Flooding Of Uk Fishing Boats

Banff and Buchan College of Further |Education

The Universities of Glasgow and Strathclyde

FLOODING OF UK FISHING BOATS

by BANFF & BUCHAN COLLEGE AND THE UNIVERSITIES OF GLASGOW AND STRATHCLYDE

JANUARY 2003

Banff and Buchan College of Further Education

The Universities of Glasgow & Strathclyde

Flooding of UK Fishing Boats www.banff-buchan.ac.uk

Page i

Banff and Buchan College of Further Education

The Universities of Glasgow & Strathclyde

THE FLOODING OF UK FISHING BOATS

Contents Executive summary EXECUTIVE SUMMARY............................................................................................... VI BACKGROUND ................................................................................................................... VI RESEARCH METHODS ....................................................................................................... VII SUMMARY OF THE MOST SIGNIFICANT FINDINGS .............................................................. IX GENERAL FINDINGS .......................................................................................................... IX FLOW DESIGN ................................................................................................................... XI MATERIALS AND FITTINGS ............................................................................................... XII ENGINE ROOM LAYOUT, LOCATION OF EQUIPMENT AND ACCESS................................... XIV OPERATION ..................................................................................................................... XIV INSPECTION OF SYSTEMS .................................................................................................. XV FREQUENCY AND COST OF FLOODING ............................................................................. XVI CONCLUSIONS AND RECOMMENDATIONS .................................................... XVIII CONCLUSIONS ............................................................................................................... XVIII RECOMMENDATIONS ...................................................................................................... XXII ACTIONS TO MEET THE RECOMMENDATIONS ...............................................................XXIV

Flooding of UK Fishing Boats www.banff-buchan.ac.uk

Page ii

Banff and Buchan College of Further Education

The Universities of Glasgow & Strathclyde

BACKGROUND.................................................................................................................. 5 OBJECTIVES AND ACHIEVEMENT ............................................................................ 6 OBJECTIVES........................................................................................................................ 6 FOCUS OF THE STUDY ......................................................................................................... 6 SOURCES OF INFORMATION....................................................................................... 9 GOVERNMENT INFORMATION ............................................................................................. 9 INSURERS INFORMATION .................................................................................................... 9 BANFF AND BUCHAN COLLEGE PREVIOUS STUDY ............................................................. 9 SURVEYS OF VESSELS AND PERSONNEL............................................................................. 9 SKIPPERS AND OWNERS REACTION TO SURVEY/RESEARCH ............................................ 10 FLOODING HAZARDS IN FISHING BOATS............................................................. 12 OVERVIEW OF SYSTEMS .................................................................................................. 12 SOURCES OF FLOODING ................................................................................................... 14 SUMMARY OF FLOODING HAZARDS ................................................................................. 16 ENGINE ROOM VOLUMES .......................................................................................... 17 MATERIAL (CORROSION) & FLOW (EROSION) PROBLEMS ........................... 18 GENERAL ......................................................................................................................... 18 GALVANISED STEEL ......................................................................................................... 19 OTHER PIPE MATERIALS AND FLOW RATES .................................................................... 20 MATERIALS FOR FITTINGS ............................................................................................... 21 FLOW EROSION: LIMITS ON FLOW SPEED ........................................................................ 21 LOCATIONS OF PIPE FLOW TURBULENCE ......................................................................... 22 THE PIPE MATERIALS FOUND IN THE SURVEYS ............................................................... 23 TESTING OF PIPING & SYSTEMS.............................................................................. 24 THE TIME REQUIRED TO SURVEY VESSELS ........................................................................ 24 ULTRA-SONIC TESTING ..................................................................................................... 24 FLEXIBLE HOSES AND COUPLINGS........................................................................ 26 HOSES .............................................................................................................................. 26 FLEXIBLE (VIBRATION) COUPLINGS ................................................................................. 27 BULKHEAD INTEGRITY .............................................................................................. 30

Flooding of UK Fishing Boats www.banff-buchan.ac.uk

Page iii

Banff and Buchan College of Further Education

The Universities of Glasgow & Strathclyde

SYSTEM DESIGN PROBLEMS..................................................................................... 31 SEA INLETS ...................................................................................................................... 31 CONDITION OF SEA INLETS ............................................................................................... 32 BILGE SUCTIONS .............................................................................................................. 33 OVERBOARD DISCHARGES ............................................................................................... 33 COOLING SYSTEMS........................................................................................................... 34 AUTOMATION AND VIGILANCE......................................................................................... 34 BILGE ALARMS ................................................................................................................ 35 ANALYSIS OF REGULATIONS.................................................................................... 36 METHOD........................................................................................................................... 36 SAMPLE ............................................................................................................................ 36 SUMMARY ........................................................................................................................ 38 FINDINGS ......................................................................................................................... 39 FINDINGS FROM GOVERNMENT AND INSURERS INFORMATION ........................................ 39 SUMMARY OF FINDINGS FROM DATABASES ...................................................................... 41 FINDINGS FROM SURVEYS OF ACTIVE FISHING BOATS .................................................... 41 SUMMARY OF FINDINGS FROM SURVEYS ........................................................................... 42 DISCUSSION, CONCLUSIONS, RECOMMENDATIONS AND ACTIONS ........... 45 DISCUSSION AND CONCLUSIONS ...................................................................................... 45 RECOMMENDATIONS ........................................................................................................ 49 ACTIONS TO MEET THE RECOMMENDATIONS .................................................................. 51

Flooding of UK Fishing Boats www.banff-buchan.ac.uk

Page iv

Banff and Buchan College of Further Education

The Universities of Glasgow & Strathclyde

APPENDIX A SUMMARY RESULTS......................................................................... 54 CORRELATIONS ................................................................................................................ 54 TABLES ............................................................................................................................ 55 APPENDIX B FREQUENCY AND COSTS ................................................................. 57 APPENDIX C TYPICAL REPORT............................................................................... 58 APPENDIX C TYPICAL REPORT............................................................................... 59 VESSEL DETAILS .............................................................................................................. 59 FINDINGS .......................................................................................................................... 62 RECOMMENDATIONS: ....................................................................................................... 62 APPENDIX D TYPICAL INTERVIEW......................................................................... 63 CRITICAL PIPE WORK AND SYSTEMS/ PROCEDURES INTERVIEW ....................................... 63 APPENDIX E..................................................................................................................... 66 EXAMPLES OF REGULATION ANALYSIS ............................................................................ 66 SCUPPERS, INLETS AND DISCHARGES ............................................................................... 66 COOLING WATER AND OTHER SEAWATER SYSTEMS ......................................................... 70 APPENDIX F EXAMPLE OF INFORMATION BOOKLET .................................... 77 TYPICAL LIST OF MATERIALS ........................................................................................... 77

Flooding of UK Fishing Boats www.banff-buchan.ac.uk

Page v

Banff and Buchan College of Further Education

The Universities of Glasgow & Strathclyde

THE FLOODING OF UK FISHING BOATS Banff and Buchan College of Further Education1 and Glasgow & Strathclyde Universities2 After a survey by the team a main sea water pipe, that had been identified as being thin although it looked “in sound condition”, was replaced at the request of the skipper. Inspection by a shore based engineer showed that it was indeed “dangerously thin and if left would have caused flooding”. EXECUTIVE SUMMARY The most effective ways to cut down losses and damage due to flooding are 1

To fit means to close sea inlet valves that would be operable in the event of significant flooding of the Engine Room.

2

To have functioning bilge alarms that can be tested and maintained.

3

To use flexible hoses and vibration couplings correctly.

4

To close all sea valves in harbour to avoid flooding and to confirm their operational effectiveness if they need to be used at sea.

BACKGROUND As part of their drive to improve safety at sea in the fishing industry, the Fishermen’s Safety at Sea Working Group commissioned a study of water ingress into fishing vessels, specifically with regards to critical pipe work. The aim was to identify and quantify crucial areas of deficiency and to make proposals that would help correct them. In the UK as a whole, foundering and flooding accounts for over 50% of all losses to all lengths of fishing boats and, in recent years, for Scottish boats, a rather higher percentage with Engine Room flooding as a significant source of loss3. The Working Group identified water ingress through possible critical pipe work deficiencies as one of the main problem areas and following a tendering exercise, employed Banff and Buchan College and the Universities of Glasgow and Strathclyde to investigate the problem and submit a report with conclusions and recommendations. 1 2 3

Departments of Nautical Studies and Marine Engineering The Department of Naval Architecture and Marine Engineering – a joint Department of Glasgow and Strathclyde Universities. MAIB Report on the Analysis of Fishing Vessel Accident Data 1992 to 2000. Issued July 2002. Engine room flooding caused 48% of all sinkings of Scottish boats over 12 metres in the years 1999 and 2000. Flooding of UK Fishing Boats

www.banff-buchan.ac.uk

Page vi

Banff and Buchan College of Further Education

The Universities of Glasgow & Strathclyde

The resulting study has focussed on flooding of fishing boats and, in particular, concentrated on raw sea water systems as these have most potential to cause large scale flooding. The study was sponsored by The Corporation of Trinity House, Maritime and Coastguard Agency (MCA), North East Fishermen’s Training Association (NEFTA), the Royal National Lifeboat Institution (RNLI) and Sunderland Marine Mutual Insurance Company. Additional assistance in carrying out the project was received from the MCA, Marine Accident Investigation Branch (MAIB) statisticians, Sunderland Marine Mutual Insurance Company, Scottish Boatowners Mutual Insurance Association, Pirie and Smith Ship Surveyors, vessel agents, skippers, owners, engineers, designers and boat yards. The project team are most grateful for all assistance given and wish to thank the individuals involved for their time and efforts. The study set out to answer two main questions:

•

Why do fishing boats flood so often?

•

What can be done to reduce the number of these floodings?

RESEARCH METHODS The work centred on practical surveys of 40 fishing boats and, in particular, their seawater piping systems as this was shown to be the most significant cause of flooding. Interviews were carried out with experienced fishermen on the issue of flooding and flooding related topics. Additional information was utilised from a previous project carried out by Banff and Buchan College1; “The Development of a Reliable Bilge Monitor and The Loss of UK Fishing Vessels Through Flooding”. This was coupled with a review of insurance claims2 related to flooding incidents provided by major fishing vessel insurers and with loss/damage reports from the MAIB to identify the most critical and commonly occurring faults in the fleet. Important considerations are;

1 The Development of a Reliable Bilge Monitor and The Loss of UK Fishing Vessels Through Flooding, Banff and Buchan College, 2001 2 Insurance claims reviewed during the past 5 years – 1997 to 2001

Flooding of UK Fishing Boats www.banff-buchan.ac.uk

Page vii

Banff and Buchan College of Further Education

•

The Universities of Glasgow & Strathclyde

at what stage the faults occur in the life of a boat: o do they derive from the initial design of the pumping and seawater systems on the vessel? o are any problems “built” into these systems at the construction stage? o are problems introduced as a result of modifications or repairs?

•

the reason for these faults occurring o is the system being used as designed? o is the system being maintained appropriately? o are any modifications to the system being monitored and checked?

•

whether the faults that occur could be found in conventional MCA or insurance surveys and indeed, what sort of survey could be used to find these faults?

The vessels sampled were typical of the UK fleet as possible sources of bias were minimised. The boats were not self-selected, nor were they subjectively chosen by the investigators from a wider pool of possibilities. They were selected from those that were available in the time available and against sponsors’ requirements to survey vessels of a certain size, age and also from different geographical regions within a given time period. It was not possible for the investigators to influence the selection. This lack of bias and the numbers surveyed are enough to give confidence that any commonly occurring faults will be typical of the fleet as a whole. At no time did any owner refuse to take part in the survey, thus underlining both the impartiality of the survey and the willingness of ordinary fisherman to work for a safer industry. The average age of the boats surveyed was 17 years – ranging from just under a year to well over 40 years old. Eleven (27.5%) had wooden hulls the remainder were of steel hull construction. The age and length distributions are shown in Figures 1 and 2. The following Report is in three parts. In this first section the main results are summarised and some solutions discussed. The summarised material is supported by the main body of the Report and detailed information may be found in the Appendices.

Flooding of UK Fishing Boats www.banff-buchan.ac.uk

Page viii

Banff and Buchan College of Further Education

The Universities of Glasgow & Strathclyde

SUMMARY OF THE MOST SIGNIFICANT FINDINGS This research has quantified many problems that have been previously highlighted by the MCA in their guidance to Seafarers. In particular many of the findings support recommendations that have previously been passed on to all the relevant players in the maritime industry via M Notices1 and some have been incorporated in recent Regulations for the 15 to 24 metre vessels2. Prior to this research it was difficult to quantify these problems, but an opportunity now exists for the commissioners of the research to plan their strategy to lessen losses at sea due to flooding through critical pipe work failure. To put the probability of flooding in perspective, consider a fleet of 1000 fishing boats. In an average year the vast majority (930) of these would not have a problem with flooding. Fifty boats out of the remainder would have a flooding incident that they controlled effectively without bringing in any outside help. Fifteen boats, however, would have to call in others for assistance and five boats would sink because of flooding.

GENERAL FINDINGS 1.

Faults and deficiencies are common across age and size groups.

2.

The most significant faults are those that could lead to full bore pipe failure or slower flooding that is undetected.

3.

Most fishing vessel pumping systems cannot cope with potential flooding rates.

4.

There is a need for a pumping capacity outside the Engine Room.

5.

Bilge alarms and the means of closing sea inlets and discharges are often deficient.

Distribution of Common Faults Fault associated with flow design associated with inappropriate use of materials Associated with rubber or polymer hoses. Associated with bellows (expansion joints). Associated with access to bilge alarm and valves Associated with pipe work support.

% of vessels surveyed 62.5 70 55 10 70 37.5

1

An excellent example of this is MGN 190 (F) which deals with the Premature Failure of Copper Pipes in Engine Cooling Water Systems 2 Code of Safe Working Practice for the Construction and Use of 15m (LOA) to Less Than 24m (L) Fishing Vessels Flooding of UK Fishing Boats www.banff-buchan.ac.uk

Page ix

Banff and Buchan College of Further Education

The Universities of Glasgow & Strathclyde

This photograph show a very simple cable attached to the valve handle and fed through the deck plates. A red plastic ring is attached to the upper end of the cable. A sharp tug on this red ring then quickly closes the valve. This is simple, cheap, reliable and effective, especially with the high visibility of the plastic ring. Nice idea, BUT….. the cable is too short!

Flooding of UK Fishing Boats www.banff-buchan.ac.uk

Page x

Banff and Buchan College of Further Education

6.

The Universities of Glasgow & Strathclyde

Almost every boat surveyed had at least one problem related to flooding or the ability to deal with potential flooding, but many were minor. 60%, however, needed immediate remedial action for defects.

FLOW DESIGN 7.

More attention should be paid to overall piping design. Pump and piping systems were sometimes found to be complicated to understand and use. This photograph shows a series of right angle bends used to align a new pipe with an existing penetration. If the replacement pipe had been installed more carefully a single straight connector could have replaced this series of bends. No comment on the dent!

Mis-use of a flexible hose and a flow problem in one picture. It joins pipes of different sizes that are very badly aligned and only one clip is used top and bottom

8.

Information about pumps and piping systems was seldom apparent in engine rooms.

9.

Pipelines had excessive flow rates for their internal diameter with too many right angle bends leading to erosion of the pipes. Flooding of UK Fishing Boats

www.banff-buchan.ac.uk

Page xi

Banff and Buchan College of Further Education

The Universities of Glasgow & Strathclyde

MATERIALS AND FITTINGS 10. Incompatibility of metal components, causing corrosion due to galvanic action, is common. This photograph shows mild steel studs being used to secure an aluminium brass pipe, flange and housing together. They were tucked out of sight; should they fail the flange and housing could separate resulting in a “full flow” incident. If the flange was sealed watertight and the studs kept dry, the situation would be greatly improved. 11. Galvanised steel as a general piping material on fishing boats is questionable in its effectiveness due to difficulties in making modifications and repairs.

This shows the reason why galvanised piping is not preferred. Repairs are not always performed correctly. Here, the pipe has not been re-galvanised after welding.

12. There is wide misuse of flexible (expansion) joints, flexible hoses and non type approved materials. 13.

Hose clips (“jubilee”) are frequently of the wrong material and are often insufficient in number.

Flooding of UK Fishing Boats www.banff-buchan.ac.uk

Page xii

Banff and Buchan College of Further Education

The Universities of Glasgow & Strathclyde

These photographs: show: On the right correct use of materials and coupling (although the quantity used seems excessive).

But, to the right the coupling was taking up the misalignment of the pipe work

Flexible coupling on the point of collapse – just a couple of years old! Note: Correct number of clips, but wrong material!

Flooding of UK Fishing Boats www.banff-buchan.ac.uk

Page xiii

Banff and Buchan College of Further Education

The Universities of Glasgow & Strathclyde

14. Poor support and fastening of pipe work is relatively common.

Poor support The pipe was rattling on the support and the whole assembly was supported at one end by a hose!

ENGINE ROOM LAYOUT, LOCATION OF EQUIPMENT AND ACCESS 15. Equipment is not always positioned with due regard to the consequences of potential flooding and, in particular, bilge alarms could be placed with more accessibility for testing and repair. 16. The potential ability to close sea inlets after even quite limited flooding was often lacking – this could make potentially dangerous situations irrecoverable and catastrophic. 17. Frequently valves and pipes were fitted in locations that made inspection and maintenance extremely difficult.

OPERATION 18. The common failure to close sea inlets and discharges in harbour is bad practice and sometimes costly.

If this seacock was used regularly in harbour to close down the sea inlet, then it would not be in this condition and would be free to close when required.

19. Knowledge of pumping systems on board a typical boat is not always as comprehensive as it should be. Flooding of UK Fishing Boats www.banff-buchan.ac.uk

Page xiv

Banff and Buchan College of Further Education

The Universities of Glasgow & Strathclyde

20. Identification marking of piping systems is limited making use by inexperienced crew/ persons difficult.

INSPECTION OF SYSTEMS 21. Ultra-sonic inspection of pipes is a valuable component of an overall vessel safety scheme.

This photograph shows the researcher using ultra-sonic equipment to test a pipe.

This photograph shows the same section of pipe but this time take from below floor plate level showing perforations that were undetected by the ultra sonic sampling above the deck plates. This section is directly connected to one of the main sea water inlets

22. MCA (and other) surveys do not always detect piping and system problems. 23. Video and photographic records of vessels, where permitted, proved advantageous in reviewing findings and stimulating thought and discussion.

Flooding of UK Fishing Boats www.banff-buchan.ac.uk

Page xv

Banff and Buchan College of Further Education

The Universities of Glasgow & Strathclyde

These two photographs were taken on similar age and type of vessels. The one on the left with the red ring around the connecting piece shows that all the materials used in its construction are compatible and no materials problems are being displayed (although there are nasty right angled bends). The pipe work below has the centre section replaced with steel (magnet stuck onto section). This section was failing over its entire surface. This is the main discharge from the pump and failure would lead to sea water being pumped via the sea suction into the vessel. This would have disabled this particular vessel’s main systems in a very short space of time.

Magnet

FREQUENCY AND COST OF FLOODING 24. The frequency of flooding incidents is high, as is the cost of flooding. 25. The frequency and cost of flooding represents both a danger to life and a financial drain on the industry. 26. There is considerable non-reporting of incidents.

Flooding of UK Fishing Boats www.banff-buchan.ac.uk

Page xvi

Banff and Buchan College of Further Education

The Universities of Glasgow & Strathclyde

Flooding of UK Fishing Boats www.banff-buchan.ac.uk

Page xvii

Banff and Buchan College of Further Education

The Universities of Glasgow & Strathclyde

CONCLUSIONS AND RECOMMENDATIONS Most critical flooding problems derive from the design, installation, operation or maintenance of raw sea water piping systems.

