Rate this article:  
Loss Prevention Circular No


Loss Prevention Circular No. 02-01

Damage to Ballast Tank Coating and the 1996 Norwegian Marine Insurance Plan


Ship Type: Tanker (built 1997)



Shipowners may experience tank coating damages of a complex nature, not only from a purely technical point of view but also when considered against the insurance conditions.  In this circular it is attempted to focus upon some insurance coverage concerns, whilst the technical aspects are summarised in short terms based on the coating experts’ findings.  For the insurance cover, there are several issues of particular interest, e.g. the question whether and when there is a physical damage/”casualty”, timing of damage in case it develops during more than one policy period, the principle of insurer’s responsibility to restore the vessel to the condition it was prior to the casualty vs. “betterment”, recurring damages etc.  We shall also briefly examine a difference between the Norwegian Marine Insurance Plan (NMIP) and other H&M conditions.


Course of Events

During the construction of the tanker, the ballast tanks were not blasted and painted in situ but in prefabricated blocks in the shipyard.  The ballast tanks were prefabricated from steel plate shot blasted and shop primed with zinc silicate primer.  A small margin of 200-300 mm was left unpainted at the open edge to be joined when the blocks were erected together.  This margin was primed with a metallic zinc coating prior to application of the main coating.  Two coats of a 2-pack modified epoxy coating were applied.  Each coat was 150 μm in thickness.


At the time of construction, the owner tried to convince the shipyard, without success, to grit blast the join up welds.  Grit blasting was prohibited in the building dock for environmental reasons.  Furthermore, the owner expressed concerns that the temperature resistance of the ballast tank coating applied to grit blasted steel would not withstand temperatures of 70-100ºC.


After only a few weeks in operation, cracks were discovered in the ballast tank coating.  Upon inspection, the ballast tank coating was found to be in good condition.  However, small cracks were present in the coatings on the join up weld seam (erection joint) in some places where white salts had formed.  Cracking was found in coating where the thickness was in the range 300-500 μm.  The coating on the join up weld seams was 500 μm to 1 mm thick.  In a few locations it was observed to be more than 1 mm thick.  In addition, cracks in the ballast tank coating were found on welds around temporary access holes in the cargo tank plate.


The temperature of the steel of the join up welds was approximately 60-65ºC.  As a result of the adjacent cargo tanks being filled with heated cargo of oil.  Coatings that were applied at block stage to blast cleaning weld seams and plate were found to be in good condition.


Internal stresses, poor adhesion and soluble corrosion materials (such as ferrous sulphate and ferrous chloride) in the coating induced corrosion.  Cracking of the coating on join up weld seams only occur where it is subject to thermal stress as the cargo tank temperature varied between 20-60ºC.  This was caused by storing and discharging heated cargoes of oil, ballasting with seawater and the mechanical flexing of the ship.


Although damage was relatively small, it took a disproportionate time to repair in dry dock.  Considerable time and effort would be required to erect and dismantle scaffolding needed to repair the damage.  Additional requirements were needed to provide full environmental controls of temperature, relative humidity and ventilation in the ballast tanks whilst repairs were being made.


The shipowner’s superintendent was concerned that the next scheduled classification society inspection was due in 2000 prior to the next scheduled dry-docking in 2002.  He was concerned that the steel thickness would be reduced by corrosion and no extra thickness had been allowed in the design of the vessel to allow for such corrosion.  The coating was considered the most important protective barrier against corrosion.


The primary causes of the damage was:


The lack of sufficient surface preparation of the join up weld seams and welds around temporary access holes before the tank coating was applied.


The variations of steel temperature in the range 20-60ºC depends upon whether the cargo tank is full of heated cargo, or whether the cargo tank is empty and the ballast tank full of seawater.  Similar manually prepared join up weld seams not subject to temperatures of 60ºC did not crack.  The heat was considered a factor that induced cracking of the coating where the adhesion was poor.


The coating thickness was excessive in some locations and was applied in a few thick coats rather than many thin coats.


The use of thinner high tensile steel may have resulted in the increased mechanical flexing of the steel plating when the ship was in service.


Coverage concerns

The vessel’s Hull & Machinery policy was based on the Norwegian Marine Insurance Plan (NMIP) of 1996.


