Gauging Hull Structure for Class: The Technician's Working Method

The instrument decides the job

UR Z17 clause 1.4 is one sentence long and it structures an entire gauging campaign. On coated surfaces, instruments using the pulsed-echo technique — either with an oscilloscope, or digital instruments using multiple echoes with a single-crystal technique — are required. Single-echo instruments may be used on uncoated surfaces which have been cleaned and ground. The permission is conditional, and the condition is physical work with a grinder.

Multiple-echo measurement works by timing the interval between two or three successive backwall echoes rather than the flight from the front face. Because the coating sits ahead of the first backwall echo, it falls outside the measured interval and contributes nothing to the reading. The gauge reports steel, not steel plus paint. That is the whole reason class permits gauging through sound coating at all.

Single-echo measurement times the round trip from the coupled surface to the backwall and includes everything in that path. Put a probe on a painted plate in single-echo mode and the coating is measured as if it were steel, inflating the reading and hiding wastage. Put it on heavy scale and the beam scatters. Neither error announces itself on the display, which is why the regime governing ESP thickness measurement fixes the equipment rule before anything else.

Couplant and contact on real hull steel

Couplant exists to remove air from the interface, and hull steel makes that harder than a test block does. A gel or paste holds on vertical shell plating and overhead deck heads where a liquid runs off. Thicker couplant fills the roughness left by pitting and grinding. Water and light oil work on horizontal, prepared surfaces where the probe can be seated and read quickly.

Contact is the variable a technician controls inside the tank. A reading that jumps between values on successive placements is reporting the coupling condition, not the thickness of the steel. The response is more couplant and a firmer, flatter seat, then a repeat until the value settles and repeats. Recording the first number that appears on the display is how a gauging report acquires values nobody can reproduce, and reproducibility is exactly what the attending surveyor tests when he asks for a reading to be taken again.

Pitted and grooved steel defeats a flat contact face entirely. Corrosion of that shape returns readings from the deepest point the beam happens to find, which is the number that matters and also the number hardest to repeat. The ESP Code survey planning meeting puts taking representative readings where uneven corrosion or pitting is found on the agenda as its own item, which tells you how routinely it goes wrong.

Surface preparation without destroying the coating

Preparation splits along the same line as the instrument. Where the coating is sound and multiple-echo measurement applies, preparation means cleaning the contact patch of mud, oil and loose product — nothing more. Grinding through a hard protective coating to take a reading destroys the corrosion prevention system the surveyor is separately assessing and creates a defect where none existed.

Where the surface is uncoated or the coating has already broken down, preparation means removing rust scale, marine growth and weld spatter until bright metal shows across an area larger than the probe face. Hard scale is the specific enemy. It bonds well enough to look like sound steel to the eye and scatters the beam badly enough to produce readings wrong in an unpredictable direction, which is worse than no reading at all.

UR Z17 clause 1.5 requires the firm's documented work procedure to address inspection preparation, selection and identification of test locations, surface preparation, protective coating preservation, calibration checks, and report preparation and content. Protective coating preservation appears as its own line item. A procedure that treats grinding as the default has already failed the audit that qualifies the firm, an issue we work through in UTM operator qualification.

Spot readings, transverse bands and five-point grids

Routine gauging takes spot readings — individual values at locations the attending surveyor has selected so that the numbers represent, on average, the condition of the structure in that area. Annex 14 of the ESP Code makes that representativeness the surveyor's responsibility. The technician's job is to place the probe where directed and report what the steel says rather than what the docking schedule wants it to say.

Once substantial corrosion appears, spot readings stop being sufficient and the pattern becomes prescriptive. Annex 10 specifies a five-point pattern for each shell panel between longitudinals, applied to the suspect plate plus the four adjacent plates. Longitudinals take three readings in line across the web and three across the flange, over a minimum of three longitudinals in way of the suspect area.

Transverse bands carry their own geometry. Lower stools are gauged in a band within 25 millimetres of the welded connection to the inner bottom and a second within 25 millimetres of the connection to the shelf plate. Transverse bulkheads take a band at approximately mid-height and a band adjacent to the upper deck or below the upper stool shelf plate, each read as a five-point pattern over one square metre of plating.

Mapping readings onto structural members

A gauging report is a map before it is a data set. Deck and bottom shell readings are located by strake, identified by its number or letter, and by plate position counted forward and aft of amidships. Port and starboard values are recorded in separate columns because a ship corrodes asymmetrically — a tank that leaks on one side, a berth that always presents the same quarter to the weather — and averaging the two erases the finding.