CONCLUSIONS General wear in pipe systems is to be expected as boats get older, but this was not always found. General pipe deterioration will usually cause leakage but rarely causes catastrophic flooding unless it goes undetected. The most critical defects are those that that could produce full bore failure - that is, catastrophic flooding. The types of failure that produce the largest flooding rates are defects that accelerate wear (corrosion/erosion), cracks that propagate and cause pipe work to shear, and failure of flexible/rubber connections or inserts. A critical evaluation should be made, during the initial design and construction stages of a vessel’s life; into the effects that catastrophic flooding would have on the vessel’s ability to cope with these situations. This risk assessment should guide the design and location of vital equipment and systems. There is inconsistency in the positioning of components, equipment and in the selection of materials for engine room systems. There is also evidence, from the boats surveyed, that suggests ‘repairs’ are performed with little regard for material compatibility and potential consequences. Over-seeing inspections during construction or in service are not picking up all the possible deficiencies. It seems that information about and understanding of fluid flow systems is not being adequately applied or disseminated in all situations. We conclude that there will always be incidents that are beyond the capability of a vessel’s systems to control. This places an emphasis on ways to detect and limit flooding. Despite this: i) the overwhelming majority of the boats surveyed (87.5%) did not have suitable means to close sea inlet valves that would be accessible if there was significant flooding in the engine room. This is such an obvious problem it is hard to understand why simple steps by the regulatory authorities, builders and vessel owners are not taken to fit appropriate means of closure; ii) the overwhelming majority (85%) of bilge alarms were positioned for early detection of water ingress but were placed awkwardly for testing and maintenance. There has to be a balance between function as an alarm and functionality of the alarm.

Flooding of UK Fishing Boats www.banff-buchan.ac.uk

Page xviii

Banff and Buchan College of Further Education

The Universities of Glasgow & Strathclyde

Most of the boats surveyed were built in accordance with the 1975 Fishing Vessels Safety Provisions Regulations and associated guidance. It is clear that these Regulations are in many instances unclear and ambiguous, requiring builders and surveyors to use their own interpretation. Because of this, the rules naturally evolved to set minimum requirements / standards and in most instances vessels surveyed were constructed to comply with these minimum requirements. Alarmingly, few owners were aware that their vessels were only meeting a minimum standard and that they were free to modify - and improve - their vessels. Many vessel operators were under the impression that their UK Fishing Vessel Certificate was definitive evidence of the vessel’s seaworthiness and that modifications / improvements could only be made if advised by the MCA. Thus, the Regulations, themselves are enforcing an artificial ceiling on good practice. It is felt that this should be made clear to the industry. We have found, moreover, that the recent Code of Safe Working Practice for the Construction and Use of 15 metre length overall (LOA) to less than 24 metre registered length (L) Fishing Vessels is not as clear and unambiguous as it could be about minimum requirements so as to avoid ‘individual interpretation’. It is very important that Regulations should be clear and that the requirements should be enforceable and enforced. Among the most immediately critical defects discovered on vessels was the misuse of flexible fittings such as hoses and flexible expansion pieces. These were frequently used as repairs and were also used in fairly new boats as a means of dealing with poor pipe alignment and fitting. Failure of these flexible fittings would usually lead to rapid flooding. The MCA have pointed out through M Notices that that these fittings should be replaced regularly as a matter of course, but a date of fitting is seldom apparent and no date of manufacture is impregnated into the fabric and, therefore, operators are unaware of the age of the item. Use of these items should be logged within a safety system and key players should be made aware of the correct way of using them. It must be stressed, however, that these fittings were never seen under floor plates in live sea water pipelines. The number of missing or broken pipe supports is also of major concern (37.5% of vessels surveyed). Failure to repair these seems to be due to lack of awareness of the consequences. These may be catastrophic as vibration induced fractures tend to be sudden and cause full bore pipe failure. In modern vessels many pumps for bilges and deck lines are electrically driven with remote switches for operating. Unfortunately, while this is convenient and beneficial in these times of reduced crew levels, it also reduces engine room visits, which in turn reduces the ability to spot leaks visually prior to reaching the bilge alarm level or that may go undetected by a malfunctioning bilge alarm. It also encourages valves to be left open which can cause back flooding in certain conditions. Progress is not always improvement.

Flooding of UK Fishing Boats www.banff-buchan.ac.uk

Page xix

Banff and Buchan College of Further Education

The Universities of Glasgow & Strathclyde

Whilst the surveys detected numerous thin pipe walls (25% of pipes were in excess of 25% average pipe wall thinning, of these 10% in excess of 40% thinning), it is felt that very few of these would have resulted in catastrophic flooding provided that bilge alarms functioned properly and gave adequate early warning. In most cases the application of an epoxy bandage would allow operations to continue until the vessel made shore and had a permanent repair made. Many deficiencies discovered during our surveys were probably found simply by “fresh eyes” specifically targeting pipes and flooding related areas. That implies there is a need for the vessel’s own engineer to devote time specifically to these issues in addition to normal operational effort. In some cases video footage and photographs were taken. These recordings were found to be invaluable for review and occasionally allowed previously unseen deficiencies to be detected either by a ‘second look’ or other opinion. In the following recommendations an attempt is made: i) ii)

to solve immediate problems; to produce systems and a regime that will reduce the number of floodings.

Our recommendations are a mixture of prescription, to set a base level of safety, and procedures / requirements that are intended to raise the amount of information available and to increase awareness of critical flooding issues. The recommendations are followed by actions for each part of the Industry.

Flexible coupling Misaligned to stretch and bend. Also under torsion. These must never be torsioned…or painted! Mild steel bolts?

Flooding of UK Fishing Boats www.banff-buchan.ac.uk

Page xx

Banff and Buchan College of Further Education

The Universities of Glasgow & Strathclyde

Flooding of UK Fishing Boats www.banff-buchan.ac.uk

Page xxi

Banff and Buchan College of Further Education

The Universities of Glasgow & Strathclyde

RECOMMENDATIONS 1)

In the short term,

1.1

A major effort should be made to inform all sectors of the industry of the hazards associated with misuse of flexible hoses and flexible expansion joints.

1.2

MCA Inspectors should be informed of the limits of, and correct use of, flexible expansion joints and hoses and should enforce removal of hazardous fittings.

1.3

Skippers should (again) be reminded of the potential consequences of failing to close sea inlets and discharges in harbour and Insurers should consider means of persuading operators to close inlets and overboard discharged in harbour.

1.4

All boats 12m and over should be required to fit means of closing sea inlets and discharges from a position that can be accessed when there is significant flooding of the engine room - say 30% of the volume of the space.

2)

In the medium term,

2.1

All boats 12m and over should have an independently powered pump situated above the main deck level in a position accessible in normal operations that can take suction from the engine room and fish room by piping that is permanently installed and independent from other piping systems.

2.2

All boats 12m1 and over should be required to carry a document that describes their engine room pumping systems in terms of layout, equipment part numbers, especially the specification of all piping. There should be a schematic of the engine room systems displayed within the engine room and there should be associated marking and colour coding of pipes and equipment. Any subsequent alterations to any one of these systems should mean an automatic update of the pipe work layout diagram in particular.

2.3

All boats 12m and over should have two independent bilge alarms fitted in the engine room at positions where they can be tested and maintained. It is clear that bilge alarms are also necessary in any major space such as the fish hold.

1

12m is the current MCA lower cut off point for the UK Fishing Vessel Certificate survey but the principles proposed here could apply to smaller vessels. Flooding of UK Fishing Boats www.banff-buchan.ac.uk

Page xxii

Banff and Buchan College of Further Education

The Universities of Glasgow & Strathclyde

3)

In the long term,

3.1

The Regulations should be framed such that: i)

ii) iii)

The capacity of the bilge pumping systems is governed by the potential rate of ingress in critical pipe failure and is defined by a required system flow rate rather than by the existing system of pump capacity. Both maximum and minimum flow speeds are defined within system pipe work for the materials used so as to minimise the effects of erosion and corrosion. A risk assessment is required of each boat when new or when a major refit is performed to assess the effects of flooding and to identify means to limit or control flooding.

3.2

An inspection regime should be implemented that checks pipes and piping systems and which should include sampling by ultra-sonics. As part of this, all owners of fishing boats 12 metres and above in length should perform regular inspections of their engine room and other raw sea water piping systems and arrange an independent audit every two years. Self certification is already accepted by the < 10m fleet.

3.3

A more detailed engine room inspection should be incorporated into the MCA survey so that more time is spent in the bilges. To free up time, regular equipment safety checks could be done by the skipper and randomly inspected by MCA if there was sufficient emphasis on deterrence for non-compliance.

3.4

Sea water systems based around a manifold of increased wall thickness should be encouraged in new builds, as these have proven advantages of simplicity and endurance.

3.5

All new vessels should be designed to reduce the possibility of back flooding from sea inlets to the bilges.

3.6

Use of galvanised steel should be discouraged except for larger pipes and manifolds.

3.7

Best practice guidance for sea water system design should be adopted.

Flooding of UK Fishing Boats www.banff-buchan.ac.uk

Page xxiii

Banff and Buchan College of Further Education

The Universities of Glasgow & Strathclyde

ACTIONS TO MEET THE RECOMMENDATIONS The Maritime and Coastguard Agency should: 1. Consider carrying out spot safety inspections on fishing vessels specifically to include bilge alarms, sea inlets and remote closure systems for sea inlets. Target time for spot check 30 minutes. 2. Consider how best to ensure that a risk assessment might be carried out on each vessel that includes risks associated with flooding. Once risks are identified and quantified effective action may be taken to reduce the overall risk. 3. Consider setting up a working party to evaluate the Code of Safe Working Practice for the Construction and Use of 15m (LOA) to Less Than 24m (L) Fishing Vessels (and associated Guidance to Surveyors), with particular regard to clarity in the rules for sea inlets, bilge pumping capacities, materials and components. 4. Examine the viability of new builds installing an auxiliary engine and bilge pumping arrangement in a dry space out with the main engine space. 5. Consider requiring all vessels 12m and over to prepare and carry on board a schedule of piping, pumps and associated equipment which includes material specifications, system drawings and recommended replacement intervals for short life fittings. The intention being to allow transfer of design intent and relevant information through the vessel’s life. 6. Consider requiring all vessels to display within the engine room a schematic diagram of the bilge and cooling water systems to assist the crew. The Fishermen’s Safety at Sea Working Group could: 7. Consider initiating research into: a) the degradation of piping systems and preferred best practice in design and maintenance. b) the value of carrying out regular critical piping system inspections by various means of testing ‘live’ pipe work. c) the effectiveness and prioritisation of various emergency systems to counter severe flooding. d) the shock loading and vibration levels in typical fishing boats at sea and the effects on pipe work. e) the true endurance of flexible fittings and hoses as they are used in typical fishing boats and methods to improve performance. f) the merits of introducing damage limitation in to statutory courses for deck and engineering officers to include suitable content and methods of delivery etc. Flooding of UK Fishing Boats www.banff-buchan.ac.uk

Page xxiv

Banff and Buchan College of Further Education

The Universities of Glasgow & Strathclyde

The UK Fishing Industry should: 8. Consider, where possible and practical, carrying an independent portable pump capable of pumping bilges in an emergency. 9. Consider carrying out regular safety drills which involves all members of the crew in basic bilge pumping and valve isolation operations and use of manual and powered emergency pumps. 10. Make a practice of shutting all sea inlets and discharges when vessels are unattended in harbour. 11. Fit a high bilge level warning system, in a position where it can be readily inspected and maintained, which would operate an external alarm in circumstances where the vessel is unmanned, i.e. when in harbour. 12. Fit two independent (and accessible) bilge alarm systems for engine room protection. 13. Fit at least one bilge alarm in all major spaces within the vessel. 14. Consider installing a “bilge monitor” system which constantly displays levels of water in bilges. 15. Test bilge alarm bilge switches daily where possible and test fish room bilge alarms by controlled flooding during cleaning operations. 16. Where practicable, install close circuit colour television cameras in the engine room. 17. Be aware of the potential for electrolytic action associated with installing low voltage negative earth electrical equipment. 18. Colour code all pipe work for easy and quick identification and label valves stating their purpose. 19. Improve weather tight and non-weather tight door working practice. 20. Apply an appropriate inspection routine in engine rooms.

Flooding of UK Fishing Boats www.banff-buchan.ac.uk

Page xxv

Banff and Buchan College of Further Education

The Universities of Glasgow & Strathclyde

Marine Insurance Companies should: 21. Encourage skippers to perform safety checks and drills and include a record of these in a vessel log. 22. Encourage vessel owners to compile video footage and still photographs of vessel’s engine room etc. for ease of visualising components and equipment to assist with potential claims. 23. Consider methods of persuading operators to shut sea inlets and discharges when vessels are unattended in harbour. 24. Carry out safety related spot checks on vessels, to include observation of safety drills, testing alarms and emergency response procedures. 25. Encourage vessel owners to carry portable salvage pumps and other risk reducing equipment such as bilge monitors and engine room cameras. Training Establishments are recommended to: 26. Offer training to fishermen in awareness of flooding dangers and in portable salvage pump operations. 27. Construct courses (certificated) and offer training in basic machinery operation, inspection and maintenance, including valve chests and pumping operations. 28. Offer short Safety and Management courses in Marine Insurance to cover possible implications upon the fishing industry of insurers’ becoming more stringent in their interpretation and application of the Marine Insurance Act clauses, warranties and definitions. 29. Offer technical awareness courses suitable for fishing boat designers and surveyors to cover fluid system design, degradation and inspection techniques. The RNLI could: 30. Consider applying their considerable experience and knowledge by becoming involved in delivery of mandatory safety courses to fishermen.

Flooding of UK Fishing Boats www.banff-buchan.ac.uk

Page xxvi

Banff and Buchan College of Further Education

The Universities of Glasgow & Strathclyde

Flooding of UK Fishing Boats www.banff-buchan.ac.uk

Page xxvii

FLOODING OF UK FISHING BOATS

by BANFF & BUCHAN COLLEGE AND THE UNIVERSITIES OF GLASGOW AND STRATHCLYDE

MAIN REPORT

JANUARY 2003

Banff and Buchan College of Further Education

The Universities of Glasgow & Strathclyde

The Flooding of UK Fishing Boats www.banff-buchan.ac.uk

Page 2 of 78

Banff and Buchan College of Further Education

The Universities of Glasgow & Strathclyde

MAIN REPORT CONTENTS BACKGROUND ........................................................................................................ 5 OBJECTIVES AND ACHIEVEMENT................................................................... 6 OBJECTIVES .............................................................................................................. 6 FOCUS OF THE STUDY................................................................................................ 6 SOURCES OF INFORMATION ............................................................................. 9 GOVERNMENT INFORMATION ................................................................................... 9 INSURERS INFORMATION........................................................................................... 9 BANFF AND BUCHAN COLLEGE PREVIOUS STUDY .................................................... 9 SURVEYS OF VESSELS AND PERSONNEL ................................................................... 9 SKIPPERS AND OWNERS REACTION TO SURVEY/RESEARCH .................................. 10 FLOODING HAZARDS IN FISHING BOATS ................................................... 12 OVERVIEW OF SYSTEMS ......................................................................................... 12 SOURCES OF FLOODING .......................................................................................... 14 SUMMARY OF FLOODING HAZARDS ....................................................................... 16 ENGINE ROOM VOLUMES................................................................................. 17 MATERIAL (CORROSION) & FLOW (EROSION) PROBLEMS .................. 18 GENERAL ................................................................................................................ 18 GALVANISED STEEL................................................................................................ 19 OTHER PIPE MATERIALS AND FLOW RATES ........................................................... 20 MATERIALS FOR FITTINGS ...................................................................................... 21 FLOW EROSION: LIMITS ON FLOW SPEED ............................................................... 21 LOCATIONS OF PIPE FLOW TURBULENCE ............................................................... 22 THE PIPE MATERIALS FOUND IN THE SURVEYS...................................................... 23 TESTING OF PIPING & SYSTEMS .................................................................... 24 THE TIME REQUIRED TO SURVEY VESSELS ............................................................... 24 ULTRA-SONIC TESTING ........................................................................................... 24 FLEXIBLE HOSES AND COUPLINGS .............................................................. 26 HOSES ..................................................................................................................... 26 FLEXIBLE (VIBRATION) COUPLINGS ........................................................................ 27 BULKHEAD INTEGRITY..................................................................................... 30

The Flooding of Fishing Boats www.banff-buchan.ac.uk

Page 3 of 78

Banff and Buchan College of Further Education

The Universities of Glasgow & Strathclyde

SYSTEM DESIGN PROBLEMS ........................................................................... 31 SEA INLETS ............................................................................................................. 31 CONDITION OF SEA INLETS ...................................................................................... 32 BILGE SUCTIONS ..................................................................................................... 33 OVERBOARD DISCHARGES ...................................................................................... 33 COOLING SYSTEMS ................................................................................................. 34 AUTOMATION AND VIGILANCE ............................................................................... 34 BILGE ALARMS ....................................................................................................... 35 ANALYSIS OF REGULATIONS .......................................................................... 36 METHOD ................................................................................................................. 36 SAMPLE .................................................................................................................. 36 SUMMARY............................................................................................................... 38 FINDINGS................................................................................................................ 39 FINDINGS FROM GOVERNMENT AND INSURERS INFORMATION............................... 39 SUMMARY OF FINDINGS FROM DATABASES ............................................................. 41 FINDINGS FROM SURVEYS OF ACTIVE FISHING BOATS .......................................... 41 SUMMARY OF FINDINGS FROM SURVEYS ................................................................. 42 DISCUSSION, CONCLUSIONS, RECOMMENDATIONS AND ACTIONS.. 45 DISCUSSION AND CONCLUSIONS ............................................................................. 45 RECOMMENDATIONS ............................................................................................... 49 ACTIONS TO MEET THE RECOMMENDATIONS ........................................................ 51 APPENDIX A SUMMARY RESULTS ............................................................... 54 CORRELATIONS ....................................................................................................... 54 TABLES ................................................................................................................... 55 APPENDIX B FREQUENCY AND COSTS ........................................................ 57 APPENDIX C TYPICAL REPORT ..................................................................... 58 APPENDIX C TYPICAL REPORT ..................................................................... 59 VESSEL DETAILS..................................................................................................... 59 FINDINGS ................................................................................................................ 62 RECOMMENDATIONS:.............................................................................................. 62 APPENDIX D TYPICAL INTERVIEW ............................................................... 63 CRITICAL PIPE WORK AND SYSTEMS/ PROCEDURES INTERVIEW ............................. 63 APPENDIX E ........................................................................................................... 66 EXAMPLES OF REGULATION ANALYSIS................................................................... 66 SCUPPERS, INLETS AND DISCHARGES...................................................................... 66 COOLING WATER AND OTHER SEAWATER SYSTEMS ............................................... 70 APPENDIX F EXAMPLE OF INFORMATION BOOKLET ........................... 77 TYPICAL LIST OF MATERIALS.................................................................................. 77

The Flooding of Fishing Boats www.banff-buchan.ac.uk

Page 4 of 78

Banff and Buchan College of Further Education

The Universities of Glasgow & Strathclyde

THE FLOODING OF UK FISHING BOATS Banff and Buchan College of Further Education and Glasgow & Strathclyde Universities

BACKGROUND In the two years, 1999 and 2000, 29 Scottish fishing boats in the over 12 metres fleet were lost. 20 foundered due to flooding and 14 of these losses were specifically due to initial engine room flooding – that is 48% of all losses were caused by engine room flooding. On most of these occasions the vessels involved sank without loss of life, but, unfortunately, sometimes there have been fatalities1. Often the vessels in question sank due to flooding from causes unknown. It is hoped that the findings from this study will go some way to establish and quantify areas of weakness where such floodings may arise and by doing so, improve safety within the fishing industry. Insurance payments for damage to, or loss of, fishing boats of all sizes exceed £15 million a year on average; insurance payments for boats that sank in 2001 exceeded £6 million2. These figures highlight the cost to the fishing industry that stems from flooding of vessels. There is a clear need for some action to limit the most damaging causes.

1 Report on the Analysis of Fishing Vessel Accident Data 1992 to 2000; MAIB 2001. See Section 4 where it is noted that, in the eight years from 1992 to 2000, 35 fishermen lost their lives when vessels sank due to flooding. 2 Figures developed in this study from confidential information. The Flooding of Fishing Boats www.banff-buchan.ac.uk

Page 5 of 78

Banff and Buchan College of Further Education

The Universities of Glasgow & Strathclyde

OBJECTIVES AND ACHIEVEMENT OBJECTIVES Contracted objectives of the project were i)

To identify, as far as possible, the most frequent and likely causes of flooding of fishing boats.

ii)

To survey or inspect no less than forty, (mostly steel) UK registered fishing vessels. Ten vessels were to be registered in the UK, but out with Scotland and two vessels were to be less than 2 years old.

iii)

The vessels were to be surveyed on the slip where possible and findings were to be made available to the sponsors. The vessels and operators names however, were to remain anonymous.

iv)

To produce a final report containing firm conclusions and recommendations.

