ApplicationsGuide

Corrosion coordinate mapping for internal vessel inspection

Corrosion coordinate mapping for internal vessel inspection: one datum head, axial station in mm plus clock position, a line any crew can find with a tape.

LumaSpec Vision11 min read Share on LinkedIn

On this page 10 sections
  1. 01One datum, two numbers: the unrolled-shell convention
  2. 02From clock position to tape distance
  3. 03Worked example: a 45 x 20 mm patch at station 3200, clock 4
  4. 04Marking the shell and drawing the developed view
  5. 05Photographs that carry the coordinates
  6. 06Trigger frames at the station marks
  7. 07When the flat map is the wrong tool
  8. 08Checks before the report leaves the vessel
  9. 09From map to repair order
  10. 10Questions

A corrosion map from an internal vessel inspection has to survive handoffs: the blaster must reach the patch, the painter must know what got coated, the welder must repair it, and next year's inspector must compare this year's depth readings with last year's. A photo alone carries none of that. The convention that holds up is coordinate mapping of the shell unrolled into a flat plate: fix one datum head, then give every defect two numbers — an axial station in millimetres from the datum weld, and a clock position read from that end. Below: the convention, the clock-to-tape arithmetic with a worked example, and the marking, photographing and archiving steps that let anyone walk to any defect with a tape and the report.

One datum, two numbers: the unrolled-shell convention

The whole map hangs on one decision made once: where the datum sits. Call the datum end Head A — the end with the manway or the nameplate, whichever your crew can name without argument — and paint a durable A beside its head-to-shell weld. The map's origin is where that weld crosses the 6 o'clock line (bottom dead centre). From there, two numbers place any point on the shell:

  • Axial station — millimetres from the datum weld, measured along the axis toward Head B. A defect 3200 mm down the shell is station 3200.
  • Clock position — stand at Head A, sight down the axis toward Head B, and read the clock projected onto the shell: 12 at top, 3 on your right, 6 at the bottom, 9 on your left. One clock hour is 30° of circumference.

Write the handedness sentence on page one of the report: clock read from Head A, stations increasing toward Head B. Every future reader inherits the same frame.

Defects in the heads leave the flat map. They take polar coordinates from the head centre — clock plus surface radius in millimetres, for example "Head A, clock 2, 450 mm from centre".

API RP 572, Inspection Practices for Pressure Vessels, is the standard reference for running the internal inspection itself. The convention above is the reporting format a crew can execute with a tape and reproduce year after year.

From clock position to tape distance

Two small formulas turn a clock position into something a tape can measure. Work in the circumference you will actually tape: inside circumference for interior work, outside for an external thickness-survey grid.

angle from top = clock hour x 30° θ = 180° − angle from top (angle above the bottom centre line) arc from bottom s = (θ / 360°) × π × D height above bottom h = R × (1 − cos θ)

D is the shell diameter you taped and R its radius. Keep the height column in play: a crew on scaffolding checks a tape result against how high the mark should sit before they commit to it.

For a 2400 mm shell, circumference π × 2400 mm = 7540 mm:

Clock (read from Head A) θ above bottom Arc from bottom Height above bottom
6 o'clock 0° 0 mm 0 mm
5 / 7 o'clock 30° 628 mm 161 mm
4 / 8 o'clock 60° 1257 mm 600 mm
3 / 9 o'clock 90° 1885 mm 1200 mm
2 / 10 o'clock 120° 2513 mm 1800 mm
1 / 11 o'clock 150° 3142 mm 2239 mm
12 o'clock 180° 3770 mm 2400 mm

Clock pairs share an arc but not a side: 5 o'clock sits on the 3-o'clock side (right, viewed from Head A), 7 o'clock on the 9-o'clock side. Write the side whenever the clock alone is not enough — "clock 4, 3-o'clock side" removes the last ambiguity.

Worked example: a 45 x 20 mm patch at station 3200, clock 4

Inputs. Illustrative vessel: horizontal shell, inside diameter 2400 mm, 6000 mm between head welds — a typical small process vessel; substitute your taped dimensions. Inspection finding, the record this example maps: a corrosion patch centred at station 3200 mm, clock 4, extending 45 mm along the shell axis and 20 mm around the circumference. Datum: the Head A girth weld at bottom dead centre.

Rule. One clock hour equals 30°; the arc distance from the bottom centre line is the angle's fraction of the circumference; circumference is π × D.

Arithmetic.