Internal members need their own identity. A longitudinal is named by its number and the frame station at which it was read. A side shell frame is named by frame number, and the reading records whether it came from the web, the flange or the end bracket. Annex 15 of the ESP Code divides side shell frames and brackets into four zones, A to D, with zones A and B treated as the most critical.

Brackets, floors, girders and stiffeners each get their member name recorded rather than being folded into the plating around them. The reason is structural: the surveyor evaluates hull girder strength from longitudinal members and local strength from transverse ones, and a number filed under the wrong category enters the wrong calculation. Which member matters on which structure is set out across our maritime structures orientation.

Diminution arithmetic against original scantlings

Diminution is the difference between original as-built thickness and gauged thickness, expressed both in millimetres and as a percentage of the original. A 16.0 millimetre plate gauging 13.6 millimetres has lost 2.4 millimetres, which is 15%. The percentage is the figure compared against the allowable margin for that member, and the margin differs between a deck plate, a bottom longitudinal and a hatch coaming on the same ship.

Substantial corrosion is defined against that margin rather than against the plate itself. The trigger is wastage in excess of 75% of the allowable margin while remaining within acceptable limits. A member with a 20% allowable margin therefore reaches substantial corrosion at 15% diminution, not at 75% diminution. Confusing the two is an expensive arithmetic error, because it either misses a trigger the surveyor finds later or invents one that puts sound structure into extended gauging.

Ships built under the IACS Common Structural Rules use a different formulation. Substantial corrosion there is a gauged thickness between the net thickness plus 0.5 millimetres and the net thickness — an absolute band rather than a proportion. A technician working a mixed fleet needs to know which regime the hull sits in before quoting a trigger to anyone.

Filling the class TM form

The reporting forms in annex 8 of the ESP Code are prescriptive down to the column. Form TM1-BC covers all deck plating, all bottom shell plating or side shell plating, and its columns run: strake position by number or letter, plate position, original thickness in millimetres, gauged forward reading port and starboard, diminution in millimetres and percent for each side, then the aft reading with the same breakdown, mean diminution, and maximum allowable diminution.

The general particulars sheet is the provenance record. It names the ship, IMO number, class identity number, port of registry, gross tonnage, deadweight and date of build; then the company performing the measurement, who certified that company, the certificate number and its validity dates, the place of measurement, the first and last dates of measurement, the survey the readings support, the equipment used, and the qualification of the operator.

It closes with names, signatures and stamps — operator and surveyor named, both signatures present, the company's official stamp and the classification society stamp applied. UR Z17 clause 1.7 requires the attending surveyor's signature as documented verification of each separate job, and clause 5.2.11 requires the report to include a copy of the firm's certificate of approval. The report is not finished when the readings stop.

The three errors that get a report rejected

Unverified original thickness is the first. A diminution figure is only as good as the number it was subtracted from, and the acceptable sources are the ship's approved hull structure plans or the minimum thickness table issued with the survey programme. Filling the original thickness column from a sister ship's drawings, a rule table or memory produces a report where every percentage is unsupported.

Missing member identification is the second. A column of thicknesses with no strake letter, frame number or member name cannot be placed on the ship, cannot be compared against the right allowable margin, and cannot be re-gauged at the next survey to establish a corrosion rate. The surveyor who countersigns the cover page is attesting to readings he can locate.

Readings taken through heavy coating with a single-echo instrument are the third, and the most damaging because the numbers look plausible. The coating is measured as steel and the plate reports thicker than it is, which is the one direction of error a survey regime cannot tolerate. UR Z17 does not permit it, and a firm found doing it faces withdrawal of approval under the clause covering measurements improperly carried out. Firms check themselves against that exposure through an NDT programme audit.

Calibration checks, records and the audit behind them

UR Z17 clause 1.5 lists calibration checks among the items a documented work procedure must cover, and clause 5.2.5 requires a record of the equipment used containing information on maintenance and the results of calibration and verifications. Where equipment is found not to conform to requirements, the society assesses and records the validity of previous measuring results — meaning an out-of-tolerance gauge invalidates work already reported and accepted.

In practice that means a calibration check on a step wedge or reference block of known thickness and comparable material at the start of a shift, after any change of probe or instrument setting, and at the end, with the results logged against the job. A drift found at the end of a shift is answerable precisely because the start-of-shift check exists to bracket it.

The personnel side is recorded with the same discipline. Clause 5.2.4 requires records of approved operators containing age, formal education, training and experience for the services they are approved for, and clause 5.2.2 requires a minimum of one year of tutored on-the-job training. Technicians building toward that record start from ASNT certification and the method hours set out in our training hours guide.

Request a consultation