FOCUS OF THE STUDY It was clear from prior information assimilated and from sources developed in the study that the consequential flooding of Engine Rooms constituted by far the largest cause of total loss or damage due to flooding. In addition, piping system failure was indicated as a primary cause of many fish room floodings – generally through "back flooding." Within the category of flooding/sinking it was possible to identify causes such as hull failure and rudder or shaft gland leakage that contributed to floodings and losses. Where information existed, however, it showed that these were almost completely associated with a shock incident such as grounding or contact with the propeller or rudder. To investigate these was clearly out with the remit of the study. An early decision in the study was, therefore, to concentrate on pipe work and associated fittings and these of course are predominately found in Engine Room raw seawater systems. A total of 40 boats have been surveyed, meeting the criteria required. The age and length distributions of these are given in Figures 1 and 2.

The Flooding of Fishing Boats www.banff-buchan.ac.uk

Page 6 of 78

Banff and Buchan College of Further Education

The Universities of Glasgow & Strathclyde

Figure 1 Age Distribution of boats sampled 25%

Frequency in each age group

20%

15%

10%

5%

0% 0 to 5

5 to 10

10 to 15

15 to 20

20 to 25

25 to 30

30 to 35

35 to 40

40 to 45

45 to 50

50 to 55

55 to 60

Age groups

The Scope of Work suggested that, because some new vessels had problems with pipe work, two vessels of between one and two years old were to be inspected, the remainder to be more than five years old. In fact, the surveys found that there was little distinction between new and old vessels and in order to confirm this, the study team accepted newer boats when they were available. Figure 2 shows the length distribution of the boats surveyed. The Scope of Work set no restriction on length except that the vessels were to be above 12 metres.

Figure 2 Length distribution of boats sampled 25%

Frequency %

20%

15% Length

10%

5%

0% 14 to 16

16 to 18

18 to 20

20 to 22

22 to 24

24 to 26

26 to 28

28 to 30

Length ranges (m)

The Flooding of Fishing Boats www.banff-buchan.ac.uk

Page 7 of 78

Banff and Buchan College of Further Education

The Universities of Glasgow & Strathclyde

Each survey consisted of an inspection of the vessel’s engine room and identification of the pipes and components within the sea water and bilge systems. A descriptive narrative and diagram for each boat was prepared. Ultra-sonic testing was performed on pipe sections where wall thinning would normally be expected over time and in certain cases (with the owners permission) video footage and still photographs taken of examples of both proficiency and deficiency. The descriptions and diagrams were used to consider the flow regimes within the systems and to help identify deficiencies. The surveys and other studies of losses have led to a series of conclusions and recommendations addressed to the Fishing Industry, Marine Insurers, Training Establishments and the Regulators (MCA).

The Flooding of Fishing Boats www.banff-buchan.ac.uk

Page 8 of 78

Banff and Buchan College of Further Education

The Universities of Glasgow & Strathclyde

SOURCES OF INFORMATION GOVERNMENT INFORMATION Summarised accident and loss reports for the period from 1990 to the end of 2001 were obtained from MAIB. The main body of this was the information presented by MAIB in their report on the Analysis of Fishing Vessel Accident Data 1992 to 2000. This was modified with more recent information quality checked against the raw incident reports. It must be noted that MAIB believe only their major incident information is complete due to non-reporting from the fishing industry of what are perceived as lesser incidents.

INSURERS INFORMATION Information on damage and loss claims were obtained from two major fishing boat insurers who cover the majority of the UK fleet. This data gave the reported causation and the cost of both the claim.. The information allowed extension of the MAIB information to include unreported (to MAIB) incidents that had caused damage to vessels as well as providing information about flooding in harbour. Because the information was given in strict confidence, only summarised information is presented in this report.

BANFF AND BUCHAN COLLEGE PREVIOUS STUDY Banff and Buchan College had previously, through questionnaire and structured interview, obtained information and views about various aspects of flooding and this information was integrated into the study.

SURVEYS OF VESSELS AND PERSONNEL Each inspection was performed alongside or on the slip/dry-dock. A typical survey of a vessel produced a report that included information on the vessel’s details and history, the location of the survey and a description of the vessel’s condition at the time, a description of the engine room and associated systems, comments on the system design, the extent and results of ultrasonic testing, the findings and recommendations. The detailed review of the engine room systems was performed by tracking each system along all routes from inlet to discharge and identifying the function. The surveys were performed on 40 vessels that included 11 wooden hulled vessels, and 29 steel hulled vessels. The vessels varied from just under a year old through to fifty six years old. Lengths ranged from 16m through to 34m. The sample included scallopers, beam trawlers, vivier crabbers, pair trawlers, twin rig trawlers and small pelagic vessels. On occasion project researchers accompanied MCA surveyors and Insurance surveyors to observe their methods and techniques therefore gaining from their experience. Likewise, on occasion and with the vessel owners approval, an MCA surveyor shadowed the research team to observe their methods and techniques again under strictest confidentiality.

The Flooding of Fishing Boats www.banff-buchan.ac.uk

Page 9 of 78

Banff and Buchan College of Further Education

The Universities of Glasgow & Strathclyde

Operator survey information from Banff and Buchan Colleges research into ‘The loss of UK Fishing Vessels Through Flooding’ and some forty new personal interviews were used to assimilate information into this report. The results of the surveys and interviews are confidential and only summaries of the results are presented in Appendix A It must be stressed that the identity of vessels and interviewees were, and will remain, anonymous, even to the sponsors of this project. Examples of a typical report and structured interview format are in Appendices B & C. Note that neither of these is a ‘real example’ to preserve confidentiality.

SKIPPERS AND OWNERS REACTION TO SURVEY/RESEARCH Of the forty different vessels surveyed, not one owner/operator was obstructive or refused access to their craft, indicating the willingness within the industry to improve vessel safety. Wherever possible boats were surveyed in the dry-dock or slipway as this allowed us to inspect hull penetrations as well as internal pipelines and, because the vessels were laid up it didn’t interfere with fishing patterns. Several of the vessels surveyed were actually undergoing their four yearly MCA survey for their UK Fishing Vessel Certificate. This gave researchers the opportunity to inspect some pipe walls internally, as often valves were withdrawn in preparation for MCA survey. Some consider the fishing industry to be irresponsible and apathetic to safety issues. We found a very different attitude. Where deficiencies were found verbal feedback was given to operators. In some cases vessel operators observed the survey work, and in a few instances where deficiencies were found, they were attended to (upon the owners instructions) before the survey was completed. In three situations where the owners could only be informed by telephone of probable pipe wall thinning, they gave carte blanche to the survey team to instruct shore engineers to attend to the problem. At least one owner is known to have fitted high level extended spindles to ships side isolation valves since the survey of his vessel took place. Comments from operators regarding surveys Operators’ opinions on the effectiveness of the pipeline surveys were found to vary from one extreme to the other. Some operators thanked us for pointing out deficiencies and were well pleased that, if nothing else, we may have saved them a lost haul. An instance arose where an inspected vessel had flooding due to a compression joint parting several weeks after our survey had been carried out. This gave the operator the impression that the survey was worthless because it did not pick up on this weakness - despite the fact that these areas were not surveyed and indeed cannot be surveyed without dismantling fittings. The Flooding of Fishing Boats www.banff-buchan.ac.uk

Page 10 of 78

Banff and Buchan College of Further Education

The Universities of Glasgow & Strathclyde

From this we can sympathise with the MCA and the predicament they find themselves in regarding perceptions of the UK Fishing Vessel Certificate. Many operators are lulled into a false sense of security by the issue of the Certificate - thinking that their vessels are absolutely safe. In fact their vessels have reached a minimum base line for safety requirements, not the ceiling. We found a similar perception with regard to our surveys; which included ultrasonic inspection with random sampling of pipe wall thickness in areas that could be expected to encounter corrosion/erosion. At no time were components taken apart for internal inspection for integrity or strength. Some of the failings of ultrasonic testing were pointed out to operators – for example that reliable sampling could not be obtained near flanges or compression fittings. The survey of the vessel’s pipes and ensuing report can not be taken as a clean bill of health, just as a car passing its MOT is not guaranteed to be free of defects or the potential to break down.

The Flooding of Fishing Boats www.banff-buchan.ac.uk

Page 11 of 78

Banff and Buchan College of Further Education

The Universities of Glasgow & Strathclyde

FLOODING HAZARDS IN FISHING BOATS OVERVIEW OF SYSTEMS The life of an engineering system can be very broadly viewed as design, construction and then operation and the headings of design, build and operate can also be used to generate a framework of broad objectives for systems on fishing boats that are related to flooding (Tables 1(a) to (c)). These objectives are based on the well known idea of avoiding a risk, reducing the effect of the event if it occurs and providing contingency measures to deal with the consequences. Design has to be good. It is expensive to change systems – other than the components – after the system has been built and, thereafter, conflicts with other systems and difficulties with space and access may hinder attempts to change things. Design, therefore, has to consider the life of the system – how it may degrade and how it may be accessed for inspection and maintenance. It is the purpose of engineering qualifications and controls such as "Safety Regulations" to ensure that a design is, at least, competent. TABLE 1(a) System Objectives to Avoid Flooding Loss DESIGN WHAT HOW Avoid flooding As far as possible sources of flooding should be minimised. An obvious example is to minimise hull penetrations that cause weakness. Limit the extent of If flooding occurs then the volume that can be flooded flooding should be minimised. For example the vessel should be subdivided into a number of watertight spaces. Limit the rate of If flooding occurs then the rate should be minimised. flooding For example, pipes should be as small as possible. Limit the effects Essential vessel systems should not be affected by of flooding anything other than major flooding. The rate of flooding will control the time it takes to put the vessel at risk. Know about If critical flooding occurs there should be a way of flooding telling the crew before the situation is irretrievable. Bilge alarms are in fighting flooding. Deal with The crew should have means to deal with the flooding flooding by isolation and removal. Once it is known that there is flooding, shutting off isolating spaces and corrective pumping are the fundamental principles of damage control. In this respect crew training is essential. Control design Regulatory systems should exist that set out clear objectives and enforce minimum standards in design. Regulatory systems should exist that provide Control mechanisms to transfer the designers intentions to knowledge construction and operation. transfer

The Flooding of Fishing Boats www.banff-buchan.ac.uk

Page 12 of 78

Banff and Buchan College of Further Education

The Universities of Glasgow & Strathclyde

A system should be designed so that it can be built and that implies a relationship of some sort between the designer and the builder. At the least this requires experience on the part of the designer, and good practice by the builder. Both the designer and builder should ensure that they do not produce systems that are going to be difficult to access for operation or for maintenance. TABLE 1(b) System Objectives to Avoid Flooding Loss BUILD WHAT HOW Understand Key building staff should know why systems have been the design designed and why certain materials have been specified. Builders must have the professionalism to use what is required rather than what is available to avoid this. Follow the As far as possible, the intent of the designer should be purpose of followed. Changes or corrections should be assessed the design against the original intent. This requires good communication promoting two way transfer of knowledge. Use good Competent tradesmen should be used. Competence implies practice training, understanding and experience. Control Inspection systems should exist that check and control the construction competence of workers and their work, i.e. quality control. Clearly, a system also has to be designed so that it can be operated and fulfil its functions. There are often many ways to make a system work, but only a few will do what was intended, and that implies that the crew need to know how the system works and what it should and should not do. Operation also includes regular checking, maintenance and repair. These are intimately linked to design in that the purpose of the design has to be communicated to, and understood by all those who are working the system. TABLE 1(c) System Objectives to Avoid Flooding Loss OPERATE WHAT HOW The crew should be aware of the effects of flooding and the Understand critical nature of flooding on fishing vessels. There exist flooding requirements for skippers to understand stability, but effects everyone onboard needs to know how crucial some contingency measures are. Understand The crew (not just one person) should know how to operate systems the systems that control or remove flooding. Maintain System flooding control should be maintained by competent systems personnel who understand why the systems are built in a certain way and from certain materials. Continuity of knowledge from design through build to repair is needed. Regulatory inspection systems should exist to check and Control control the crews’ understanding and the maintenance of the operational systems. factors The Flooding of Fishing Boats www.banff-buchan.ac.uk

Page 13 of 78

Banff and Buchan College of Further Education

The Universities of Glasgow & Strathclyde

SOURCES OF FLOODING Water ingress into a vessel can be from, i)

boundary failures such as leaks through the hull, bulkheads and tank boundaries.

ii)

envelope failures in the fluid system pressure envelope (this includes all pipe line systems).

iii)

system faults where fluid flow is inappropriate or misdirected.

These are the broadest possible generalised categories and the vessel’s own pumping and piping systems are involved in two of them – highlighting the importance of these systems in maintaining the vessel’s integrity. Boundary failures The main sources of boundary failures are hull leaks through planking or at penetrations such as sea inlets and discharges, shaft and rudder glands. Whereas hull seepage is a fact of life on wooden boats, good maintenance keeps it in check and careful monitoring of bilges warns of any changes. Shaft and rudder gland failures affect all vessels, but dominate boundary failures on steel boats. Our review of insurance information suggests that shaft and rudder gland leaks are almost always associated with ground or gear contact or shock impact of some kind. Where vessels were in dry dock or slipped at the time of survey their hulls were inspected visually at potential suspect areas. Only one vessel inspected reported having this type of damage prior to dry docking and this was being attended to at the time. At another level are bulkhead and tank penetration leaks and open Weather tight (WT) doors and hatches. Tank leaks are naturally limited in volume so that, although they can cause damage, they are unlikely to lead to a total loss. Down flooding through doors and hatches is not usually the primary event in a flooding. It is clear, however from descriptions and reports of losses, that flooding through a door or hatch that is designated as ‘closed at sea’ is involved in many sinkings. The matter is not always so clearcut as ‘failure to shut a weather tight door’. Many vessels cannot be operated effectively with all doors closed and there are vessels where a door would have to be open to allow action or passage – and yet is meant to be closed to provide buoyancy. That is a design fault rather than an operational failure, and is out with the scope of this study.

The Flooding of Fishing Boats www.banff-buchan.ac.uk

Page 14 of 78

Banff and Buchan College of Further Education

The Universities of Glasgow & Strathclyde

Pressure envelope and system faults Envelope and system failures can affect all boats. Of a vessel’s envelope systems those with the largest flows present the most obvious hazard, but it is the size of the leak as well as the flow within the system that makes for a catastrophic failure. In a fishing boat the hydraulic and fuel lines may be discounted as they are limited in their effect. Equally, sanitary systems are small bore and should not cause major problems, although faulty installation and improper operation has led to hazardous flooding1. The systems of interest in typical boats that were studied in the surveys of working vessels are listed in the tables below. Vessel Sea Water Systems System Cooling water

Comment Continuous use from sea inlet when running machinery

Deckwash

Sea water pumped when required for deck/gear cleaning.

Process water

To clean catch - may be common with deckwash.

Ballast water

Occasional use to balance vessel.

Refrigeration

Usually continuous sea water supply at sea (if fitted).

Bilge water

Although the bilge system is a contingency system to deal with any flooding, there may be regular use of this system if there is continuous, known leakage into the vessel – say through a wooden hull.

Flooding due to system failures stems from those systems that are connected at one point to the sea and at another to a space or spaces – typically at a bilge suction. The potential exists for flow to be misdirected or for gravity/pressure to cause flow if control valves fail or if they are not used correctly. The systems that are most susceptible to this are those that take entry from or discharge to the sea below the water line.

1

Overflow of sanitary holding tanks through toilets has produced dangerous free surfaces within enclosed spaces and blocking of scuppers by waste. The Flooding of Fishing Boats www.banff-buchan.ac.uk

Page 15 of 78

Banff and Buchan College of Further Education

The Universities of Glasgow & Strathclyde

Possible System Failures System Ballast

Cooling Bilge

Comment Potential to pump to the wrong space, allow gravity flow to the wrong space from the sea or to allow gravity flow between spaces. Although generally enclosed, connections may pass water and allow flow through other systems. Failure, or trapped debris in a non-return valve can and often does allow flow directly from the sea into the bilge.

This type of flooding is especially connected to harbour flooding. Of 19 claims studied for machinery damage by flooding against one insurer, 12 were from flooding in harbour and in 11 of those cases, the sea inlet valves had not been closed when the vessel was left unmanned. All that is then required is a valve that is passing water for some reason and flow will take place to a low point – generally the engine room or fish room bilge. Surprisingly, perhaps, the deck wash system can also suffer from misdirection of flow when the tail of a hose is left in the sea and a siphon effect arises through a passing valve. It can happen at sea or harbour and, because the backflow suction rate is high the ensuing flooding can be rapid. Finally, in the study we found that most of the boats surveyed had a bilge system of the type only protected, from the sea by one non-return valve in a valve chest – making this component absolutely vital to the boat’s integrity. We comment on that later.

SUMMARY OF FLOODING HAZARDS From the design overview the key words are competence, control, communication and understanding. Mechanisms have to be found that will i)

Ensure competent design of systems;

ii)

Transfer knowledge through the life of the vessel so that, for example, correctly specified materials are used in repair;

iii) Ensure that the crew have knowledge of how to operate the systems; iv) Audit the competence of those who work on the system. From the review of flooding hazards it is clear that system failures can also be addressed by increasing the crew’s awareness of the hazards. That leaves failures of the pipes or piping fittings themselves as a primary cause of flooding that could be addressed to reduce losses - with concentration on cooling, bilge, deck/process and ballast systems. Larger bore pipes with potential for rapid catastrophic ingress and fittings that could fail full bore are the most critical. The Flooding of Fishing Boats www.banff-buchan.ac.uk

Page 16 of 78

Banff and Buchan College of Further Education

The Universities of Glasgow & Strathclyde

ENGINE ROOM VOLUMES As part of the study we measured the volumes of the Engine Rooms as a whole and up to the level at which we judged a vital system – electrical power or main engine - would fail (in calm seas). In any sort of rough weather with water moving in the bilges, systems could be lost sooner. Taking account of structure and large pieces of machinery, we found that (on average) just over 23% of the engine room volume on older style boats had to flood before our critical level was reached. There was, however, a significant problem found in newer steel vessels where the engine room layout includes wing fuel tanks that are built up from the shell to a height level with, or indeed higher than, the main engine.

Main engine

Fuel tanks (P&S)

This produces a central well around the main engine that has limited volume. The effect is that, for these boats only 8% of the volume of the space would have to flood before the critical level is reached – one third less than for the other older style layouts. Realistically the higher style of wing tanks not built into a vessel’s lower shell plating can reduce the time flooding takes to disable a boat by two thirds. Modern design however, allows the auxiliary engines to be sited on top of these tanks and that can be an advantage in flooding situations whereby pumping capacity is maintained after the main engine is submerged. Nevertheless, it is likely that this layout will lead to rapid loss of control of the situation if permanently attached isolation valve spindles do not reach up to the top of these fuel tanks. When these vessels age, and as systems start to deteriorate, the full impact of these design changes may start to reveal themselves in incidents where even moderate rates of flooding disables a vessel rapidly. In fairness to designers fisheries legislation has caused operators to build ‘rule beaters’ to maintain status quo of power and catching efficiency etc. Boat design has been driven to give owners alternatives that may not be best suited to the prevention of catastrophic flooding. e.g. Rules for 20m and over vessels “reporting in” prior to landing.

