite directions passing over a small isolated target.

Processing steps:

  1. Load both yaw lines.
  2. Select the "Yaw" or "Heading" calibration type.
  3. Manually or automatically pick the target in each line.
  4. The software measures the horizontal shift between the target positions.
  5. From the shift and the water depth, it computes the yaw offset.

Interpretation: A positive yaw means the transducer is pointing slightly to starboard relative to the vessel's heading. The correction rotates the swath to align the target.

Yaw patch test – target shift in opposite lines
Figure 3: Yaw correction – the same target appears shifted in opposite lines; the shift is used to compute yaw offset.
🎯 Critical: Yaw error is the most important to fix because it affects horizontal positioning of all beams. After applying yaw correction, the target should appear at the same coordinate in both lines.

⏱️ Latency Processing

Data used: Lines over a target at two different speeds (or the same target as yaw).

Processing steps:

  1. Load the latency lines (e.g., 4 knots and 8 knots).
  2. Select the "Latency" calibration type.
  3. Pick the target in each line.
  4. The software computes the along‑track shift between the two speeds.
  5. From the shift and speed difference, it calculates the time delay (latency).

Interpretation: A positive latency means the sonar ping is delayed relative to the GNSS time stamp. The correction shifts the sonar data forward in time.

Latency patch test – target shift at different speeds
Figure 4: Latency correction – the target shifts along‑track at different speeds; the shift is used to compute time delay.

💻 Applying Corrections in Software

Once you have the correction values (roll, pitch, yaw, latency), you need to enter them into your acquisition software. The process varies by software:

Software Where to Enter Corrections Notes
Hypack Settings → MBES Configuration → Alignment tab Enter roll, pitch, yaw offsets. Latency is usually entered in the "Timing" section.
QINSy Sensor Configuration → MBES → Alignment QINSy uses a "Calibration" file that can be updated.
Teledyne PDS System Setup → MBES → Offsets & Alignment Enter values directly; PDS applies them in real time.
Qimera (post‑processing) Vessel Configuration → Sensor Offsets Corrections are applied during data import.
Hypack patch test processing interface screenshot
Figure 5: Hypack Patch Test processing interface – where you load lines and compute corrections.
💻 Important: After entering the corrections, save your configuration and restart the acquisition software to ensure the new values are loaded.

✅ Validation – Cross-line Analysis

After applying the corrections, you must validate that the patch test was successful. The primary validation method is cross‑line analysis.

Procedure:

  1. Run a few cross‑lines (perpendicular to the main survey lines) over a flat area.
  2. In post‑processing, compare the depths from the main lines and cross‑lines at their intersections.
  3. If the patch test corrections are correct, the depth differences at intersections should be within the survey's uncertainty budget (e.g., <0.1 m for Order 1a).
  4. If systematic differences remain, you may need to iterate the patch test (re‑process with refined values).

Most software (Hypack, QINSy, PDS) includes a cross‑line analysis tool that automatically computes statistics.

Cross-line validation – depth differences at intersections
Figure 6: Cross‑line validation – depth differences at intersections should be within tolerance.
✅ Pro tip: If cross‑line differences are systematic (e.g., all intersections show a constant offset), you may have a remaining vertical offset (draft or tide error) rather than an angular error.

🔗 External Resources & Tools

Refer to these official guides for detailed processing steps:

Resource Description Link
Hypack Patch Test ManualStep‑by‑step for collecting and processing patch test linesHypack Help
Teledyne RESON Patch Test GuideDetailed PDF with line patterns and analysisTeledyne RESON
QINSy Calibration ModuleCalibration manual for roll, pitch, yaw, latencyQPS QINSy
NOAA Patch Test ProceduresField procedures for MBES calibrationNOAA PDF (Chapter 6)
🔗 Authority linking: Hypack, Teledyne, QPS, and NOAA are trusted references.

📋 Homework & Fieldwork Checklists

📘 Day 14 – Homework Checklist

  • ✅ Read the entire Day 14 post.
  • ✅ Watch a video tutorial on processing patch test data in your preferred software.
  • ✅ If you have access to patch test data, practice loading the lines and computing corrections.
  • ✅ Write a short explanation of why cross‑line analysis is essential after applying corrections.
  • ✅ Share this post with #Hydrography #PatchTest.

🛠️ Fieldwork Readiness Checklist – Patch Test Processing

  • ✅ Ensure you have backed up all raw patch test data before processing.
  • ✅ Load the roll line and compute roll correction. Record the value and apply it.
  • ✅ Load the pitch lines and compute pitch correction. Record and apply.
  • ✅ Load the yaw lines and compute yaw correction. Record and apply.
  • ✅ Load the latency lines and compute latency correction. Record and apply.
  • ✅ After applying all corrections, run a cross‑line analysis on a small test area.
  • ✅ If cross‑line differences are within tolerance, finalise the calibration. If not, re‑examine the processing.
🖨️ Pro tip: Use the Print / Download PDF button to keep these checklists in your field binder.

❓ Frequently Asked Questions

🔹 What is the correct processing order for patch test parameters?
The recommended order is: Roll → Pitch → Yaw → Latency. Roll and pitch affect the basic geometry; yaw and latency corrections are more sensitive and should be computed after roll and pitch are fixed.
🔹 How do I know if my patch test corrections are correct?
The primary validation is cross‑line analysis. If the depth differences at intersections are within the survey's uncertainty budget (e.g., <0.1 m for Order 1a), the corrections are correct. Also, the swath should appear symmetric in water column display.
🔹 Can I process patch test data in post‑processing software like Qimera?
Yes, Qimera and other post‑processing software have patch test modules that can compute corrections from raw data. However, the corrections are typically applied during data import or vessel configuration.
🔹 What if the computed corrections are very large (e.g., >2°)?
Large corrections indicate a significant mechanical misalignment. You should re‑check your physical mounting and re‑measure offsets before relying on software corrections. Very large angular corrections can degrade beamforming quality.
🔹 How often should I repeat the patch test?
At the beginning of every project, after any sensor removal or re‑installation, after significant vessel modifications, and if you suspect a change (e.g., after a grounding or impact). Many hydrographers also perform a quick patch test at the start of each survey season.

✅ Action Items & Internal Linking

🎯 Your Tasks for Today

  • 📌 Step 1: Bookmark the Course Homepage and the Hypack patch test guide.
  • 📌 Step 2: Download the checklist PDF using the button at the top.
  • 📌 Step 3: Practice processing patch test data if you have access to a sample dataset.
  • 📌 Step 4: Leave a comment on this post: "The most challenging part of patch test processing for me is ____."
  • 📌 Step 5: Proceed to Day 15: Module 2 Review & Quiz using the link below.
◀ Day 13: Patch Test (Part 1) Day 15: Module 2 Review ▶
🔗 Internal linking note: Day 13 and Day 15 are correctly linked. This builds a strong internal network for SEO.

👨‍🔧 Engr. Rokib Hossain | River Engineering Solutions

LinkedIn | Portfolio