Introduction
This paper considers the general principles of surveying for corrosion and coating breakdown. Detailed corrosion surveying of specific areas or vessel types is a huge topic. Corrosion occurs in all structural areas of the vessel. Tanks and holds are subject to many kinds of corrosion problems, as in the outer hull, superstructure and deck structure. Pumps, pipe work and other fittings all corrode in the marine environment.

It would be impossible to cover all these areas in a short paper, so for this paper, examples will be taken from coated ballast tanks in bulk carriers and tankers. Corrosion surveying is often a matter of taking a technical “snapshot” of the state of the tanks and carrying out metal thickness gaugings to ensure that sufficient structural steel is still present to meet the requirements of the Classification Societies.

Understanding the mechanism and probably progress of the corrosion phenomena present allows surveying to be taken some steps further into the field of prediction. Identifying imminent failure is dealt with in some detail. The techniques and mechanistic models have been developed to enable a clear and quantitative assessment of the remaining usable ballast tank lifetime to be reached. The ability to make an accurate decision in this area enables:

Understanding Coating Breakdown
In general terms, surveying for corrosion starts with a quantification process. It is important to locate what type of breakdown is occurring, to what severity and in which locations. This is a visual process, which is aided by the experience of having seen similar breakdown in other vessels of the same generic type.

The second, and more important part of the surveying for corrosion process, is that of developing an understanding of the failure mechanism and why it is occurring. It is important to be able to answer the questions: why?, and how? Answers to these questions will enable the experienced surveyor to be able to put some kind of time-scale on the process and to be able to predict how soon catastrophic failure will occur.

Breakdown Time Scales

Over its lifetime, a coating undergoes a series of changes in its barrier properties. These are set out schematically in figure 1.

As a coating ages, it goes through three major stages: initiation, stabilisation and breakdown. The time that is spends in each stage is extremely variable. The initiation phase can be said to last between six months and two and a half years. During this time, the barrier properties tend to increase in a somewhat erratic manner.

The coating then enters a stabilisation phase where its barrier properties slowly decline over a number of years. When the barrier properties attain a certain level, a failure point is reached where catastrophic breakdown is initiated and the electrochemical loading in the tank increase dramatically. The corresponding changes in the level of coating area breakdown are shown in figure 2.

It can be seen that a very small percentage of breakdowns occur over the first seven to eight years. As the coating passes the failure point, progressive failure occurs; firstly on the edges and welds, and then on the flat areas.

Many factors influence the time spent in the three phases, but the major ones include:

· Surface and edge preparation at new building
· Shop primer type and retention levels
· Coating type and thickness distribution
· Sacrificial anode density and distribution
· Vessel trading characteristics

Breakdown Mechanisms

Ballast tank steel corrosion and the breakdown mechanisms of coatings are all electrochemical in nature with a large cathode driving a small anode. The large areas of the flat surfaces are usually the cathodes with the welds and edges being the anodic sites. This results in a strong tendency for the corrosion current to focus on any areas of weakness or non-uniformity in the coating.

The corrosion situation is most severe when the tank is drying out, as the transport of oxygen through the residual moisture film on the surface of the steel or paint is at its greatest. This situation can be changed dramatically both for better or worse by the presence of sacrificial anodes in the ballast tank.

Coating barrier properties are affected by the electrochemical processes occurring in the tank. The resistance of the coating to corrosion remains high for several years over the majority of the surface, but as some areas begin to decline in their corrosion resistance with time, this increases the level of electrochemical stress. After a period of time, there is a rapid loss of corrosion barrier effect in the flat areas. This generates the large and invisible cathode. The rate-controlling step is the barrier properties at the cathodic site. This is an invisible location on the coating where dissolved oxygen reacts to become hydroxyl ions.

Areas that become corrosion anodes are usually welds and cut edges. The rate at which these areas break down is usually determined by new building quality factors. For welds, residual oxides, poor preparation, irregularity and over-thick paint are major issues. For cut edges, residual oxides, rough and poor preparation, together with thin stripe coats, are important factors in determining the rate of breakdown.