The Flooding of Fishing Boats www.banff-buchan.ac.uk

Page 17 of 78

Banff and Buchan College of Further Education

The Universities of Glasgow & Strathclyde

MATERIAL (CORROSION) & FLOW (EROSION) PROBLEMS GENERAL Corrosion We found a range of problems that could be traced to poor use of materials either when built or during repair. 70% of all boats surveyed had what we judged to be a material problem that would eventually lead to a pipe failure. 25% of vessels surveyed were actually displaying advanced corrosion due to dissimilar metal contact. Initially, these could be leaks that would be caught early by a functional bilge alarm, however a few may cause a complete pipe work section to collapse and in a pumped system this could produce quite rapid flooding. There was evidence of corrosion and erosion problems within the critical pipe work. 11 out of the 40 boats (27.5%) had what we would class as a pure corrosion problem caused by use of inappropriate materials. Frequently the problem had been inappropriately repaired by patching and not replacing the “suspect” pipe. On one vessel there was evidence of multiple levels of patching. In some cases there seems to be a tendency within the industry to replace material rather than solve the deep rooted design/construction problem that caused the initial failure. We found seven vessels (17.5%) with patches on their pipe work, including one on a vessel only fifteen months old. Under such a ‘patch’ erosion continues due to the ‘broken’ pipe wall. The underlying problem is either in the understanding or the implementation in practice of corrosion theory. In the worst cases of repair a dangerous problem was built into the system. Galvanic corrosion arises if two different metals are put in contact in a wet environment - and sea water is one of the worst. If this happens, there will be a flow of electrical current from one to the other. Electrical current is a flow of electrons and each electron has to come from one of the metals. As the electrons are taken away some metal is also lost and one of the metals will corrode away; the amount of this corrosion being proportional to the flow of current between the metals. If this is small then only a small amount of metal is lost – if it is large then corrosion will be evident on one of the metals. There is also a form of corrosion due to electrical faults such that the electrical system forces a current to flow and metal to be lost. This is called stray current or electrolytic action corrosion and is best thought of as the opposite of electro-plating. It is different from galvanic action and we did not see any obvious traces of it. If there are different metals in direct contact – say mild steel and aluminium brass pipe work - then the steel will usually corrode away. It is possible to The Flooding of Fishing Boats www.banff-buchan.ac.uk

Page 18 of 78

Banff and Buchan College of Further Education

The Universities of Glasgow & Strathclyde

design systems with dissimilar metals that don’t cause problems, but it’s a lot easier to avoid the problem by keeping metals apart. On one boat we found three, possibly four different metals in one length of pipe (the doubt arose because it wasn’t possible to test the metals in the allotted time and we had therefore to rely on observation and experience.). Erosion The research found that a majority of boats (62.5%) had what we judged to be problems associated with flow design of the sea water systems that was leading to erosion of the pipe walls. Here the problem lies with accelerated turbulent flow caused by sharp bends, sudden changes in flow and badly designed connections. To understand this it must be noted that, in general, materials gain resistance to erosion (and corrosion) by forming a thin film of oxide coating. If this coating is taken away by high speed flow then the metal itself will wear away dramatically. Some of this erosion was very severe and would eventually have led to reasonably large holes. There is an interaction here. Because erosion usually occurs where flow rates are high, the leaks that result could have high flow. Different materials have varying capabilities with regard to flow and our views and findings are discussed in the sections below.

GALVANISED STEEL The recently issued Code of Practice (issued November 2002) states that galvanised steel is generally unacceptable for sea water piping. Correctly fitted and in carefully designed flows, the galvanising has a finite life and any deficiencies in practice reduce the life. In practice, (i) (ii) (iii)

the pipes will at some time require replacing or re-galvanising by removal ashore, thorough internal cleaning and hot dipping in zinc; The zinc wears off at sites of turbulent flow and erosion/corrosion is then enhanced at these sites; Onboard repair is seldom effective unless a complete pregalvanised piece can be fitted between flanges.

We found galvanised steel to be acceptable in large diameter, thick-walled pipe as used on sea inlet suction manifolds and here only because the flow rates are relatively low and wall thickness is high. Sections of this piping can be replaced between flanges and the steel is upstream of the other piping so cannot be affected by copper related attack. Where galvanised steel is used in a manifold, great care should be taken that the exits from the manifold are by smooth concentric reducers and the manifold is effectively isolated at flanges from the downstream pipe work. No thickness reduction should be claimed for the galvanising – instead a margin of at least 1.5 mm should be added to the design life thickness.

The Flooding of Fishing Boats www.banff-buchan.ac.uk

Page 19 of 78

Banff and Buchan College of Further Education

The Universities of Glasgow & Strathclyde

There should, moreover, be a minimum flow rate of 1 m/sec to avoid stagnation. This is because very low flow rates may be a particular problem in galvanised steel suction manifolds if organic fouling occurs leading to aggressive sulphide attack. We found one large diameter galvanised steel manifold that was experiencing fouling problems, but no corrosion so far. The problem can be addressed by either fitting a back flushing connection or regular inspection and cleaning. Galvanised steel should certainly not be used for cooling systems where the temperature can get close to 60 degrees Celsius. This is where the zinc stops protecting and becomes protected by the steel! We found one system of closed loop cooling where this had possibly occurred.

OTHER PIPE MATERIALS AND FLOW RATES Candidate materials We conclude that either aluminium brass or one of the copper nickel alloys is used (70/30 or 90/10). There are other possible materials such as Duplex stainless steel and titanium, but these materials require specialist welding and fitting skills that are not always available at most build and repair yards. These findings reflect the new 15 to 24 metre Code of Practice which also reaches the same conclusions. Flow rates Where a maximum flow rate is discussed below, it should be clear that we are talking about flow through straight pipe. This automatically builds in an allowance for bends and turbulence downstream of fittings – as long as the bends and fittings are themselves built to good practice! If, for example, there is a series of bends that are tighter than they should be then flow erosion will take place. Aluminium brass Our review of published material and tests and the results of our inspections suggest that aluminium brass is acceptable provided that high flow rates or turbulent flow are avoided by good design. Flow rates for this material should be about 3.0 m/sec maximum to allow for enhancement at bends (an absolute maximum speed limit is 4 m/sec). Copper nickel alloys The 90/10 cupronickel material also performs well across the board, but cannot resist flow erosion as well as the aluminium brass or its cousin the 70/30 cupronickel. The maximum flow rate for the 90/10 material should be maximum 2.5 m/sec to allow for bend speed enhancement. Because of the lower flow limit to avoid stagnation, this limits the utility of the material. The 70/30 speeds are as for aluminium brass. Because there is a deficiency of iron in the sea water within wooden boats, iron anodes should be fitted to encourage the formation of the protective film The Flooding of Fishing Boats www.banff-buchan.ac.uk

Page 20 of 78

Banff and Buchan College of Further Education

The Universities of Glasgow & Strathclyde

that gives these alloys their positive corrosion characteristics. This might be used as an argument for use of a galvanised steel manifold, but that also argues for a breakdown on the zinc layer! Stainless steels It is most important that only acceptable marine grades of stainless steel are used because the standard 304 or 316 grades are not suitable for sea water systems and are only protected against pitting and crevice corrosion if there is a sufficiently large area of another corroding metal connected to it. It is not even clear that the high molybdenum (super) grades are free from crevice/pitting corrosion and, although such corrosion is more likely to lead to pinhole weeping, full bore failure at a flange cannot be discounted. We found stainless steel in use, but could not be sure of its integrity. The view from the offshore industry where techno-economic comparisons have been made is that “the use of stainless steels for sea water piping systems needs to be approached with great caution”.

MATERIALS FOR FITTINGS We observed few problems with alloy pipe or valve casings or chests (generally aluminium bronze) as these tended to be provided by specialist suppliers to meet marine standards and they would not usually corrode themselves in contact with steel. Steel valve chests, however, were a source of problems as there tended to be a concentration of corrosion at outlets and flanges that was usually associated with very poor flow into and out of the chests.

FLOW EROSION: LIMITS ON FLOW SPEED The effects of the fluid velocity can be shown in the following example. For carbon steel, liquids travelling at low velocities (< 2.0 m/s) show a general corrosion rate of 0.1 mm per year. As flow increases to a medium velocity (say 5.0 m/s) this rate increases ten fold to approximately 1.0 mm per year and as the flow rate further increases to high velocities (say 30.0 m/s) the increase is a hundred fold to 10.0 mm per year. Copper piping, as seen on 5% of the surveyed vessels, is adversely affected by water speeds above about 0.6 m/s as this causes deep pitting to occur. Fortunately these vessels did not have high flow rates and pipe wall reduction was tolerable. It is often said that stainless steel avoids these flow limits because, although it is susceptible to pitting corrosion at speeds below 1.0 m/s, above this speed the incidence of pitting decreases even up to 40 m/s. The problem is that, although the general corrosion rate remains very low, at any crevices in the surface the steel will continue to deteriorate. There are many crevices in piping systems at connections between pipes, and connections to valves – The Flooding of Fishing Boats www.banff-buchan.ac.uk

Page 21 of 78

Banff and Buchan College of Further Education

The Universities of Glasgow & Strathclyde

and stainless steels do not provide the clear advantage as might be imagined. As noted above, however, there is also a lower flow speed limit to avoid stagnation and possible fouling. It is because of this lower limit that redundant piping runs (say for an emergency cooling supply) should be blanked and drained until needed.

LOCATIONS OF PIPE FLOW TURBULENCE On the vessels surveyed, numerous areas were found where we consider there had been wall thickness loss due to turbulent flow erosion/corrosion. In general it is necessary to minimise turbulence in the system by avoiding design and fabrication details that cause turbulence. Such details are: •

• • • • • • •

Small radius bends. We found excessive use of these – particularly where engines were shoe-horned into small engine rooms and where pipes had been poorly aligned. As a general rule, the minimum bend radius should be about three times the diameter of the pipe. Bends immediately downstream from a pump where the turbulent flow from the pump would bear on the bend. Misaligned pipes. See the comment above, but also where pipes were fitted into the wrong size flanges so that the contact was not square. Partly protruding joints. Especially where pipes of different diameters joined and in some flanges that we were able to remove. Partly throttled valves Reductions in pipe diameter on full flow. Poor welds which leave rough and protruding surfaces inside pipes. Internal pipe corrosion – especially with galvanised pipes.

On the limited evidence, the same problem of flow exists from manifolds. It is essential that the flow out of a manifold is through a smooth concentric reducer welded into the manifold pipe. Directly welded stub pieces produce sharp edges and high turbulence just downstream of the entry from the manifold. It is particularly bad to have two turbulence raisers in close proximity, for example, two tight bends or elbows sited next to each other. An adequate straight length of pipe should be provided between turbulence raising components. Our surveys showed that some critical pipe work systems are less than satisfactorily designed. For example, many are fitted with small radius rightangled pipe bends for neatness and ease of fitting. The sharper the bend the higher the tendency there is for turbulence and erosion/corrosion. This was exacerbated in some boats by fitting a series of bends in very close proximity to each other. Pipe wall thinning was also found to be excessive when tight bends were placed too close to the inlet or out let of a pump. This is poor design practice. The Flooding of Fishing Boats www.banff-buchan.ac.uk

Page 22 of 78

Banff and Buchan College of Further Education

The Universities of Glasgow & Strathclyde

THE PIPE MATERIALS FOUND IN THE SURVEYS The survey team encountered a very wide range of metal pipe material during the course of the research. By far the most common were galvanised steel and aluminium brass. Among the others were mild steel, other aluminium alloys, copper, copper alloys, and stainless steel. It was not possible with the alloys to identify the exact make up of the metal in any particular pipe sample although this can be done very effectively, but expensively, with portable material analysers. The precise make up can, however, have a great effect on a pipe’s ability to cope with various water speeds and turbulence. It must be stated that the only materials that were found deficient across the board were mild steel and galvanised mild steel. There was, however, occasional thinning in sharp bends and 90 degree elbows for all metals whether steel or copper alloy. Aluminium brass alloy is a fairly common medium for pipelines and the research found few sections that were in immediate need of replacement. However this does not mean that it is without its faults. Aluminium brass does not cope well with water velocities in excess of 3.5 m/sec. If the surface of aluminium/brass alloy becomes broken for any reason it looses its protective oxide coating and deterioration is then rapid. This was apparent in aluminium brass 90° elbows where the bend caused turbulence and increased velocity. A good limiting speed (3.0 m/sec) has been quoted above that includes an allowance for bend speed enhancement. A small amount of stainless steel piping was encountered in the research and seemed to be in sound condition. Because, however, stainless steel can suffer from pitting and crevice corrosion in ‘quiet’ areas such as flanges and couplings it is hard to be relaxed about its use. The methods used in the project did not have the ability to detect thinning in these areas.

The Flooding of Fishing Boats www.banff-buchan.ac.uk

Page 23 of 78

Banff and Buchan College of Further Education

The Universities of Glasgow & Strathclyde

TESTING OF PIPING & SYSTEMS THE TIME REQUIRED TO SURVEY VESSELS Competent inspectors required about 2 hours to review the system layouts on board a typical vessel surveyed and a further 6 man hours to perform ultra-sonic testing of samples of the piping. A report with schematic layout requires at least a further 4 man hours of effort. Thus, a competent engine room system survey will take approximately twelve man hours. The surveys in this project were not exhaustive and therefore cannot be assumed to give a clean bill of health to any vessel surveyed. There is no doubt these man hour times could be improved if more information was available about the systems. In many vessels, the pipe work systems were found to have multiple pumps and a multiplicity of sea inlets and cross connections causing confusion over how to use the system efficiently. These systems were vastly more complex than required and, for example it took over five man hours on one twenty metre vessel to establish the function of all the valves and pumps within the engine room. These complexities also have a secondary effect of reducing crew confidence in their ability to set up the system for emergency pumping. Only one surveyed vessel carried a clear “user guide” to valve settings for pumping the bilges.

ULTRA-SONIC TESTING The study used ultra-sonic testing as a means of non-destructively identifying thin areas of pipe work. Sections were found that sounded and looked in good condition, but were quantifiably thinned by erosion or corrosion. This was confirmed on the removal of the pipe to make repairs. This revealed that hammer testing is not always a reliable means of finding thin sections - other than on large areas. Conversely there have been occasions where hammer testing has disproved ultrasonic results especially on hull inspections. It was found, however, that ultra-sonic testing is not a panacea for all problems. It is best deployed as a predictive tool that can be an indicator of problems or be used to monitor wall thickness loss though regular sampling would be required for it to be of best value. The problems inherent to ultra-sonic testing were found to be: i)

The need to know the exact material of the pipe being tested as different materials have different speeds of sound.

ii)

The difficulty of access to pipes that are probably corroding because the roughness of the external pipe wall can cause contact problems between the probe and the pipe and the nature of the inner wall surface of the pipe can cause spurious echoes.

The Flooding of Fishing Boats www.banff-buchan.ac.uk

Page 24 of 78

Banff and Buchan College of Further Education

The Universities of Glasgow & Strathclyde

iii)

The range of pipe diameters and materials that have to be tested so that the pipe curvature may not match the probe head. It is difficult to ensure the probe is normal to the surface and this can cause the pipe wall to be measured thicker than it actually is.

iv)

The skills and special equipment required to accurately gauge the wall thickness of thin pipes. There is a doubling effect as the probe nears its lower limits and this means that the sound double echoes to give a reading of double the wall thickness, this is exacerbated as further thinning of the wall will produce a quadrupling effect!

In addition, welds and other non homogeneities in the materials produce unreliable results and both crevice corrosion and pitting - among the most destructive forms of corrosion - cannot be detected because of the proximity of flanges and fittings and because it is difficult to detect localised corrosion. It is of course, in the area of flanges and compression fittings that pipe damage frequently occurs. MCA guidelines are that when the wall loses 25% of the original thickness the item should be replaced. Before a true sample of wall thinning can be calculated, however, the original wall thickness must be known. This information was not available to the researchers; therefore the next best thing was to sample areas which should not be worn – that is extended straight runs well away from bends or pumps or other turbulence raisers. It has to be noted, however, that even with new pipes wall variations may occur. Samples of bends “off the shelf” gave thickness variations up to 12.5% of nominal - in line with quoted manufacturing tolerances. The problem of dealing with thin pipes means that any inspector would require both special training and test equipment with special heads to deal with the likely piping. When a pipe is very thin and especially if it is corroding, the sound echoes from front and back wall and from internal deposits can be hard to separate. This requires training and experience but can be eased with purpose made equipment. Allowing for these parameters we still detected some 10% of vessels where the general wall thickness loss (weight loss) exceeded 40% of what we expect to have been the original thickness – a very significant loss of material. In particular areas, of course, the loss of wall thickness was much greater. 23 of the boats surveyed (57.5%) had lost more than half of the pipe wall in localised sample positions that were obvious erosion/corrosion sites. We found, however, that ultra-sonic inspection was very useful in sampling probable problem areas for weight loss erosion/corrosion – that is, general loss of pipe wall thickness as opposed to localised loss. We consider, therefore, that while ultra-sonic testing does have a value as a regular part of a boat’s integrity testing, it requires care. In addition, because ultra-sonic testing is giving a wall thickness estimate, there is a need to know the nominal thickness. We very strongly advise a requirement for a piping booklet that defines materials, schedules and gives a layout diagram. The Flooding of Fishing Boats www.banff-buchan.ac.uk

Page 25 of 78

Banff and Buchan College of Further Education

The Universities of Glasgow & Strathclyde

FLEXIBLE HOSES AND COUPLINGS HOSES Theory Flexible hose can be used to take up vibration and can also be used as a connector if there is limited space. It is not suitable as a permanent repair for damaged piping because a hose failure is likely to be full bore – or very near full bore. That means the hose is a critical component in any piping system. Used properly, a flexible hose should not be longer than one metre between end fittings that are permanently attached1. If the end fittings are not permanently attached then metal hose clips are usually used. It is very important that the correct stainless steel material is used and that at least two clips are used at each end of the hose, with each clip bearing correctly on the metal beneath. The hose must be specially selected for resistance to the fluid being carried and aging in the environment of use. Thus, the hose and its connectors should be suitable for the design pressure, the design temperature, the fluid being carried, the mechanical loads and pressure impulses, contact forces and abrasion and the environment outside the hose. Every hose has a manufacturer’s recommended life, but for marine use it is unwise to take this as any longer than 5 years. In summary, the use of hoses must be formalised. If they have to be used, then they have to be correct for the purpose and their type and date of manufacture should be impregnated on the hose. Hoses found Flexible hoses were often used wrongly to take up severe misalignment in pipe work and as a repair for failed piping. Many such hoses seemed long term fittings as they were of various indeterminate rubber/polymer materials, were often perished due to heat, age, vibration and oil or paint contamination. Swaged hoses were rare and only found as engine manufacturers supply so that the vast majority of hoses were fitted using jubilee clips. Probably the most common problem encountered was these clips. We found the wrong material more times than the correct i.e. mild steel as opposed to stainless. In many cases there were an inadequate number of couplings or the couplings were not properly bearing on the pipe beneath. Because mild steel clips had been used, we found couplings that had completely corroded leaving a “push fit” connection. The nature of the hose 1

See for example, Lloyds Register of Shipping guidance on use of flexible hoses The Flooding of Fishing Boats www.banff-buchan.ac.uk

Page 26 of 78

Banff and Buchan College of Further Education

The Universities of Glasgow & Strathclyde

material allowed it to be fitted on incompatible pipe diameters by stretching or squeezing on to the pipe stub in several instances. As far as length was concerned, we found many hoses that were too short to allow appropriate coupling but only two instances of a flexible hose longer than the recommended maximum length. Hose materials covered a wide range, unfortunately some were found totally inappropriate for pressurised sea water use in an engine room with the worst being no better than common garden hose. Whilst we appreciate these things may be essential for a "get you home," their condition often indicated fairly prolonged usage. 55% of vessels were seen to have flexible hoses with inappropriate connections or in need of replacement which included: • • • • • • • •

Inadequate number of couplings (jubilee clips); Wrong material for couplings (mild steel as opposed to stainless) Couplings completely corroded off, leaving a “push fit” connection. Inappropriate hose material Hoses too short to allow appropriate coupling. Hose of wrong diameter for pipe stretched/ squeezed on to pipe. Flexible hoses used to take up severe misalignment in pipe work. Rubber/polymer material perished due to: o Heat o Age o Vibration o Oil contamination

12% of interviewees reported flooding caused by flexible pipes either perishing, coming loose at jubilee clip fittings or splitting open to allow in each instance full bore flooding.

FLEXIBLE (VIBRATION) COUPLINGS Used correctly, flexible couplings isolate flexibly mounted, vibrating machinery from piping that is fixed to the hull (or other machinery). The couplings should be joining pipes that are well aligned and well supported so that the coupling can do its job of accepting, but not transmitting vibration – and not any other job such as taking up misalignment. The fittings are often to be found in the main cooling systems around the engines where there are very high flow rates of water. Failure occurring here can be a major problem.