Cause of damage

Causes Nos. 2 and 4 above cannot be relevant factors under the insurance policy.  The variations in temperatures were known and foreseen, and the paint system was designed to withstand them.  In this respect, there was no suggestion that the temperature had been outside the expected range.  The high tensile steel used is an approved material.  There was no damage outside the affected welding areas, which could indicate that the steel and coating were not able to take the strain where the preparation work and paint application had been applied.  Thus causes Nos. 1 and 3 were the crucial factors as far as H&M coverage is concerned.


Is the damage recoverable?

It is important to identify the relevant defects (or “faults”) and damages to be rectified and dealt with during repairs.


During construction, the parties involved deliberately chose to apply a surface preparation in the affected areas, which eventually turned out to be inadequate.


During construction, too much paint was applied in too few and too thick coatings.


As a consequence of the above factors, the painting began to crack shortly after the vessel commenced her operation.


Due to the paint cracking, light corrosion developed under the cracks.


The “defects” under item 1 and 2 above were present before inception of vessel’s H&M policy, and in any event, under the NMIP § 12-1 the insurer is only liable for restoring the vessel to the condition she was in prior to the occurrence of the damage.  Therefore, any costs related to correcting the two first items must be regarded as “betterment” and is not recoverable under the H&M policy.  Whether there in fact are any extra costs is another question.


The paint cracking under item 3 above is “damage” in the sense of the NMIP § 12-1, whilst the perils causing the damage (item 1 and 2) were “defects” and not damage from the building period.  It is reasonably clear that the peril/cause under item 1 above must be classified as “error in design” in the sense of § 12-4 of the NMIP.  Item 2 above may probably be called “faulty workmanship”, which however is not a term of the Plan.  Faulty workmanship may, in some circumstances, fall within the category “error in design”, but not necessarily.  It is likely that the paint cracking occurred after the vessel commenced operations and definitely before expiry of the first H&M policy year.  Consequently, the cracking had occurred in the first policy year.  If the insurers want to free themselves from liability, they must show that the peril (cause) is excluded. 


Being caused by error(s) in design, the relevant exclusions to consider for the casualty are found in § 12-4 of the NMIP.  § 12-4 deals with error in design and faulty material, which normally are “original” causes from the building period (although it is possible that such defects may be “built into” the vessel at a later stage).  It will appear that insurers are not liable for repairing or renewing part(s) damaged as a result of error in design or faulty material unless the affected part(s) had been approved by the classification society.  The crucial question is therefore whether or not the paint was approved by the classification society in the sense of § 12-4.  Under this provision it is not required that the actual part (the paint) shall be approved by the class over and above the approval implicit in the general supervision of the construction.  It is sufficient that the part is included in the class’ general checking procedure.  In this case, the class society required the ballast tanks to be coated and that the coating specification be submitted for approval.  This had been done.  Consequently, it is clear that the class society had approved the coating according to § 12-4, and thus the paint cracking was recoverable.


The light corrosion under item 4 above developed as a result of the paint cracking.  Again, this was caused by error in design (as explained above).  Consequently, the comments above are relevant here.  One may add that exclusions under § 12-3 of the NMIP should also be considered, as we are faced with “ corrosion”.  § 12-3 excludes “costs incurred in renewing or repairing a part or parts of the hull, machinery or equipment which were in a defective condition as a result of wear and tear, corrosion, rottenness, inadequate maintenance and the like”.  It is important to remember that the rule excludes ordinary wear and tear, corrosion or similar.  The corrosion in the present case was extraordinary and a consequence of a recoverable casualty, i.e. cracks in painting.  Therefore, § 12-3 is not relevant, and repairing the corrosion is recoverable.


Timing of damage – which policy is applicable?

With the insurance aspects in mind, the sequence of events can be repeated and summarised as follows: The vessel was delivered from the yard with an unknown “defect” which resulted in damage during of the first H&M policy period.  Repairs were attempted and stretched into subsequent policy periods.   Each time, it was hoped that repairs successfully would be of a permanent nature.  However, damage reoccurred during several policy periods, and developed in new areas due to the same underlying defects to the affected welding areas.


The question of timing of damage is dealt with in the NMIP § 2-11, and the principal rule in the 1st subparagraph provides that “The insurer is liable for loss incurred when the interest insured is struck by an insured peril during the insurance period.” If tested under this provision, there would hardly have been any claim, since it could be argued that the peril “struck” at the time when the error in design was committed, i.e. during the building period prior to inception of the first H&M policy period.  Such a result would clearly not respond to the assured’s needs for insurance.  However, § 2-11, 2 sub-paragraph contains an important modification to the general rule, It states, inter alia, that if a defect, which is present but unknown at the inception of a policy, later results in a casualty, then the defect shall be deemed to be a marine peril which strikes when the casualty occurs, or when the defect became known, if this was earlier.