Once early coating damage has occurred then the failure progresses through a “jacking” mechanism, whereby voluminous corrosion products form beneath exposed edges of the coating and lever it from the steel. Two types of jacking occur: calcareous scale, or deposit jacking and rust jacking. Calcareous deposit and rust jacking are shown diagrammatically in figure 3 and figure 4.

Each of these phenomena behaves differently with time. Scale or deposit formation tends to be self-limiting, whereas corrosion breakdown is not. The self-limiting nature of calcareous deposit formation can be seen in figure 5.

The run-away mechanism driving rust jacking is cyclic and tends to result in the type of breakdown shown in figure 6.

Quantifying Corrosion Severity

The extent of corrosion is usually independent of the level of penetration or metal loss. In some areas, such as tank bottoms, large amounts of metal are lost by a pitting mechanism. Pitting on welds is often difficult to measure electronically, and depth gauges are usually employed. More uniform metal loss is usually quantified by ultrasonic thickness gauges. Cracking is often detected by magnetic particles or dye penetrants.

Corrosion rate estimation is difficult as corrosion rates can sharply increase when fresh metal is introduced, or partial refurbishment is undertaken.

Quantifying Breakdown

The normal Classification Society method of quantifying ballast tanks gives three categories: “Good”, “fair” and “poor”. However, when a ballast tank has reached a “fair” condition the useable lifetime of the coating has probably been exceeded and steel repair will become inevitable as the tank will be well out of the “stabilisation” phase, and will be in the “breakdown“ phase.

It is important to be able to detect coating breakdown before the point that extensive refurbishment become necessary. In order to do so, it is necessary to assess the breakdown in a manner that enables the breakdown mechanism itself to be understood.

The vessel structure also needs to be understood in terms of how the breakdown on the edges and welds intersects with that on the flat areas. Once breakdown has initiated it is important to quantify its level and severity. Two methods are available: visual and instrumented assessments.

Visual Quantification of Breakdown

Traditionally, percentage area charts, or scatter diagrams, have been used to determine a percentage area for the whole tank. Typical charts are shown in figure 7.

These charts often look nothing like the real thing, as in coated ballast tanks, the flat areas rarely break down first, and corrosion tends to spread in form the edges. To reflect this, a new set of charts has been developed and typical examples are shown in figure 8.

As stated previously, the onset of catastrophic failure is often indicated visually by weld and edge breakdown. As quantification of the extent of these phenomena is of paramount importance, new linear extent diagrams have been developed to assess the extent of breakdown in these areas. The edge diagrams in figure 9 are also used to quantify edge calcareous deposit jacking.

The weld diagrams shown in figure 10 can also be used to assess the extent of weld pitting.

The most important point of the survey is to identify if the failure point has been exceeded. This is usually indicated by the extent of breakdown exceeding 1% and an increase in the rate of sacrificial anode consumption that usually results in pitting of the anodes. At this point, maintenance by crew touch up or repair becomes increasingly more difficult.

Instrumented Quantification of Breakdown

Instead of frequent, close-up, visual examination, barrier property measurements on the flat (cathodic) areas gives the earliest indication that the failure point has been exceeded. Comparisons over a large number of vessels shows a good correlation with visual methods.

Instrumented methods of measuring coating barrier properties give more reliable and consistent results. Electrochemical patch probes designed for coating assessment can be used to give a quantitative measurement of the quality of the coating.

Measurements of the substrate potential also provide information on the type of corrosion reaction occurring. The combination of both instrumented methods allows areas of weakened coating to be located for repair.

Some areas of the tanks begin to fail sooner than others and picking up this failure at the earliest possible time enables preventative remedies to be used that can extend the coating lifetime. Failure shows first in instrumented measurements, then in sacrificial anode consumption rates and finally in visual observations of coating breakdown at edges and welds.

Both voltage and current patch probes need to be connected back to the hull or ground, as shown diagrammatically in figure 11. A current patch probe in use is shown in figure 12.

These patch probes are the only method (to date) of locating weak cathodic areas. The first point at which site of failure can be detected is usually at the one year guarantee inspection. The first intermediate survey after 2½ years in service provides another opportunity for monitoring the condition at those locations and also the detection of other weak sites. The most crucial time is that of the first Special survey, after 5 years of service, when coatings often look good visually but their corrosion resistance may have declined dramatically to the point where total repair is unavoidable at the next dry-docking.