The Flooding of Fishing Boats www.banff-buchan.ac.uk

Page 27 of 78

Banff and Buchan College of Further Education

The Universities of Glasgow & Strathclyde

The coupling, like the hose, has to be of the correct material for its application and it should be fitted to the maker’s recommendations with the pipes cut and positioned to suit the coupling, not the other way around. Often termed ‘rubber bellows,’ flexible couplings perform just like metal expansion joints to compensate for axial, lateral or angular movement and they will expand when under pressure especially if unrestrained. The ‘end thrust’ exhibited on a bellows can be substantial at pressures above 2 bar on a 65 mm nominal bore unit. Generally the thinner the walls of the bellows become the more flexible it becomes. If the bellows become axially overextended this ‘flexibility’ will increase. The immediate pipe work on either side of the bellows must be anchored to prevent it from extending too far. These couplings are made from a range of different materials and all have a limited service life of around 5 years, but the main influence on life expectancy in use is temperature - the higher the temperature the less the life expectancy. Their maximum pressures should be lowered when the working temperature exceeds 500 C. Due to the difficult conditions onboard fishing vessels, ship builders and bellows manufacturers normally recommend a 4-year replacement cycle for marine use. When the pipe work cannot be adequately anchored within manufacturer’s guidelines the unit must be tied by support rods (tie bars). In the case of installation on pump suction inlets vacuum support rings may need to be fitted. Tie rods limit lateral vibration to +/- 5mm maximum and axial and angular movement is prevented. Tied units are associated with ‘guided’ pipe work and untied bellows are associated with anchored pipe work. Untied bellows are recommended by manufacturers for bellows up to 150mm bore with pressures of 3 bar (45 psi) in non-engine cooling connections. If a pipe rises out from an engine and the first anchor is not within a straight run from it then tie bars should be fitted. These rods are also required where: 1. Pump surges occur. 2. Anchoring or pipe supports are not fitted. 3. Alignment guides are not fitted (in cases where wide temperature fluctuations occur) Best practice 1. Select the appropriate bellows for the job intended. 2. Fit the unit using metal spool pieces to ensure correct alignment at the mating to the pipe work flanges. Ensure the dimension between flanges meets manufacturers recommended bellows installation size and that the gap between them is not excessive. 3. When installed on a system for engine cooling either side must be ‘guided’ or ‘anchored’ according to manufacturers recommendations – thus ensuring that the bellows is acting as an acoustic break and not the associated pipe work. 4. The joint should not be subject to torsion or torsional vibration. 5. It should not be used for the purposes of correcting misalignment.

The Flooding of Fishing Boats www.banff-buchan.ac.uk

Page 28 of 78

Banff and Buchan College of Further Education

The Universities of Glasgow & Strathclyde

Inspection of bellows in service We consider that these fittings should be renewed after 4 years regardless of operation or condition. However, that requires the date of fitting to be logged. The problems we found included, o Profile changes (10% of vessels) – usually associated with excessive misalignment, torsion or excessive face-to-face dimensions. In this case the pipe work should be re-aligned and the coupling replaced. o Flange bolts contacting the rubber and causing wear (5% of vessels) – usually a result of poor fitting. o Painted bellows (20% of vessels) - the solvents in paint destroys the polymeric material and replacement is needed. o Lack of adjacent support (17.5% of vessels) - causing severe pipe work vibration problems due to excessive “spring” effect. Other possible problem with this type of fitting, o Cracking – usually as a result of overextension, angular or lateral movements – this requires immediate replacement. o Blistering, deformation or ply separation so that reinforcement material was visible – again this requires immediate replacement. o Rubber Deterioration – when a bellows feels ‘Gummy’ it should be replaced immediately. These fittings were used to mask badly fitted pipes in 4 boats out of the 40 surveyed (10%). Most often they were on larger bore pipes which were probably misaligned during fitting and the ensuing distortion caused by inserting the bellows was an easy option. They were found stretched, squeezed and twisted to compensate for the errors in the original pipe work. Failure of these items would probably give full bore flooding.

The Flooding of Fishing Boats www.banff-buchan.ac.uk

Page 29 of 78

Banff and Buchan College of Further Education

The Universities of Glasgow & Strathclyde

BULKHEAD INTEGRITY It is a part of any damage control strategy – especially important for flooding situations – to isolate compartments. For flooding, this means containing water in the compartment concerned; this is vitally important because any free flow of water into another compartment will increase free surface which may also produce critical trims which further increase inflows. Flooding between spaces can jeopardise a vessels chances of survival by turning recoverable situations into catastrophic. Review of MAIB and Insurance reports shows that this free flow of water through supposedly watertight bulkheads or conduits has been a contributory factor to many losses. As part of the vessel surveys, bulkhead penetrations and the condition of watertight bulkheads were inspected. 5% of vessels inspected revealed bulkhead penetrations that would allow water to transfer freely between compartments in a flooding situation. Only one vessel had a watertight bulkhead that showed a small area of significant localised thinning (by some 40%). This was caused by corrosion in an area that was alternately wet and dry and it affected unprotected steel plate. Although this study did not provide evidence for lack of bulkhead integrity as a common flaw, it is known from experience that boats have sunk when flooding progressed from one enclosed space to another.

The Flooding of Fishing Boats www.banff-buchan.ac.uk

Page 30 of 78

Banff and Buchan College of Further Education

The Universities of Glasgow & Strathclyde

SYSTEM DESIGN PROBLEMS SEA INLETS Problems The most common problems with these were, i) ii) iii) iv) v) vi)

Multiple inlets so that the chance of a failure is increased. Poor access – so that the valves and filters were difficult to operate or clean. Under floor plate positioning. Failure to fit methods of closing the sea inlet valves in the event of moderate flooding. Failure to fit a strainer in line before valve chests and pumps. Failure to vent the sea inlet boxes to reduce air locking.

Each inlet should follow established marine practice by having a thick walled pipe section before a screw down isolation valve followed by a strainer and a second valve (which may be in a valve chest). The thick walled pipe should be of a thickness not less than the hull plating or 12.5 mm whichever is the greater. There should be a maximum of two sea inlets for general operations unless a very good case is made for an additional inlet for a specific purpose such as, vivier or refrigerated sea water (RSW) systems where incompatible flow rates may result from mixing functions. The valves and strainer must be accessible and the ships side valve should be capable of operation when a significant proportion of the volume of the space e.g. 30% is flooded. The position of the filter/strainer is important. It should be placed in a position that will remove solids before any components are reached that could be damaged by them. Both pump impellers and non-return valves are seriously impaired by solids or build up of materials. The regulations allow individual builders to choose any type of ships side inlet. Our surveyors encountered three variations of the theme, namely large bore ring-main manifolds with only two major ships side penetrations; multiple smaller bored inlets with penetrations anywhere in the length of the engine room, and, multiple inlets feeding from dedicated sea chests/boxes built into both sides of the hull, therefore keeping all the valves in two localities. Advantages of a ring main manifold system with only two shipside penetrations were seen to be that in the event of flooding only two inlet valves need to be closed to isolate the vessel from the sea; the down side being that the bores of the pipes in this type of system are so great that if The Flooding of Fishing Boats www.banff-buchan.ac.uk

Page 31 of 78

Banff and Buchan College of Further Education

The Universities of Glasgow & Strathclyde

they rupture, ensuing flooding could be severe and possibly catastrophic. This risk is minimised by using heavy gauge pipe. The advantages of multiple smaller inlets from numerous ships side penetrations and each serving a dedicated unit are that flooding arising from any one pipe rupturing would be less severe. The problem arises in quickly identifying which pipe is damaged and possibly having to close upwards of six valves before complete isolation is achieved. Another downside is that these inlets were in some vessels, spread over the length of the engine room bilge port and starboard, making location equally difficult before isolation can begin. Many sub ten year old vessels were seen to have this location problem addressed by simply taking sea in via two main chests or boxes with multiple inlets confined to these boxes. We have strong feelings against the multiple sea inlets spread around the engine room due to their inconsistent positioning and because, where smaller bore pipes are used, pipe walls are usually thinner therefore prone to faster erosion. Access and means of closing Clearly, the valves and strainer must be accessible, but there should also be a way of closing the sea inlet valve when a significant proportion of the Engine Room volume i.e. 30% is flooded. This 30% figure is based on the calculated volumes of the Engine Rooms of the boats surveyed. Poor access leads to poor maintenance only 17.5% of the boats surveyed had sufficient access to work easily on the sea inlet valves and filters, these were predominantly the newest vessels in the survey. 12.5% of the boats we surveyed had one or more valves that were excessively corroded. This corrosion rendered them inoperable without a serious risk of them breaking. MCA have highlighted this problem for years and the industry knows of many cases where boats have been in severe difficulties or lost because the inlets could not be closed. Of the boats surveyed, 87.5% (35 out of 40) did not have a suitable system for shutting off the inlets in the event of high flooding levels in the engine room. (The extending of spindles etc.)

CONDITION OF SEA INLETS Where access could be obtained, none of the steel vessels of the thirty inspected revealed areas of unacceptable thinning on their sea inlet trunkings or sea chests – that is the entries directly adjacent to the ship’s side through to, but before, the isolation valves. General weight loss did not appear excessive and was commensurate with surrounding plate thickness. Two wooden vessel of the eleven inspected raised concerns on the condition of their main inlets, which also incorporated their isolation valves. One of these vessels’ was found to have a sea cock seized and it was felt that normal practices to free this would result in damage. The other vessel had evidence of direct dissimilar metal contact in her recently fitted sea cocks and galvanic action was already evident. The Flooding of Fishing Boats www.banff-buchan.ac.uk

Page 32 of 78

Banff and Buchan College of Further Education

The Universities of Glasgow & Strathclyde

BILGE SUCTIONS We usually found only one bilge suction and often they were fitted with very poor access. Because of the nature of bilges with the ever present possibility of fouling of their suctions, this can be at its worst when the foot of the line is open. Current regulations do allow this: in practice this has caused problems when this line becomes choked during an emergency situation. Second lines could be built into the systems at no great expense, which would allow for contingency. On many vessels there was only one non return valve in the system from the sea inlet through to the overboard and the non return valve was often a component within the valve chest. O/board

Figure showing sea inlet to bilge arrangement

Valve chest

Sea inlet valve

Sea Inlet

Engine Room bilge

Malfunction or debris within these valves have been the direct source of back flooding thus causing substantial losses to the industry. Interview and insurance company records confirmed that this single area of weakness causes substantial problems. A re-arrangement that fits a two way cock between pump and valve chest (with the other line to the sea inlet) solves this problem at the expense of causing pump priming problems when taking suction from the bilge.

OVERBOARD DISCHARGES These can become inlets if the boat develops a back flooding situation and must, therefore be protected as for inlets. The difference being that here a non return valve is fitted to prevent flow into the hull. The configuration should, therefore be, (from the hull) a thick walled pipe, an isolation valve that can be closed from a position that can be accessed with a significant volume (30%) of the engine room flooded, a screw down nonreturn valve and then the system. We suggest separation of the isolation function from the non-return function thereby allowing maintenance of the NRV as these are notoriously unreliable mechanisms.

The Flooding of Fishing Boats www.banff-buchan.ac.uk

Page 33 of 78

Banff and Buchan College of Further Education

The Universities of Glasgow & Strathclyde

In general, flow is continuous through overboard discharges and they can often be hard to repair due to access. We consider that any dispensation to the hull thickness in steel vessels is not advisable. The discharge pipe should be of a certain minimum thickness, but not less than the hull thickness in way of the discharge. Where accessibility for ultra-sonic sampling allowed, none of the vessels surveyed revealed thinner inlet lines (when they were new) than the adjacent hull plating.

Valve

NRV

Hull

COOLING SYSTEMS Sea water cooling systems have continuous flow when engines are running. They will probably be the systems that shows wear first and yet the pipes and valves are often routed with no thought for inspection or maintenance. However, on newer vessels (under five years old) we found these systems to be more accessible and user friendly. A particular problem we have found with galvanised steel pipes is where a looped cooling system is used so that warmed water re-circulates with a dramatically increased corrosion rate due to this temperature increase. Problems were found with redundant standby cooling systems and the connecting pipes into the heat exchangers. These little used items can lie dormant for considerable periods, but contain some sea water. Due to stagnation the chemical properties of the sea water changes and increases corrosion within the pipe. It is advisable to break, drain and blank the piping and connect it if and when it is needed. It is normal to paint these emergency spool pieces red to ensure visibility.

AUTOMATION AND VIGILANCE In modern vessels many pumps for bilges and deck lines are electrically driven with remote switches for operating. To utilise this capability to the full valves which must be operated manually are usually left in the open position. This is clearly convenient and beneficial in times of reduced crew levels, but it unfortunately also reduces engine room visits which in turn reduce the ability to spot leaks if undetected by the bilge alarm. The implementation of this type of progress requires careful consideration within risk assessment. The Flooding of Fishing Boats www.banff-buchan.ac.uk

Page 34 of 78

Banff and Buchan College of Further Education

The Universities of Glasgow & Strathclyde

BILGE ALARMS Most bilge alarms were found to be inaccessible with only six vessels (15%) giving easy access for testing and maintenance. These figures are in line with previous research by Banff and Buchan College which indicated that 35% of bilge alarms gave false alerts and that 36% of interviewees had encountered flooding of which their bilge alarm gave no warning. This study again highlights the fact that efficient operational bilge alarms are imperative. It is only when adequate early warning is given that successful remedial action – such as alerting the crew and shutting off flow from the sea – can be taken. Although every boat had a bilge alarm as required by Regulation, the fitting position appeared to have been selected to be as low in the bilge as possible. Many of these bilge alarms, however, had been sited in positions recommended by the MCA or, and some cases, agreed by insurance companies. It is unfortunate that, in the effort to achieve optimum placing for earliest warning of ingress, the alarms are often inaccessible. There has to be a balance between function as an alarm and functionality of the alarm. No vessel inspected had more than one bilge alarm apparent in the Engine Room. The November 2002 15 -24 metre Code addresses this issue but of course this does not apply to vessels smaller than 15 metres or over 24 metres.

The Flooding of Fishing Boats www.banff-buchan.ac.uk

Page 35 of 78

Banff and Buchan College of Further Education

The Universities of Glasgow & Strathclyde

ANALYSIS OF REGULATIONS METHOD A new Code of Practice for fishing vessels between 15 and 24 metres in length was published in 20011. Guidance for Surveyors related to the Code is being prepared at present. As part of this study we have analysed an example section of the Code of Practice and the results of this are given in Appendix D. The analytical technique is not new in itself, but has been reviewed and extended by John Lawson in a recent PhD thesis from Aberdeen University2. Essentially The aim is to grammatically analyse the text and identify verb clauses and thereby isolate these as clear instructions to take account of information, perform calculations or do things in some way thus removing ambiguity. When set against a clear set of functional requirements we suggest that this is an extremely powerful way to achieve clarity in Regulations and, from there, into Guidance.

SAMPLE A typical rule analysis taken from the section in the recent Code of Practice that deals with scuppers, inlets and discharges is given below where the original text is highlighted in bold, discussed and then a proposed new text is given, highlighted in bold italics. 2.2.6.2 Each scupper or discharge leading through the hull from spaces below the freeboard deck or from within an enclosed superstructure or deckhouse on the freeboard deck should have an automatic non-return valve fitted at the hull with a positive means of closure from an accessible position. N This consists of 2 rules and a modifying clause to define where they apply. The rules apply to scuppers and discharges from water or weather tight spaces leading through the hull and they are, a) scuppers and discharges to have an automatic non return valve fitted at the hull to prevent back- flow of sea water into the vessel b) scuppers and discharges to have a means of closure that can be operated from an accessible position when the nonreturn valve is not functional. Both rules are ambiguous because hull is undefined. The meaning is ‘hull below the freeboard deck,’ because no other section of a vessel’s hull requires this protection. 1 2

See MSN 1770. Aberdeen University thesis, 2002 John Lawson, The Flooding of Fishing Boats www.banff-buchan.ac.uk

Page 36 of 78

Banff and Buchan College of Further Education

The Universities of Glasgow & Strathclyde

Much of the definition of the spaces is redundant as it is clear that it is only watertight or weather tight spaces that require such discharges. Drainage from open decks to discharges below the freeboard deck is covered in 2.2.6.7 where an unspecified increase in wall thickness is required. See comments at that rule. The position of the NRV is not clearly defined with respect to access for operation or maintenance. The need to maintain these notoriously frequently unreliable valves in fact conflicts with the phrase at the hull. There is a need for an isolation valve near the hull and a non-return valve inboard of that. This allows maintenance of the NRV. Combining isolation with the NRV makes for complexity and inefficiency and consequentially poor reliability. Remote closure from an 'accessible position' is also not clear, as accessible is not defined either in terms of location or for ease of operation. It could usefully read ‘positioned above the freeboard deck and within the normal working area of vessel, located so that it is easy to use.’ If, however, this were accepted then a dispensation for engine room overboard discharges would be required. Either the isolation valve or the NRV could be subject to the 'remote closure' requirement, but it would be best to make this the isolating valve – again because of the unreliable nature of NRV’s. It would also be necessary to define certain requirements for the closing apparatus and ensure that there is, a way of specifically knowing when the valve is closed! The engine room dispensation would allow the remote closure to be from a location within the engine room but readily accessible after significant flooding of the space. This could apply to any machinery space, but it would be necessary for the space to be protected by bilge alarms. Significant flooding would be a level that would be expected to remove critical bilge pumping capacity – from our studies about 30% of floodable volume. The rule might then be written as 2.2.6.2(a.1) Each scupper or discharge leading from a water or weather tight space through the hull below the freeboard deck should have an isolating valve fitted near the hull. This valve should be capable of local control at the valve and also be fitted with a positive means of closure that can be operated from a position above the freeboard deck and within the normal working area of vessel, located so that it is easy to use. Indication of valve position should be provided at the remote closure position. N The Flooding of Fishing Boats www.banff-buchan.ac.uk

Page 37 of 78

Banff and Buchan College of Further Education

The Universities of Glasgow & Strathclyde

2.2.6.2(a.2) Notwithstanding the requirements of 2.2.6.2(a.1), if the discharge is within a machinery space that is fitted with a bilge alarm and serves machinery within that space, then the remote means of closure may be operated within the machinery space from a position that is readily accessible after significant flooding of the space. As a guide, significant flooding may be taken as more than 30% of the floodable volume of the space. N. 2.2.6.2(b) Each scupper or discharge leading through the hull below the freeboard deck from water or weather tight spaces should be fitted with an automatic non-return valve fitted inboard of the isolation valve to prevent back flow into the vessel. N

SUMMARY It should be clear that the Regulations as they exist – even this recent (Nov 2002) Code of Practice – can be improved. We consider that this analysis should be extended to cover all relevant sections of the Code and the results fed into the Regulations through Guidance.