Although it was known that the surface preparation was not ideal during the construction period, the parties believed that the paint would hold.  It has not been suggested that the assured knew of the defects before paint cracking was discovered.  We have also concluded above that the initial paint cracking must have occurred within the first H&M policy period.  Therefore, the repairs of the cracking discovered during the first policy period should be honoured by the insurers at risk during that first year.  For damages incurred during subsequent policies, as a result of the same underlying cause, it is important to remember that the causative defect had been known to the assured in the sense of § 2-11 of the NMIP since the first H&M policy period.  In accordance with § 2-11 2. sub-paragraph, the peril must be deemed to have struck at the time the defect became known, i.e. during the first policy period.  Consequently, all damage falls on the policy in force at the time of discovery of the first cracking.


Number of casualties and deductibles to be applied

It should be noted that § 2-11 of the NMIP only addresses the question as to timing of damage.  It does not address the extent of cover, number of deductibles to be applied, etc.  Consequently, the question arises as to whether the failed repair attempts would constitute a new casualty that is subject to a separate deductible.  For the H&M insurance, the application of deductible per casualty is dealt with in NMIP § 12-18.  This section with relevant Commentary goes far in accepting failed repair attempts and redoing repairs as part of the original casualty.  Only where the repairers have committed an “unforeseeable” error (e.g. gross negligence) can a separate deductible be applied.  This is hardly the case in the present matter.


Differences between the NMIP and other conditions

In this particular case, there is little significant difference between the extents of cover under the most frequently used H&M conditions (e.g. NMIP, ITCH and the American form (AIHC)).  On the other hand, there is a difference as to timing of damage, i.e. in the question as to which policies the claim should be charged to.  As we have shown above, under the NMIP, the decisive factor is when the “peril struck” (supplied by the modification in § 2-11, 2. subparagraph).  Under the other forms mentioned, the question is when the damage occurred.  In the vast majority of cases the result will be the same.  For the casualty under review, it appears that the damage has progressed during more than one policy period.  Therefore, the claim would be apportioned over the various policy periods under ITCH and AIHC.


Although it is not necessarily relevant to this case, another issue that has generally attracted little attention is the question of the “chain of causation” being broken (e.g. repairer’s negligence).  Under ITCH and AIHC (“named perils” cover), repairers’ negligence is an insured peril.  In the case where damage occurs as a consequence of a negligent act by the repairers (e.g. during repairs of a recoverable casualty), the negligent act may well break the chain of causation resulting in another casualty with another deductible being applied.  On the face of it, an equivalent result might appear under the NMIP.  It will follow from the NMIP § 12-18 with the associated relevant Commentary that the decisive factor is whether the error committed is “foreseeable”.  Only where the error is “unforeseeable”, shall the new damage be deemed a new casualty and thus giving rise to a separate deductible.


The Commentary goes on with some examples, and e.g. if the error is “grossly” negligent, it is “unforeseeable” On the other hand, it is stated that it is “not necessarily unforeseeable that a part is installed the wrong way in an engine, cf. the VESTFOLD I case.  Substandard work, e.g. poor welding work, will normally also be foreseeable.  If the yard’s error is foreseeable, both the repairs of the same damage and the further development of the damage must be recoverable without any new deductible  It is submitted that certain errors, which would be accepted as “foreseeable” under the NMIP, would break the chain of causation when considered under other conditions.


Lessons to be learned

If a shipowner is considering changing the insurance conditions for his fleet, attention should be given to the difference between the various conditions.  This is particularly so in respect of timing of damage.  To illustrate, one may consider a casualty such as the example presented above under the ITCH, where it would be apportioned over more than one policy year (e.g. 3 policy periods with 1/3 on each year).  Under NMIP the claim would fall under the first year’s policy in full.  If we further assume that the vessel was insured on ITCH during the first year, and on the NMIP during the subsequent years, the assured would recover 1/3 from the first policy and nothing from the subsequent policies.


Needless to say, the situation may well be opposite – there can be double cover, which is not a problem for the assured.  Nevertheless, it is possible for shipowners to want to change conditions without running the risk of falling between two sets of conditions.  Brokers and underwriters can assist with providing a “change in conditions” clause or similar, which takes care of such pitfalls.