  • Clock 4 = 4 × 30° = 120° from the top, so θ = 180° − 120° = 60° above the bottom, on the 3-o'clock side.
  • Circumference = π × 2400 mm = 7540 mm.
  • Arc = (60° / 360°) × 7540 mm = 1257 mm.
  • Height check: h = 1200 mm × (1 − cos 60°) = 1200 × 0.5 = 600 mm above the vessel floor.
  • Bounding-box area = 45 mm × 20 mm = 900 mm².

Result — the report line:

C-03 · station 3200 mm from Head A weld · clock 4 (60° above bottom, 3-o'clock side) · 45 mm axial x 20 mm circumferential · 900 mm² · photos C-03-w, C-03-m

Decision. The blaster walks the bottom line with a 5 m tape: 3200 mm lands inside one tape length from the weld. Wrapping 1257 mm around toward the 3-o'clock side should land on the paint mark, about 600 mm off the floor; if the tape ends elsewhere, re-measure before the first grit passes. The reverse check proves the record: 1257 / 7540 × 360° = 60°, two hours from 6 o'clock, which is clock 4. Every defect gets the same treatment — a line that cannot be walked to with a tape is not finished.

Marking the shell and drawing the developed view

Coordinates only hold if the shell carries them. During the inspection:

  • Paint each patch outline with a paint pen. The outline is what survives into the blasting and repair scopes.
  • Paint station marks along the 6 o'clock line: a tick every 500 mm, the metre number at every 1000 mm. Repeat ticks at 12 o'clock for work off the top.
  • Give defects IDs — C-01, C-02 and so on on the shell, in the table and in the photo file names; head defects take A-01, B-01.

The drawing is the same map on paper: the developed view. Unroll the shell into a rectangle — 6000 mm long by 7540 mm of developed circumference for the example vessel. At 1:25 that prints as 240 mm x 302 mm, which fits an A3 landscape sheet, with the two head circles at 96 mm diameter drawn beside it. Defects go in as dots with their IDs, the bottom centre line runs along one long edge, and the station grid runs across.

An orange-peel sketch still works for the heads, but clock-plus-radius on a plain circle is faster to draw and faster for a welder to read. The paint, the table and the sketch are three copies of one coordinate set; when two disagree, the shell wins.

Photographs that carry the coordinates

A report line puts a crew within a tape length; the photo records what they found. Take two frames per defect: a context shot that includes a painted station mark, and a close-up with a scale in frame — a ruler or the tape itself. Name the files after the defect ID (C-03-w, C-03-m) so the map, the table and the storage folder index each other.

Pixels become millimetres as soon as a ruler is in the scene. The image scale calibration from a ruler method reads mm per pixel straight off it, and if the reference is a tape line seen at an angle, camera distance from a tape line's pixel width recovers the standoff as well. As an estimate: a sensor 7.2 mm wide with 1600 pixels across has 7.2 / 1600 = 4.5 µm pixels; a 12 mm lens at 300 mm working distance covers about 7.2 mm × 300 / 12 ≈ 180 mm, so 180 mm / 1600 = 0.11 mm per pixel, and the 45 mm patch of C-03 spans roughly 400 pixels.

Lens focal length, field of view and working distance
Units
Solve for
Focal length
– mm
Field of view
– mm
Working distance
– mm
Magnification
– x
Angle of view
– °
Pixel on the part
– mm

Thin-lens model: m = sensor width / field of view, f = WD × m / (1 + m). Accurate to a few percent above about 150 mm; check the lens's minimum object distance.

Find lenses with this focal length Open the full calculator

Corroded and passivated steel throw glare back at the lens and hide the pits you are mapping; the pipe internal surface glare control techniques translate directly to vessel walls.

Set focus, exposure and working distance on the bench, then lock them — nothing gets re-focused inside a dark shell. The camera stands category covers the hardware: the LS-MNT-0040 slide-table stand, height 700 mm, 400 x 450 mm base, lets you rehearse camera height and light angle exactly as they will sit, while the portable LS-MNT-0006 carries a camera and a small light at a repeatable offset from the wall on staging inside the vessel.

Camera standsMachine vision camera stand, professional slide-table type, height 700 mm, 400 x 450 mm base (LS-MNT-0040)LS-MNT-0040
Type
Camera stands
$629.00View partBuild a system with it
Camera standsPortable optical test stand (LS-MNT-0006)LS-MNT-0006
Type
Portable optical test stand (LS-MNT-S04)
$629.00View partBuild a system with it

Trigger frames at the station marks

If a crawler or rail rig captures the shell instead of a hand-held camera, tie frames to the same station marks. Set a reflective flag at every 500 mm tick and run a diffuse-reflective sensor on the carriage: each flag fires the camera, so every frame files itself under a known station. The MV-PE5502 diffuse-reflective sensor reads a flag anywhere from 100 mm to 1 m away and its NPN output drops into a camera's opto-isolated trigger input. PNP or NPN trigger wiring covers the hookup, and keep the pulse shorter than the exposure you want — otherwise the pulse width becomes the exposure.