The Flooding of Fishing Boats www.banff-buchan.ac.uk

Page 38 of 78

Banff and Buchan College of Further Education

The Universities of Glasgow & Strathclyde

FINDINGS FINDINGS FROM GOVERNMENT AND INSURERS INFORMATION Causes of flooding As noted, the main body of data had already been summarised and presented by MAIB. Additional information extended the results of the MAIB analysis of 1992 to 2000 incidents to include 2001. These new data were quality checked against raw reports. The insurance information was used to obtain mean average and upper values of claims for floodings and total losses due to flooding. This allowed the research to obtain estimates of the financial burden on the industry that is borne through insurance premiums. To avoid infringement of commercial confidentiality we have presented losses as a percentage of all claims in the figure below. Floodings and Sinkings (as percent of overall insured loss for each year) 70.0%

60.0%

% of cost for that year

50.0%

40.0%

Sinking and flooding

30.0%

All mechanical damage

20.0%

All other causes

10.0%

0.0% 1997

1998

1999

2000

2001

Year

It will be seen that partial loss (repairable damage) due to flooding and total loss due to flooding are equivalent to the claims for all mechanical damage making these two categories by far the largest impingement on underwriters. This figure is derived by looking at all the available evidence and producing a “one off” figure that does not provide a statistical basis for deriving any trends so it must be taken at face value with flooding and sinking averaging 31% of claims and mechanical damage 28% of claims (by value). The MAIB and insurance information was also used to help determine the major sources of flooding, so far as it was known, through study of reported incidents. This revealed Engine Room flooding to be the predominantly affected area and, that a significant majority of fish hold flooding was caused The Flooding of Fishing Boats www.banff-buchan.ac.uk

Page 39 of 78

Banff and Buchan College of Further Education

The Universities of Glasgow & Strathclyde

by bilge system problems. It would be expected that a sinking where the precise cause was not known would follow the same distribution. Frequency and costs The MAIB record flooding as the primary causation for 16 % of all serious incidents reported to them1. Moreover, flooding and foundering incidents from whatever cause account for 50% of all losses and engine room flooding can be firmly established in half of those cases. Thus, based on the MAIB reports, Engine Room flooding accounts for more (possibly significantly more) than 25% of all losses. Because MAIB receive only notification of serious incidents they cannot comment on the occurrence of other incidents such as flooding at sea when controlled by the crew or flooding in harbour. Only compulsory reporting (which is not at present mandatory) of hazardous incidents and near misses would give specific causes of problems whereas the cause of a sinking may always be in question as the vessel is no longer available for inspection. It is possible, however, to estimate the number of unreported flooding incidents overall using other information2. These unreported events range from leaks that are big enough to be remarked upon to flooding that gives everybody a reasonably large fright. The method is presented in Appendix E. The average results for boats over 12 metres in length are: Category (and source) 1 MAIB report: losses due to Flooding 2 MAIB report: flooding at sea 3 Uncontrolled harbour flooding (marine insurance stats) 4 Other flooding incidents (estimate Appendix E)

Boats per year 1 in 210 1 in 60 1 in 115 1 in 20

It can be seen that flooding incidents of some kind are relatively common.3 Many of these, not reported to MAIB, could have had more severe consequences. Our survey of insurance claims supports the rates given above. This study gives a reasonably complete picture of flooding incidents on larger fishing boats. Like many forms of accident/incident there is a broad base of incidents that have been resolved without significant damage to the vessel – albeit the boat may lose fishing time while the flooding is sorted out and temporary repairs are made. About 20% of the flooding incidents, however, 1

MAIB Report on the Analysis of Fishing Vessel Accident Data 1992 to 2000. Issued July 2002. It is important to note that, despite Regulatory requirements for Skippers/Owners to report incidents, the MAIB do not receive reports on all incidents. 2 Banff and Buchan College Report on the Development of a Reliable Bilge Monitor. December 2001. The questionnaire and interviews associated with this work provide a reference for flooding frequency. 3 As far as losses go, they are similar to reported losses for other fleets such as Iceland. The Flooding of Fishing Boats www.banff-buchan.ac.uk

Page 40 of 78

Banff and Buchan College of Further Education

The Universities of Glasgow & Strathclyde

lead to a need for intervention by rescue services and become visible to the MCA. Total Losses due to flooding cost the industry at least £6 million In the year 2001 alone.

SUMMARY OF FINDINGS FROM DATABASES 1)

Every year about one boat in every twenty (1 in 20) suffers a flooding incident that is more than just a leak.

2)

Every year about one boat in every one hundred and fifteen (1 in 115) is flooded because the sea cocks or overboard discharges are not closed.

3)

The MAIB only receive reports on severe floodings and losses. They report that, every year, one boat in every sixty suffers flooding at sea and one in every two hundred and ten (1 in 210) is lost due to flooding.

4)

Financial costs to the industry are high – averaging £4.5M every year and much more in a bad year.

FINDINGS FROM SURVEYS OF ACTIVE FISHING BOATS A typical survey report is attached as Appendix B. These surveys were performed on 40 vessels which included 11 wooden hulled vessels, and 29 steel hulled vessels. The vessels varied from just under a year old through to fifty six years old. Lengths ranged from 16m through to 34m. The sample included scallopers, beam trawlers, vivier crabbers, pair trawlers, twin rig trawlers and smaller pelagic trawlers. Of the 40 vessels surveyed 60% showed defects that could have been a prime cause of catastrophic flooding in port or at sea. Most vessels surveyed revealed at least one hazard that if addressed would lessen the chances of flooding or improve the vessels crews ability to cope with flooding. These figures corroborate the findings of Banff and Buchan College’s previous work which revealed that some 98% of fishermen interviewed had at some time encountered flooding (rather than just leaking) on their vessels. Bearing in mind that on almost all these occasions the flooding was coped with efficiently without loss or major damage. In many cases this was dealt with as a matter of routine and therefore the insurers were not aware of it happening. Deficiencies in pipe work were often initially detected by close visual inspection, a fresh pair of eyes looking at the job with the sole objective of finding faults in the pipe work system. We feel that the fact that the survey team revealed faults has no reflection or slight on crews’ abilities, as we were focusing on one single objective whereas they are usually carrying out a multiplicity of other tasks. The old saying “you can not see the wood for the trees” applies more than ever. The Flooding of Fishing Boats www.banff-buchan.ac.uk

Page 41 of 78

Banff and Buchan College of Further Education

The Universities of Glasgow & Strathclyde

Although it is possible to use experience and knowledge to make judgements as to the areas where thinning problems usually occur, they were sometimes found in the most benign of positions, for instance on long straight pipe runs. The majority of pipe work faults were found in carbon steel. Vessels using copper based alloys pipes revealed few localised problems but showed more general signs of “weight loss” degradation. The survey team, however, were shown samples of these alloys with severe pitting corrosion around the inside of swaged flanges which can not be seen externally or picked by ultra-sonic sampling.

SUMMARY OF FINDINGS FROM SURVEYS The findings of the inspections of fishing vessels include: 1)

25% of all vessels surveyed showed thinning in pipes that was in excess of the 25% limit as per MCA replacement guidelines. Ultrasonic testing is a useful tool for the detection of “weight loss” corrosion.

2)

20% of all vessels surveyed had perforations (pin holes), within their critical pipe work systems.

3)

5% of all vessels surveyed revealed small holes in excess of 2 mm which appeared as a result of preparing the pipe for ultra-sonic testing.

4)

17.5% of all vessels surveyed displayed deficiencies in common cast elbows.

5)

One vessel had localised thinning of her engine room/fish room bulkhead.

6)

5% of vessels had bulkhead penetrations which would allow free flow of water between compartments.

7)

Galvanic corrosion (dissimilar metals) was found on 27.5% vessels.

8)

20% of wooden vessels surveyed showed high levels of earth strapping

9)

47.5% of steel vessels surveyed showed high levels of earth strapping (note: newer vessels are now carrying a higher level of bonding as standard)

10)

12.5% vessels had valves within the piping systems badly corroded

11)

55% of vessels surveyed had deficiencies associated with flexible hoses The Flooding of Fishing Boats

www.banff-buchan.ac.uk

Page 42 of 78

Banff and Buchan College of Further Education

The Universities of Glasgow & Strathclyde

12)

10% of vessels surveyed had deficiencies associated with flexible vibration couplings.

13)

37.5% of all vessels surveyed revealed inadequate supports to pipe work systems.

14)

27.5% of vessels surveyed had misaligned pipe work.

15)

7.5% of vessels had cracked welds in their pipeline systems.

16)

15% had systems that were considerably more complex than required.

17)

12.5 % of vessels revealed problems with valves.

18)

61% of vessels surveyed in the harbour had left their sea inlet valves open

19)

36% of the wooden vessels had appropriate remote closure fittings on their sea cocks.

20)

No steel vessels had a full complement of adequate extended spindles permanently attached to their sea inlets.

21)

85% of bilge alarms were found to be difficult to reach for testing and maintenance.

22)

No vessel inspected was seen to be carrying more than one bilge alarm in the engine room.

23)

Only one vessel was seen to carry a competent description of the fluid systems in a conspicuous place.

24)

Only one vessel (not the same vessel as above) had pumping guidelines displayed openly.

25)

Only 37.5% of vessels had a consistent and useful system for identifying pipe use.

26)

Only 27.5% of vessels had labels on their valves to show their purpose.

The Flooding of Fishing Boats www.banff-buchan.ac.uk

Page 43 of 78

Banff and Buchan College of Further Education

The Universities of Glasgow & Strathclyde

The Flooding of Fishing Boats www.banff-buchan.ac.uk

Page 44 of 78

Banff and Buchan College of Further Education

The Universities of Glasgow & Strathclyde

DISCUSSION, CONCLUSIONS, RECOMMENDATIONS AND ACTIONS DISCUSSION AND CONCLUSIONS In this section of the report we bring together our conclusions about the most critical things that we found. We then make recommendations for the short and longer terms that set the priorities as we see them and suggest actions for the various parts of the industry. We consider that the most critical flooding problems derive from the design, installation, operation or maintenance of raw sea water piping systems. The problems are often fundamental in nature and it seems to us that understanding and information about these systems is not adequate in the industry. In some ways, also our findings may reflect the current situations within the industry which include; Financial constraints within the white fish and shellfish sectors of the industry causing: • o Vessels to operate continuously allowing little time for maintenance. o Crews to leave fishing for other careers o Vessels to sail with minimal crews. o In some cases only essential maintenance to be carried out. •

Varying interpretations of regulations: o That encourage operators to build “rule beater” and “paragraph boats” o A misunderstanding that the four yearly UK Fishing Vessel Certificate is a definitive statement of vessels seaworthiness coupled with a misconception that the MCA recommendations and guidelines are a standard to be achieved rather than a minimum standard to be exceeded where practicable. o Shipbuilders operating in competitive environment are building vessels to be (capital) cost effective within the rules and regulations. o Reflect badly on the knowledge and skills applied to fishing boat systems.

•

A general lack of piping design and layout skills in the marine - not just the fishing - industry.

The deaths and losses that the MAIB have reported and the costs and frequencies of flooding we have calculated should be too high for the industry to bear. We hope that by quantifying the problems we have helped the industry to focus on ways to reduce the toll.

The Flooding of Fishing Boats www.banff-buchan.ac.uk

Page 45 of 78

Banff and Buchan College of Further Education

The Universities of Glasgow & Strathclyde

In reaching these views an important consideration has been that there does not seem to be an age relationship with flooding problems – apart from the general decay of materials. For example, a new boat is just as likely to have a major corrosion problem as an old boat – in fact, the survival of the old boat may be because it has good system. The general uncertainty in design knowledge is also evidenced in the inconsistency in the positioning of components and in the selection of materials for engine room systems. When sister ships were surveyed there were marked variations in the placing of systems onboard each vessel. There was, however, a thread of evidence from the younger vessels surveyed that some of these quality control issues are being addressed. It is clear, however, that over-seeing inspections during construction or in service are not picking up all the possible deficiencies. It seems that understanding of and information about fluid flow systems is not adequately being applied or disseminated in all situations. There is, then, a need to concentrate the thoughts of the design/build teams on flooding and we believe this should be led by the Regulatory body, although all branches of the industry can contribute. A critical evaluation should, therefore, be made, during the initial design and construction stages of a vessel’s life, into the effects that flooding would have on the vessel’s ability to cope with these situations. Such a risk assessment would set the basic parameters for safe operation. We consider this sort of approach essential, but have noted that the present legal position discourages it. In the short to medium term, therefore, we consider there will be a need for MCA to re-focus their surveys to place more emphasis on critical engine room systems and bilge systems elsewhere on the vessels. That will probably (almost certainly) require training and new skills for the surveyors. On the other hand, there is also evidence, from the boats surveyed, that suggests ‘repairs’ are sometimes performed without due regard for materials and not enough thought of potential consequences. The poor and indiscriminate fitting of flexible hoses and couplings is a prime example of this as failure of these fittings could cause the most catastrophic flooding. Immediate emphasis on this problem by all parties would certainly repay the effort in reduced floodings. It is clear that no design makes a boat proof against all flooding, but we found that the majority of the boats surveyed did not have suitable means to close sea inlet valves that would be accessible, should there be any significant flooding in the engine room. If this is taken with the fact the a majority of bilge alarms are positioned so that they are hard to test and maintain then the first line of defence of the boat is breached. It is, perhaps, unfortunate that many of the bilge alarms are sited in positions recommended by the MCA and in some cases by insurance companies. It seems that, in the effort The Flooding of Fishing Boats www.banff-buchan.ac.uk

Page 46 of 78

Banff and Buchan College of Further Education

The Universities of Glasgow & Strathclyde

to achieve optimum placing for earliest warning of ingress, these items are regularly fitted in inaccessible positions and the balance between function as an alarm and functionality of the alarm has been lost. We have concluded that there is a need for a final line of defence against flooding – for those situations that are too severe or are found too late for normal remedial action. We consider that each vessel should carry a suitable pump in a well ventilated area outside the machinery space and that hard plumbed bilge suctions should be fitted to every major space at risk to allow pumping out in critical situations. The scene for minimum standards is, of course set by the Regulations and the 1975 Fishing Vessels Safety Provisions Regulations and associated Guidance to Surveyors were in some instances unclear and ambiguous requiring builders and surveyors to use their own interpretation of the rules and guidelines. Alarmingly, few owners who were interviewed were aware that their vessels were only meeting a minimum and that they were free to modify and improve their vessels. Indeed, many vessel operators were under the impression that their UK Fishing Vessel Certificate was definitive evidence of the vessel’s seaworthiness and that alterations/improvements could only be (and would be expected to be) made by the MCA. Given those beliefs, it is important that the regulations are both concise and precise. The recently published Code of Safe Working Practice for the Construction and Use of 15m (LOA) to Less Than 24m (LOA) Fishing Vessels was analysed and found to be something less than clear. We consider that these Rules require clarification that might be achieved in the interim through the Guidance to Surveyors. We do not apologise for returning to the subject of flexible fittings. Among the most immediately critical defects we discovered was the misuse of flexible fittings such as hoses and expansion pieces. These were frequently used as repairs, but also appeared in fairly new boats to deal with poor pipe fitting. Failure of these would usually lead to rapid flooding. The MCA have pointed out through M Notices that that these fittings should be replaced regularly as a matter of course. Unfortunately there is seldom a date of fitting applied or a date of manufacture impregnated into the fabric and, therefore, operators are seldom aware of the age of the item. We also found that problems were made worse by a lack of information about the fittings that should be used for repair or replacement. To replace a pump by another – off the shelf – without matching capabilities is to modify the whole behaviour of the flow system. We consider that there is a need for transmission of design information through a formal route – a document that is kept on board the vessel describing components and giving specifications. As an example to show it can be done consider Appendix F.

The Flooding of Fishing Boats www.banff-buchan.ac.uk

Page 47 of 78

Banff and Buchan College of Further Education

The Universities of Glasgow & Strathclyde

Among the design flaws we found a common problem concerning sea inlets that makes back flooding into the bilge(s) more likely. We have suggested possible changes, but it is the lack of thought that is most serious. This should be addressed within the Code of Practice and Guidance. Similarly, the problems we found with use of materials are matters for Regulation and the new Code of Practice for 15 to 24 m boats addresses some of them. We have, however, concluded that there is only a limited role for galvanised steel piping in sea water – not because of any theoretical disadvantage, but because, in practice, it seems never to be repaired or maintained correctly. Flow problems were common (62.5%) and seemed to us to be evidence of inadequate thought in design. Too many sharp bends and changes in flow speed leads to erosion of the pipe walls – and in extreme cases not just to leaks, but to full bore failures. Whilst the research detected numerous thin pipe walls, few of these would have resulted in catastrophic flooding provided that bilge alarms functioned properly and gave adequate early warning. That highlights all the more the importance of bilge alarms and the ability to take action to isolate the boat. If the sea inlets can be shut off in time and the pumps operated by someone who knows what they are doing, then in most cases the application of an epoxy bandage would allow the vessel to make the shore and have a permanent repair made. The number of broken pipe supports is, however, of rather more concern. We consider there are two main reasons why we found a significant number of boats with this problem (37.5%). Either, at some time the piping has been disconnected to allow replacement of pipes or other fittings and has not been secured again, or the supports have broken adrift due to vibration and have not been re-fixed by the vessels operators. The fact that broken anchor points are not replaced immediately by the operators may not be deliberate negligence, but rather a matter of casualness and lack of awareness on their part. That is, a sin of omission rather than commission. Usually, when an anchor point comes adrift there is little, if any, damage done at the time. Nothing apparent happens, the equipment carries on functioning as before and it is understandable that repair is put off or even forgotten about. The ensuing vibration over a period of months will, however, gradually affect the pipes causing further damage. Eventually the pipeline may fail catastrophically as vibration induced fractures tend to be sudden and allow ingress at full bore. Many deficiencies discovered during our surveys were probably found simply by “fresh eyes” specifically targeting pipes and flooding related areas. That implies there is a need for the vessel’s own engineer to devote time specifically to these issues in addition to normal operational effort. Unfortunately present day operational pressures may prevent this. The conclusion, however, is that this sort of work must be done – the challenge for the industry is to achieve it at reasonable cost. The Flooding of Fishing Boats www.banff-buchan.ac.uk

Page 48 of 78

Banff and Buchan College of Further Education

The Universities of Glasgow & Strathclyde

One of the techniques we used for inspection was ultra-sonic thickness measurement. This helped us, but requires some thought as to how it could be used to help the fleet as a whole. If this sort of testing was incorporated for long term monitoring in a safety system that logged the original thickness values and if the surveyors had the equipment and training to use it then we conclude that it has very real value. In some surveys video footage and photographs were taken. These were found to be invaluable for review and occasionally allowed previously unseen deficiencies to be detected. That experience leads us to say that it would be of benefit to both owners and insurers to hold video records of vessels that would assist in identifying problem areas and possibly assist when dealing with claims.

RECOMMENDATIONS In the following recommendations an attempt is made to solve immediate problems and then produce systems and a regime that will reduce the number of engine room floodings. The recommendations are a mixture of prescription to set a base level of safety and requirements that are intended to raise the amount of information available and the knowledge of the critical engine room systems. The recommendations are: 1)

In the short term,

1.1

A major effort should be made to inform all sectors of the industry of the hazards associated with misuse of flexible hoses and rubber expansion pieces.

1.2

MCA Inspectors should be informed of the limits of, and correct use of, these items and should enforce removal of hazardous fittings.

1.3

Skippers should (again) be reminded of the potential consequences of failing to close sea inlets and discharges in harbour and Insurers should consider means of persuading operators to close inlets and overboard discharged in harbour.

1.4

All boats 12m and over should be required to fit means of closing sea inlets and discharges from a position that can be accessed when there is significant flooding of the engine room - say 30% of the volume of the engine room

2)

In the medium term,

2.1

All boats 12m and over should have an independently powered pump situated above the main deck level in a position accessible in normal operations that can take suction from the engine room by piping that is permanently installed and independent from other piping systems.

The Flooding of Fishing Boats www.banff-buchan.ac.uk

Page 49 of 78

Banff and Buchan College of Further Education

The Universities of Glasgow & Strathclyde

2.2

All boats 12m and over should be required to carry a document that describes their engine room systems in terms of layout, equipment part numbers and, especially the specification of all piping. There should be a schematic of the engine room systems displayed within the engine room and there should be associated marking and colour coding of pipes and equipment.

2.3

All boats 12m and over should have two independent bilge alarms fitted in the engine room at positions where they can be tested and maintained. [Although out with the scope of this study it is clear that bilge alarms are necessary in any major space such as fish holds.

3)

In the longer term,

3.1

The Regulations should be framed such that: a. The capacity of the bilge pumping systems should be governed by the potential rate of ingress in critical pipe failure and should be defined as a required system flow rate rather than by the existing system of pump capacity. b. Both maximum and minimum flow speeds are defined within system pipe work commensurate with the materials used so as to minimise the effects of erosion and corrosion.

3.2

An inspection regime should be implemented that checks pipes and piping systems and which include sampling by ultrasonic. As part of this, all owners of fishing boats 12 metres and above in length should perform regular inspections of their engine room and other sea water piping systems and arrange an independent audit every two years. Self certification is already accepted by the < 10m fleet.

3.3

A more detailed engine room inspection should be incorporated into the MCA survey by apportioning the surveyors time so that more time is spent in the bilges and less time on equipment check lists. Regular equipment safety checks could be done by the skipper and randomly inspected by MCA if there was sufficient emphasis on deterrents for non-compliance.

3.4

Sea water systems based around a manifold of increased wall thickness should be encouraged as these have proven advantages of simplicity and endurance.

3.5

All new vessels should be designed to reduce the possibility of back flooding from sea inlets to the bilges.

3.6

Use of galvanised steel should be discouraged except for larger pipes and manifolds.

3.7

Best practice for sea water systems should be adopted.

The Flooding of Fishing Boats www.banff-buchan.ac.uk

Page 50 of 78

Banff and Buchan College of Further Education

The Universities of Glasgow & Strathclyde

ACTIONS TO MEET THE RECOMMENDATIONS The Maritime and Coastguard Agency should: 1. Consider carrying out spot safety inspections on fishing vessels specifically to include bilge alarms, sea inlets and remote closure systems for sea inlets. Target time for spot check 30 minutes. 2. Consider how best to ensure that a risk assessment might be carried out on each vessel that includes risks associated with flooding. Once risks are identified and quantified effective action may be taken to reduce the overall risk. 3. Consider setting up a working party to evaluate the Code of Safe Working Practice for the Construction and Use of 15m (LOA) to Less Than 24m (L) Fishing Vessels (and associated Guidance to Surveyors), with particular regard to clarity in the rules for sea inlets, bilge pumping capacities, materials and components. 4. Examine the viability of new builds installing an auxiliary engine and bilge pumping arrangement in a dry space outwith the main engine space. 5. Consider requiring all vessels 12m and over to prepare and carry on board a schedule of piping, pumps and associated equipment which includes material specifications, system drawings and recommended replacement intervals for short life fittings. The intention being to allow transfer of design intent and relevant information through the vessel’s life. 6. Consider requiring all vessels to display within the engine room a schematic diagram of the bilge and cooling water systems to assist the crew. The Fishermen’s Safety at Sea Working Group could: 7. Consider initiating research into: a) the degradation of piping systems and preferred best practice in design and maintenance. b) the value of carrying out regular critical piping system inspections by various means of testing ‘live’ pipework. c) the effectiveness and prioritisation of various emergency systems to counter severe flooding. d) the shock loading and vibration levels in typical fishing boats at sea and the effects on pipework. e) the true endurance of flexible fittings and hoses as they are used in typical fishing boats and methods to improve performance.

The Flooding of Fishing Boats www.banff-buchan.ac.uk

Page 51 of 78

Banff and Buchan College of Further Education

The Universities of Glasgow & Strathclyde

f) the merits of introducing damage limitation in to statutory courses for deck and engineering officers to include suitable content and methods of delivery etc. The UK Fishing Industry should: 8. Consider, where possible and practical, carrying an independent portable pump capable of pumping bilges in an emergency. 9. Consider carrying out regular safety drills which involves all members of the crew in basic bilge pumping and valve isolation operations and use of manual and powered emergency pumps. 10. Make a practice of shutting all sea inlets and discharges when vessels are unattended in harbour. 11. Fit a high bilge level warning system, in a position where it can be readily inspected and maintained, which would operate an external alarm in circumstances where the vessel is unmanned, i.e. when in harbour. 12. Fit two independent (and accessible) bilge alarm systems for engine room protection. 13. Fit at least one bilge alarm in all major spaces within the vessel. 14. Consider installing a “bilge monitor” system which constantly displays levels of water in bilges. 15. Test bilge alarm bilge switches daily where possible and test fish room bilge alarms by controlled flooding during cleaning operations. 16. Where practicable, install close circuit colour television cameras in the engine room. 17. Be aware of the potential for electrolytic action associated with installing low voltage negative earth electrical equipment. 18. Colour code all pipe work for easy and quick identification and label valves stating their purpose. 19. Improve weather tight and non-weather tight door working practice. 20. Apply an appropriate inspection routine in engine rooms.

The Flooding of Fishing Boats www.banff-buchan.ac.uk

Page 52 of 78

Banff and Buchan College of Further Education

The Universities of Glasgow & Strathclyde

Marine Insurance Companies should: 21. Encourage skippers to perform safety checks and drills and include a record of these in a vessel log. 22. Encourage vessel owners to compile video footage and still photographs of vessel’s engine room etc. for ease of visualising components and equipment to assist with potential claims. 23. Consider methods of persuading operators to shut sea inlets and discharges when vessels are unattended in harbour. 24. Carry out safety related spot checks on vessels, to include observation of safety drills, testing alarms and emergency response procedures. 25. Encourage vessel owners to carry portable salvage pumps and other risk reducing equipment such as bilge monitors and engine room cameras. Training Establishments are recommended to: 26. Offer training to fishermen in awareness of flooding dangers and in portable salvage pump operations. 27. Construct courses (certificated) and offer training in basic machinery operation, inspection and maintenance, including valve chests and pumping operations. 28. Offer short Safety and Management courses in Marine Insurance to cover possible implications upon the fishing industry of insurers’ becoming more stringent in their interpretation and application of the Marine Insurance Act clauses, warranties and definitions. 29. Offer technical awareness courses suitable for fishing boat designers and surveyors to cover fluid system design, degradation and inspection techniques. The RNLI could: 30. Consider applying their considerable experience and knowledge by becoming involved in delivery of mandatory safety courses to fishermen.

The Flooding of Fishing Boats www.banff-buchan.ac.uk

Page 53 of 78

Banff and Buchan College of Further Education

The Universities of Glasgow & Strathclyde

APPENDIX A SUMMARY RESULTS CORRELATIONS Correlations show if there are any relationships between variables. For example you would expect age to be correlated with height for teenagers. Correlations were calculated between age and all other variables and between hull material and all other variables. The results are: Positive correlations The only positive significant correlations are: AGE to Comment Hull material The wooden boats sampled were older – as would be expected Average corrosion Average weight loss on older boats was greater than on younger – again as would be expected. Pipe coding and valve tagging Newer boats have better systems – probably because old boats have lost track of the systems. Earthing Newer boats have better bonding systems HULL MATERIAL to Length Pipe coding and valve tagging Earthing Sea cocks open

Comment Steel boats were longer. Steel boats had better systems but the correlation was marginal. Steel boats had (marginally) better bonding systems Steel boats had their seacocks open more often than wooden boats. This, perhaps, represents a greater awareness of flooding in general on wooden boats.

Negative results The most interesting variables that showed no significant correlations with age are: AGE to High hazard Extreme corrosion problems Flow problems Flexible fittings *New: In this context, under 5 years old.

Comment There is no clear age effect on the most significant hazards. New* boats are just as likely to flood as old. This is not confined to old boats. New* boats are just as likely to have severe corrosion and leaks. New boats have as many flow design problems as older boats Poorly fitted hoses and bellows found on both old and new* boats. For the old boats these could be poor repairs.

The Flooding of Fishing Boats www.banff-buchan.ac.uk

Page 54 of 78

Banff and Buchan College of Further Education

The Universities of Glasgow & Strathclyde

TABLES Materials and Corrosion Observation Age (years) 20 10 34 35 35 5 5 13 20 2 5 6 1 36 22 20 2 22 1 28 12 33 34 56 13 3 21 22 15 14 3 2 12 13 35 30 15 21 9 14

Average Max Pipe Hull High Pipe Pipe Cracked Material Galvanic Earthing Corrosion Leaks? support material Hazard? thinning? weight weld? problem? action? problem? % problem? loss % Steel Yes Yes 35 100 Yes No Yes No No Yes Steel Yes No 15 100 Yes No Yes No Yes Yes Steel Yes Yes 45 50 No No No No No Yes Steel Yes No 50 60 No Yes No No No No Wood No Yes 25 45 No No Yes No No No Steel Yes Yes 12 100 Yes Yes Yes No Yes No Steel No Yes 25 70 No No Yes No Yes No Steel No No 20 65 No No Yes No No No Wood Yes No 15 25 No No Yes Yes No No Steel No No 15 30 No No No No Yes No Steel No Yes 25 65 No No Yes No Yes Yes Steel Yes Yes 12 60 No No Yes Yes Yes Yes Steel Yes No 15 20 No No No No Yes No Wood Yes Yes 10 45 No No Yes Yes Yes No Wood Yes Yes 20 100 Yes No Yes Yes No Yes Steel Yes No 25 100 Yes No No Yes No Yes Steel No Yes 15 70 No No No No Yes No Wood Yes Yes 30 66 No No Yes Yes No Yes Steel No Yes 10 28 No No Yes No Yes No Wood Yes No 22 25 No No Yes No No Yes Steel Yes No 15 47 No No Yes Yes Yes Yes Steel No Yes 30 52 No No No No Yes No Steel No Yes 30 57 No No Yes No Yes No Wood Yes Yes 30 100 Yes No Yes No No No Steel No No 10 10 No No No No Yes No Steel Yes Yes 20 57 No No Yes No Yes Yes Steel No No 10 50 No No Yes No Yes No Wood Yes No 20 23 No No No No No No Steel No No 15 66 No No No No No No Steel Yes Yes 25 100 Yes Yes Yes No Yes Yes Steel No Yes 20 64 No No Yes Yes Yes No Steel No No 5 20 No No Yes No Yes No Steel Yes Yes 25 43 No No Yes No No No Wood Yes No 10 25 No No Yes No Yes No Steel Yes Yes 50 100 Yes No Yes Yes No Yes Wood No No 33 100 Yes No No Yes No Yes Steel Yes Yes 50 100 Yes No Yes No No No Steel Yes Yes 25 55 Yes No Yes Yes No Yes Steel Yes No 20 50 No No Yes No Yes No Wood No No 10 10 No No No No Yes No

Steel Yes Yes >40% >50% Yes Yes Yes Yes Yes Yes 72.5% 60.0% 55.0% 10.0% 57.5% 27.5% 7.5% 70.0% 27.5% 55.0% 37.5% Notes Ages > 35 years are not given individually to avoid possible recognition of the boat concerned High Hazard means that there existed a defect or problem that, if left unchanged, presented a significant flooding hazard. Earthing problem means that earth straps were missing from critical locations – not that signs of earth leakage corrosion existed. Pipe support problem means that either there were existing problems caused by poor pipe support or that the surveyor judged that a significant hazard existed. The Flooding of Fishing Boats www.banff-buchan.ac.uk

Page 55 of 78

Banff and Buchan College of Further Education

The Universities of Glasgow & Strathclyde

Flow and Fittings Problems Age years

Hull

20 10 34 35 35 5 5 13 20 2 5 6 1 36 22 20 2 22 1 28 12 33 34 56 13 3 21 22 15 14 3 2 12 13 35 30 15 21 9 14

Steel Steel Steel Steel Wood Steel Steel Steel Wood Steel Steel Steel Steel Wood Wood Steel Steel Wood Steel Wood Steel Steel Steel Wood Steel Steel Steel Wood Steel Steel Steel Steel Steel Wood Steel Wood Steel Steel Steel Wood

Flex Flow Bellows Bends Poor hose problem problem problem alignment problem ? ? ? ? ? Yes Yes No No No No Yes No Yes Yes Yes Yes No No Yes Yes No No No No Yes No No No No Yes No No No No No No No No No No No No No No Yes Yes No No No No No No No No Yes No No Yes No Yes No Yes Yes Yes Yes No No Yes No No Yes No No Yes No No No No Yes Yes Yes No No Yes Yes No No No No Yes Yes No No Yes Yes No No No No Yes Yes No No No Yes Yes Yes Yes Yes Yes Yes No No No Yes Yes No No No No No No No No No No No No No Yes Yes No No Yes Yes No No No No Yes Yes No No No Yes Yes Yes No No Yes Yes Yes No Yes No Yes No No No No No No Yes No Yes Yes No No No No Yes No No No No Yes No Yes No No No No Yes No Yes Yes No No No No Yes No No Yes Yes Yes No No No No No No No No

Yes Yes Yes Yes 62.5% 55.0% 10.0% 20.0%

Yes 27.5%

Pipes Valves Remote Corroded Poor coded tagged closure valves? Access? ? ? ? No No No No No No No No No No Yes No No No No No No No No No No No No No No No Yes Yes No Yes No No No No Yes No No No No No

Yes 12.5%

Poor Poor OK Poor Poor Poor Poor Poor Poor OK Poor Poor Poor Poor Poor Poor OK Poor OK Poor Poor Poor Poor Poor Poor Poor OK Poor Poor Poor OK Poor Poor Poor OK Poor Poor Poor Poor Poor

No No Yes No No No Yes Yes No Yes Yes Yes Yes No No No Yes No Yes No No No No No No Yes Yes No No Yes Yes Yes No No No No No No Yes No

No No No No No No Yes Yes No Yes Yes No Yes No No No Yes No Yes No No No No No No Yes Yes No No No Yes No No No No No No No Yes No

No No No No No Yes No No Yes No No No No No No No No Yes No No No No No No No No No No No No No No No No No Yes No No No No

Sea Cocks open? No On slip Yes On slip No On slip On slip Yes Yes On slip Yes No Yes No Yes Yes On slip No On slip No On slip No Yes Yes On slip Yes On slip Yes On slip Yes On slip On slip On slip No Yes No On slip Yes On slip On slip

Yes Yes Yes Yes Poor 37.5 27.5 10.0% 60.9% 82.5% % %

Notes Flow problem means that there existed a defect that would cause pipe degradation that stemmed from poor design of (or repair that affected) the system flow characteristics. Pipes coded/valves tagged means that the system was not sufficient for purpose. Remote closure applies to the inlet isolation and means that such a system did not exist or could not be used for some reason. One boat had high inlets & did not need remote closure. Sea cocks open 17 boats were on the slip with systems being maintained. The percentage given here is of the boats that were afloat when inspected. The Flooding of Fishing Boats www.banff-buchan.ac.uk

Page 56 of 78

Banff and Buchan College of Further Education

The Universities of Glasgow & Strathclyde

APPENDIX B FREQUENCY AND COSTS A survey by Banff and Buchan College1, found that 28% of senior crew surveyed had experienced uncontrolled flooding at sea and a further 17% when in port. Of the same group, moreover, 98% had experienced flooding that had been controlled, dealt with efficiently and professionally with little detriment to the vessel. These figures may be used to obtain an estimate of unreported flooding incidents overall. The MAIB average rate for serious flooding of boats of all lengths in the interval 1992 to 2000 is 0.042 incidents per boat per year. For boats greater than 12 metres in length the incident rate is 0.015 per boat per year and the total loss rate 0.005 per boat per year. Taken with the average number of boats licensed in the same period, the number of boats affected on average in a year is estimated. If we concentrate on the >12 m boats we can ratio the events using the Banff and Buchan figures. The underlying assumptions are that: i)

the Banff and Buchan interviews have sampled typical fishermen

ii)

any of these fishermen has the potential to be involved in either a controlled or uncontrollable flooding

iii) the average number of crew per boat doesn’t change over the years. On these bases we assume that the serious floodings reported to MAIB and the uncontrolled floodings reported by fishermen to Banff and Buchan are the same things and then we ratio to unreported, less serious floodings. Using all this information, the expected flooding proportions, for a fleet of 1000 boats in an average year, are shown on the figure overleaf. In addition, it should be noted that uncontrolled harbour flooding would affect 9 boats in a thousand in an average year.

1

Banff and Buchan College Report on the Development of a Reliable Bilge Monitor. December 2001. The Flooding of Fishing Boats www.banff-buchan.ac.uk

Page 57 of 78

Banff and Buchan College of Further Education

The Universities of Glasgow & Strathclyde

930 BOATS NO PROBLEMS 5 BOATS SINK

15 BOATS NEED HELP

50 BOATS FIX PROBLEM WITHOUT HELP

FLOODING PROPORTIONS PER 1000 BOATS

The Flooding of Fishing Boats www.banff-buchan.ac.uk

Page 58 of 78

Banff and Buchan College of Further Education

The Universities of Glasgow & Strathclyde

APPENDIX C TYPICAL REPORT Document Use. This document is purely for research use to assist the Fisherman’s Safety at Sea Working Group – Flooding of UK Fishing Vessels Study: and it is not in any form binding upon vessel owner(s) to carry out recommendations offered in the report. Strathclyde University and Banff & Buchan College accept no responsibility or liability as regards these findings and recommendations. Confidentiality. All findings will be given to the owner or their authorized agent(s). Further disclosure can only be by their permission. All names of people and vessels will remain strictly anonymous and will not be revealed to any third party including the sponsors. Ultrasonic Sampling: No samples of the original pipe work were available. This means that the extent of thickness loss has been estimated by sampling an area of the same system that should not be prone to general thinning. New pipe bends and elbows have been sampled and gave thickness variations of between 1% and 12%. New straight pipes were within 2% of nominal. This uncertainty must be factored in to any practically determined values from the vessel. MCA guidelines state that a pipe with a reduction in wall thickness of 25% should be considered for replacement. Although this is an exact value it is only determinable by starting ultrasonic testing in the construction phase and keeping an accurate chronological record thereafter.

VESSEL DETAILS Steel vessel built in Scotland. Age 17 years Length 25m. Approximate gross tonnage, 160. Soft nose stem, rounded bilge and transom stern. Hull subdivided from forward into chain locker, forecastle, fishroom, engine room and cabin. Non-watertight full-length shelterdeck Galley etc situated on main deck with wheelhouse above. Main engine: engine – Caterpillar 3412, 600Hp @ 1800rpm Auxiliary engines: Volvo Penta Skipper owner. Type of fishing: Whitefish and prawn trawling (twin rig) Fishing pattern: 7-day landings. 5 crew

The Flooding of Fishing Boats www.banff-buchan.ac.uk

Page 59 of 78

Banff and Buchan College of Further Education

The Universities of Glasgow & Strathclyde

Abstract Galvanised steel pipe mostly sampling at less than 25% thinning. Aluminium brass pipe wall thinning found to be 25% loss where sampled at main engine seawater discharge cooling pump. Porous pipe by way of the port auxiliary sea suction. Weeping welds in discharge of bilge pump (starboard). Weeping weld at port side bilge pump discharge. No valve tagging or colour coding. No engine room layout schematic. 3 Instances of misaligned pipes - Two sections of overboard discharge pipes were severely misaligned. Engine Room The main engine is a Caterpillar 3412 developing 600Hp@ 1800rpm and it is located within a central well between 2 wing tanks. The total volume within this well minus the main engine and associated plant is 16m³. The generators are located above to port and starboard. Introduction Engine room pipe work was inspected for wall thickness in places deemed likely to fail through corrosion, erosion or vibration and in general the pipe thickness was not in a poor condition. Systems The sea water system had the following sea inlets: ‘2 off’ - 3” sea suctions for the main engine ‘2 off ’ - 1 ½”’’ Sea inlets serving the ‘2 off ‘ alternators ‘2 off ‘ - 3’’ Sea inlets serving the bilge pumps port and starboard and the valve chest. Other small sea inlets for services. The system has an inherent weakness in that if the bilge valve leaks the vessel will experience back flooding from the main sea inlets if these are open to the bilge line valve chest – as is normal on board. Indication of this will be the constant operation of whatever bilge pump is used. Conversely, this will also seriously hamper the pumping ability of the vessel, as the leaking bilge valve will draw air when empty. In practical terms this could mean the engine room bilges will be always partially full thereby reducing the time to combat a serious flood or ingress of seawater into the engine room before the level reaches the 3 phase alternator. The bilges and sea valves would be difficult to reach in a flooding situation, as there are no extended spindles. The overboard valves are hard to reach and the fitting of extended spindles to these valves would be difficult. The Flooding of Fishing Boats www.banff-buchan.ac.uk

Page 60 of 78

Banff and Buchan College of Further Education

The Universities of Glasgow & Strathclyde

Design The design of the engine bilge system was sketched and the layout examined. The practice of tagging valves has been missed and there is the possibility of a mistake due to the complicated array of valve positions which if not memorised could lead to confusion and delay in an emergency. Practice The practical aspects of bilge pumping requires closure of valves and cocks when alongside - the complexity of the present system, in the hands of unfamiliar personnel, could lead to non-operation of these components for fear of mix-up and subsequent loss of pumping capability. All shipside valves should be closed when plant is shutdown alongside. Ultrasonic Sampling: Areas of main seawater suction pipes were sampled at points likely to experience severe turbulence and erosion: Seawater to the main engine and general service pumps enters the vessel via 6” diameter stub trunking. This trunk wall gave ultrasonic thickness to be abnormally thick – approximate external measurement from the dry-dock showed the internal diameter of this trunking to be approximately 3 inches, indicating that the sounding was reasonable. Cooling water to the main engines, gearbox and refrigeration is supplied through aluminium brass piping, as is raw seawater piping to the starboard valve chest. Most other piping is of galvanised steel. A seawater pipe (galvanised) was found porous at inlet to the port auxiliary engine cooling system. Sampling of aluminium brass pipes to both main engine and showed little evidence of wall thinning at inlets, although the main engine discharge pump had 25% reduction in thickness. Both port and starboard valve chests may be suffering from severe external corrosion. Ultrasonic sampling was attempted on the strainer casings but no results could be obtained – this usually indicates pitting internally. The casting was observed encrusted in rust with a black colouration in several places. A discharge pipe from the starboard bilge pump shows weeping at several welds with a general thinning of some 25% was found in this pipe. The port side valve discharge chest, supplied with seawater via a 3” galvanised steel pipe and this was found to be in reasonable condition. All other galvanised steel pipes sampled at less than 25% reduction in thickness. The engine room/fish room bulkhead sampled between 5.4 mm by way of 1 metre from keel and 6.5 mm at 1.5 metres.