Photoelectric sensorsHikrobot MV-PE5502 Diffuse-reflective sensor(10-350mm, Red LED, settable distance, NPN)MV-PE5502
Output
NPN
Light source
Red LED
Sensing range
100 mm - 1 m
Type
Diffuse-reflective
$49.00 · quotedView and quoteBuild a system with it

When the flat map is the wrong tool

  • Vessels too small to enter. The clock position has no audience when nobody climbs inside. Map by axial station with a fixed camera roll angle, one photo per station, and let the photo series be the map.
  • Internally lined vessels. Coordinates carry the coating condition too; what to log per defect is laid out in the internal lined pressure vessel inspection and coating integrity methods reference.
  • A replaced head. If Head B comes off and goes back different, re-datum the vessel, paint the new datum, note the change with its date, and re-check one archived defect with a tape before comparing old lines against new ones.
  • Vertical vessels. The convention rotates: datum at the bottom tangent line, station becomes height in millimetres, clock read from the top.

Checks before the report leaves the vessel

  • The datum is painted at the weld and named on page one, with the handedness sentence beside it.
  • Every line carries defect ID, station in mm, clock with side, axial x circumferential size, depth reading and photo IDs.
  • One defect per vessel has been walked back with a tape from the written line alone; the C-03 reverse check above is the template.
  • Every photo carries a scale; every context shot shows a station mark.
  • The developed view matches the table — same IDs, same stations, head circles present.
  • Depth readings sit beside the coordinates, not in a separate file.

From map to repair order

The map earns its keep downstream. The blaster scopes from paint and stations, the painter records coverage station by station, and the welder's repair file starts from the report line; when a patch goes to repair, the localized corrosion repair on pressure vessel shells per NBIC guidance covers what the repair documentation needs, and the coordinates are its backbone.

Everything the rig needs sits in two drawers: the trigger sensors category and the trigger-sensor finder for station-triggered capture, with the accessories finder and the stand configurator behind the support hardware, and the trigger configurator for the sensor side. For the inspection conditions themselves, imaging a pipe under pressure without draining covers live internal imaging, and 3D profiling of formed vessel heads covers the head geometry this flat map deliberately flattens. More field setups sit on the applications topic page; if a rig question survives this article, ask us.

Questions

Which end of the vessel should be Head A?

Whichever end your plant can name without argument: the end with the manway or the nameplate, or the north end of a horizontal vessel. Paint or stamp an 'A' beside the datum girth weld during the first inspection and state the choice on page one of every report. If a head is ever replaced, re-datum the vessel, note the change with its date, and re-verify one archived defect with a tape before trusting the old lines.

Inside or outside diameter for the clock-to-arc conversion?

The tape decides the diameter: use the surface the crew measures, so the inside circumference for interior work and the outside for an external thickness-survey grid. On a 2400 mm inside diameter with a 12 mm wall, outside exceeds inside by pi x (2400 - 2376) = 76 mm, about 1% of circumference. A defect mapped with the wrong one sits worst case about 38 mm off at the 12 o'clock line and exact at 6 o'clock.

How do I record corrosion in the vessel heads?

Switch to polar coordinates from the head centre: clock position plus surface radius in millimetres, for example 'Head B, clock 10, 450 mm from centre'. Measure the radius along the head surface with a flexible tape, not as a straight line through space. Draw each head as a circle beside the unrolled shell rectangle; the knuckle is the boundary — inside it head coordinates, outside them station plus clock.

Does the same convention work on a vertical vessel?

Yes, rotated ninety degrees. The datum becomes the bottom tangent line — or the manway nozzle standing in for Head A — the axial station becomes height above that datum in millimetres, and the clock is read looking down from the top. One hour stays 30 degrees of circumference, the arc formula is unchanged, and the dished bottom takes radius plus clock from its centre.

How accurately can a crew re-find a defect from the written line?

With painted station marks every 500 mm and the tape conversion, about 25 mm is a realistic re-find target on a clean shell. That is why the paint outline matters: mark the patch boundary itself, photograph it, and let the written line bring the crew within a tape length. After blasting removes the outline, the photo keyed to the defect ID becomes the only geometric record of where the corrosion sat.