The Flooding of Fishing Boats www.banff-buchan.ac.uk

Page 61 of 78

Banff and Buchan College of Further Education

The Universities of Glasgow & Strathclyde

Flexible joints/sections. Several flexible sections were sighted. These were inserted within the cooling piping arrangements to compensate for vibration. The sections were either of reinforced bellows type construction or straight hose. In all cases the flexible sections were applied with type approved clips with a minimum of two at each end. Two sections of suction pipes were severely misaligned. In particular the 3” emergency line.

FINDINGS Aluminium brass pipe wall thinning found to be no greater than 25% where sampled. Galvanised steel pipe mostly sampling at less than 25% thinning. Porous pipe by way of port valve chest discharge. Colour coding was painted over in many places. Weeping weld by way of first bend after port side bilge pump. Several instances of misaligned pipes with misalignment being taken out by flexible pipe hose; one of which showing signs of fretting. Three overboard flanges missing nuts on bolts holding valves to ships side.

RECOMMENDATIONS: Consider fitting extension spindles to all bilge sea suction valves. Align pipe sections on 3” overboard discharge (starboard side) by way of flexible pipe section – use a purpose made spooled flange before inserting flexible expansion joint if necessary. Replace porous pipe in suction of port auxiliary seawater cooling pump Replace section of pipe from port side bilge pump. Replace 3” galvanised steel pipe from starboard bilge pump discharge to overboard. Replace missing nuts on overboard discharge valves (3) at ships side (port). Consider monitoring thickness of pipe walls by having ultrasonic samples taken at regular intervals. Consider having aluminium brass pipe from main engine discharge pump renewed. Consider constructing and affixing piping diagrams and bilge pumping instructions to engine room bulkhead. Consider changing auxiliary galvanised pipe work to aluminium brass. Shut all non-essential sea cocks and ships side valves in harbour, when vessel is unmanned. Consider fitting a duplicate bilge alarm sensor in engine room Consider fitting externally sited warning light or klaxon to bilge alarms for harbour use. Consider repainting engine room pipe work with approved colour codes for respective services.

The Flooding of Fishing Boats www.banff-buchan.ac.uk

Page 62 of 78

Banff and Buchan College of Further Education

The Universities of Glasgow & Strathclyde

APPENDIX D TYPICAL INTERVIEW CRITICAL PIPE WORK AND SYSTEMS/ PROCEDURES INTERVIEW Information given in this interview will be treated as confidential. The names of people and vessels will be strictly anonymous and will not be revealed to any third party, including the sponsors. If it is felt that any question puts you in a compromising situation, simply answer, “pass”. 1

Vessel age

8 years old

2

Hull construction material

Steel

3

W.F.

5 6

Fishing type e.g. white fish Fishing method e.g. pair trawl Fishing routine e.g. weekly Number of crew

7

Your job

4

Pair trawl Weekly, 1½ crews 4 Engineer

Crew turnover (1=low; 4 5=high) Duration onboard of 3 years Driver/Engineer

8 9 10

Driver/ Engineer qualifications

11

General history of vessel: Ownership Burst pipework Re-engining Comments

None 2nd Yes – not uncommon Original Flooding due to mis-alignment of pipes into cooler causing erosion and in one case cracked pipe. Found during visit to engine room for other reason. Bilge alarm sometimes not spot on! About 5 real leaks in bilge piping in my time and replaced same section twice (galvanised steel) due corrosion/perforation) ME cooler replaced – that’s where we got problems.

The Flooding of Fishing Boats www.banff-buchan.ac.uk

Page 63 of 78

Banff and Buchan College of Further Education

12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37

The Universities of Glasgow & Strathclyde

Is a planned maintenance routine in place for: Pipework None – sort of check it out when I can Valves None – when operated Seacocks Monthly to work valves Strainers Monthly with valves Mud boxes Weekly Anodes; Monthly Pipes Annually Hull Pumps None – if they stop! Is a regular visual inspection routine in place for: Pipework Yes Valves Yes Seacocks Yes Strainers Yes Mud boxes Yes Anodes in pipes; On engine monthly Yes, annually On hull; Pumps Yes – regular check Are seacocks closed in harbour, if Yes vessel is unattended for more than 12 hours? Is a sign displayed indicating No but getting one painted seacocks shut? Is pipework colour coded? Painted - e.g. green for sea water Is a pipework diagram displayed No – would be handy in the engine room? Are suction valves clearly marked Some for ease of use? Is a pumping diagram/directions No displayed in engine room? Have you ever encountered Yes leaking water pipes? If “yes”, what caused the Corrosion and misalignment problem? What onboard repair material is Belzona bandage, rubber gasket carried for potential pipe leaks? sheets, Bandit system ; jubilee clips etc repairs ongoing. Are pipe repairs ever carried out Yesby welding or brazing on this Replacement when ashore vessel? Have water pipes ever been Yes – same pipes more than once replaced on the vessel? If the vessel has been re-engined, N/A for main engine but main were cooling pipes also renewed cooler replaced at the same time? The Flooding of Fishing Boats

www.banff-buchan.ac.uk

Page 64 of 78

Banff and Buchan College of Further Education

The Universities of Glasgow & Strathclyde

38

Has backflooding through valves Yes, bilge in harbour – why we ever been a problem on this boat? shut seacocks

39

Have you encountered flooding Cooler corrosion from other sources? If “yes” what was the source? Corrosion & electric leaks Annual non destruction testing of Can you suggest pipework. Galvanised steel piping anything which would improve bilge pumping seems to be sub standard. systems and Annual testing for electrical leaks. pipework reliability?

40 41

The Flooding of Fishing Boats www.banff-buchan.ac.uk

Page 65 of 78

Banff and Buchan College of Further Education

The Universities of Glasgow & Strathclyde

APPENDIX E EXAMPLES OF REGULATION ANALYSIS In this Appendix, sections of the Code of Practice for Fishing Vessels between 15 and 24 metres in Length that was published in 2001 are analysed as an example of how the Regulations might be clarified therefore by and leading to more effective guidance. The analytical technique is not new in itself, and has been well reviewed and extended by John Lawson in a recent PhD thesis from Aberdeen University. The aim is to grammatically analyse the text and identify verb clauses, therefore isolating these as clear instructions to take account of information, perform calculations or do things in some way thus removing ambiguity. In the example sections below, the original rule is in bold the discussion in normal font and proposed new rules in bold italics.

SCUPPERS, INLETS AND DISCHARGES 2.2.6

Scuppers, Inlets and Discharges

2.2.6.1 The number of inlets and discharges should be kept to the operational minimum. N This consists of 2 rules (a) Discharges to be minimised (b) Inlets to be minimised This is the first mention in the Code of any of these items so they are undefined. The intention is that openings for these purposes below the upper freeboard deck should be minimised, but operational minimum is fairly hazy as a concept. A reference to redundancy would be useful so we propose this should be reworded as 2.2.6.1 The number of openings through the hull below the upper freeboard deck used for inlets and discharges should be the minimum number that meets requirements for redundancy of supply or discharge. N ___________ 2.2.6.2 Each scupper or discharge leading through the hull from spaces below the freeboard deck or from within an enclosed superstructure or deckhouse on the freeboard deck should have an automatic non-return valve fitted at the hull with a positive means of closure from an accessible position. N

The Flooding of Fishing Boats www.banff-buchan.ac.uk

Page 66 of 78

Banff and Buchan College of Further Education

The Universities of Glasgow & Strathclyde

This consists of 2 rules and a modifying clause to define where they apply. The rules apply to scuppers and discharges from water or weather tight spaces leading through the hull and they are, c) scuppers and discharges to have an auto non return valve fitted at the hull to prevent backstop -flow of sea into the vessel d) scuppers and discharges to have a means of closure that can be operated from an accessible position when the non-return valve is not functional. Both rules are ambiguous because hull is undefined. The meaning is ‘hull below the upper freeboard deck’ because no other section of hull requires this protection. Much of the definition of the spaces is redundant as it is clear that it is only watertight or weathertight spaces that require such discharges. Drainage from open decks to discharges below the freeboard deck is covered in 2.2.6.7 where an unspecified increase in wall thickness is required. See comments at that rule. The position of the NRV is not clearly defined with respect to access for operation or maintenance. The need to maintain these notoriously frequently unreliable valves in fact conflicts with the phrase at the hull. There is a need for an isolation valve near the hull and a non-return valve inboard of that. This allows maintenance of the NRV. Combining isolation with the NRV makes for complexity and inefficiency. with NRV makes for poor reliability. Remote closure from an "accessible position" is also not clear, as 'accessible is not defined either in terms of location or for ease of operation. It could usefully read ‘positioned above the freeboard deck and within the normal working area of vessel, located so that it is easy to use.’ If, however, this were accepted then a dispensation for engine room overboard discharges would be required. Either the isolation valve or the NRV could be subject to the remote closure requirement, but it is would be best to make this the isolating valve – again because of the unreliable nature of NRV’s. It would also be necessary to define certain requirements for the closing apparatus and in particular, a way of knowing that the valve is closed! It is also necessary to define certain requirements for the closure – particularly that there is a way of knowing that the valve is closed! The engine room dispensation would allow the remote closure to be from a location within the engine room but readily accessible after significant flooding of the space. This could apply to any machinery space, but it would be necessary for the space to be protected by a bilge alarm. Significant flooding would be a level that would be expected to remove critical bilge pumping capacity – from our studies about 30% of floodable volume. The rule might then be written as: The Flooding of Fishing Boats www.banff-buchan.ac.uk

Page 67 of 78

Banff and Buchan College of Further Education

The Universities of Glasgow & Strathclyde

2.2.6.2(a.1) Each scupper or discharge leading from a water or weather tight space through the hull below the freeboard deck should have an isolating valve fitted near the hull. This valve should be capable of local control at the valve and also be fitted with a positive means of closure that can be operated from a position above the freeboard deck and within the normal working area of vessel, located so that it is easy to use. Indication of valve position should be provided at the remote closure position. N 2.2.6.2(a.2) Notwithstanding the requirements of 2.2.6.2(a.1), if the discharge is within a machinery space that is fitted with a bilge alarm and serves machinery within that space, then the remote means of closure may be operated within the machinery space from a position that is readily accessible after significant flooding of the space. As a guide, significant flooding may be taken as more than 30% of the floodable volume of the space. N. 2.2.6.2(b) Each scupper or discharge leading through the hull below the freeboard deck from water or weather tight spaces should be fitted with an automatic non-return valve fitted inboard of the isolation valve to prevent backstop -flow into the vessel. N _____________ Inlets 2.2.6.3 Each sea inlet valve should be fitted with a positive means of closure from an accessible position. N 2.2.6.4 In machinery spaces, controls for main and auxiliary sea inlets essential for the operation of machinery may be controlled locally. The controls should be readily accessible, above the floor plates, and be provided with indicators showing whether the valves are open or closed. N 2.2.6.3 is a simple rule that but is ambiguous because accessible is undefined. It also confuses local control at the valve with remote closure. 2.2.6.4 is a modifier to 2.2.6.3 and should be attached to it as with scuppers and discharges above. All the comments on rule 2.2.6.2 also apply here. The rule should read, 2.2.6.3(a.1) Each sea inlet valve should be capable of local control at the valve and fitted with a positive means of closure from that can also be operated from a position above the freeboard deck and within the normal working area of vessel, located so that it is easy to use. Indication of valve position should be provided at the remote closure position. N The Flooding of Fishing Boats www.banff-buchan.ac.uk

Page 68 of 78

Banff and Buchan College of Further Education

The Universities of Glasgow & Strathclyde

2.2.6.3(a.2) Notwithstanding the requirements of 2.2.6.3(a.1), if the sea inlet is within a machinery space that is fitted with a bilge alarm and serves machinery within that space, then the remote means of closure may be operated within the machinery space from a position that is readily accessible after significant flooding of the space. As a guide, significant flooding may be taken as more than 30% of the floodable volume of the space. N. ______________ 2.2.6.5 If valves are not fitted above the floor plates, rapid and practical means should be provided to allow for the valve to be operated from floor plate levelIf inlet valves are not fitted above below the floor plates, rapid and practical means should be provided to allow for the valve to be operated from above floor plate level when more than 30% of the floodable volume of that space is reached. . E AND

2.2.6.9 Existing vessel arrangements will continue to be acceptable provided that valves fitted at hull penetrations remain both accessible and efficient in service. E Existing vessel will have to meet the requirements of 2.2.6.5 for inlets however present arrangements will continue to be acceptable for scuppers and discharges provided that valves fitted at hull penetrations remain both accessible and efficient in service. E Rule 2.2.6.9 should be combined with 2.2.6.5 as they both refer to existing vessels. Rule 2.2.6.5 is intended to bring existing vessels into line with respect to isolation. It stands alone with, perhaps, an implicit understanding that it covers sea inlets valves because of the position of the rule in this section. It is not particularly clear which valves are covered, but it is all the valves within a machinery space that can shut off the vessel from the sea or isolate critical systems that are intended – not just valves below floor plates. In fact, floor plate level, while obvious as a way to define access, has no good relationship to flooding. In many cases of engine room flooding, the floor plates will be quickly submerged. This rule should be defined in terms of volume of flooding as with the other rules above. Rule 2.2.6.9 allows existing vessels to continue with the same systems, but with a rather weak requirement for access and efficiency. . Ideally Tthe rules should read, 2.2.6.5(a) Valve arrangements in existing vessels will continue to be acceptable provided that valves fitted at hull penetrations can be shown The Flooding of Fishing Boats www.banff-buchan.ac.uk

Page 69 of 78

Banff and Buchan College of Further Education

The Universities of Glasgow & Strathclyde

to be readily accessible for local operation and for maintenance and are regularly checked for operation. E 2.2.6.5(b) Notwithstanding the requirements of 2.2.6.5(a), valves within a machinery space that control or isolate flow of sea water are to be fitted with rapid and practical means of closure that can be operated from a position that is readily accessible after significant flooding i.e.30% of the space. E ________________ 2.2.6.6 Soil and other waste water drainage should be so arranged and fitted with such water seals, air vents and storm valves as are necessary to prevent siphoning, blowback or ingress of water. The hull closing arrangements should be as detailed in paragraph 2.2.6.2. N No change required. ____________ 2.2.6.7 If scuppers from open decks penetrate the hull below the freeboard deck they should be made from piping of substantial thickness. Allowing such scuppers is poor practicenot ideal, but may be unavoidable in some cases. Although this rule is relatively clear, the word substantial is undefined. The intention is to defeat corrosion by providing a margin on wall thickness. This is not sufficient on its own and a table of pipe wall thickness for given pipe diameters and material should be referenced. The rule should be, 2.2.6.7 Scuppers from open decks should not be led below the freeboard deck unless no other drainage route is practicable. Where a scupper from an open deck penetrates the hull below the freeboard deck it should be made from piping of the thickness given in Table X. _________________

2.2.6.8 Refer also to paragraphs 4.1.11 (seawater systems), 4.3.2.10 and 4.3.2.11 (bilge systems) and 11.1.3 (pollution). No change required. _________________

COOLING WATER AND OTHER SEAWATER SYSTEMS A typical rule analysis taken from the section in the recent Code of Practice that deals with cooling water and other seawater systems is given below. The Flooding of Fishing Boats www.banff-buchan.ac.uk

Page 70 of 78

Banff and Buchan College of Further Education

The Universities of Glasgow & Strathclyde

In the example sections below, the original rule is in bold the discussion in normal font and proposed new rules in bold italics. 4.11.1.1 All new or replacement installations of sea water piping and fittings for cooling water systems should be of aluminium bronze, cupro-nickel or similar corrosion resistant material. The intention is to ensure that all sea water piping and equipment that contains sea water is of material that resists corrosion. An admirable intent, but it must be clear and also coupled with advice on galvanic action and on allowable flow rates. There is a great deal of confusion in the industry about the various causes of corrosion with more apparent emphasis on avoiding stray currents than on correct use of materials. Aluminium brass is probably intended for piping. Aluminium bronze is not a suitable piping material, but would be correct for valves and other such items. The requirements for fittings should be separated from the requirements for pipes as these will include valves, and could include polymeric fittings. For example, rubber impregnated with a carbon filler can cause significant damage to aluminium or copper piping. It is not clear why replacement installations are specially mentioned and it causes confusion. The term or similar corrosion resistant material is imprecise. The intention should be to allow either a new material of proven corrosion resistance or an alternative lower quality material with increased wall thickness to accept greater thickness loss. The latter is not preferred, but may be an effective strategy in some cases. For new boats it should be clear that all sea water systems are to be of suitable materials. For existing boats the replacement in a repair should take account of the cause/need for the repair and should be both compatible with existing materials and achieve the corrosion performance in sea water of aluminium brass. There should, therefore, be a requirement for the vessel to carry a schedule of all fittings to allow compatible repair. A corrosion resistant standard is not defined – a schedule of suitable materials could be given in the associated guidance. Limiting flow rates should be quoted. The rule might then read,

The Flooding of Fishing Boats www.banff-buchan.ac.uk

Page 71 of 78

Banff and Buchan College of Further Education

The Universities of Glasgow & Strathclyde

4.1.11.1(a) All sea water piping should be of aluminium brass, cupro-nickel or a material that has been clearly shown to provide similar corrosion resistance in sea water. 4.1.11.1(b) Associated fittings and equipment should be of compatible materials of equivalent corrosion resistance selected or arranged so that galvanic corrosion between dissimilar metals or between a metal and a polymer is avoided. 4.1.11.1(c) It should be demonstrated that flow rates in all normal operational conditions do not fall below or exceed those given in tableXXXX. 4.1.11.1(d) A schedule should be prepared and kept on board the vessel giving the details of all pipes, fittings and equipment used in the sea water systems. _______________

4.1.11.2 ‘Heavy wall’ mild steel pipe for ‘cross vessel’ inlet mains may be used, provided that the internal diameter is 100mm or greater and the pipe is galvanised internally after all fabrication work is complete. This is a dispensation from 4.1.11.1 and this should be noted. It is intended that a sea inlet manifold may be of galvanised mild steel provided that an additional wall thickness allowance is applied and the galvanising is correctly applied. It is implicit that the flow rate in the manifold will be lower than in the daughter systems and that a larger diameter pipe is easier to galvanise or regalvanise than small diameter pipes. The term Cross vessel’ inlet mains is unclear and should be supplemented with or inlet manifold. There should be a cross reference to the regulations covering scuppers, inlets and discharges. It is very important for such manifolds that suction exits from them be smooth reducers rather than pipes set directly onto pipes. In addition, the flow rate in normal operation should not fall below a given minimum. The Regulation might then be, 4.1.11.2

Where a ‘cross vessel inlet main’ or ‘sea water inlet manifold’ is used then,, notwithstanding the requirements of 4.1.11.1(a) above, mild steel piping will be considered, provided that: i) The internal diameter is 100mm or greater, ii) The pipe is galvanised internally after all fabrication work is complete iii) The wall thicknesses are not less than those given in TableYYYY. The Flooding of Fishing Boats

www.banff-buchan.ac.uk

Page 72 of 78

Banff and Buchan College of Further Education

iv) v) vi)

The Universities of Glasgow & Strathclyde

Turbulence is avoided where pipes are taken off the manifold by the use of smooth reducing sections. It is demonstrated that flow rates in all normal operating conditions do not fall below 1 m/sec. Full attention is paid to the avoidance of galvanic corrosion at the junctions between the mild steel pipe and other pipes or fittings of dissimilar materials in the vessel. ___________

4.1.11.3 Care should be taken to ensure that galvanic corrosion effects from dissimilar metals are prevented, by such means as isolation packing, washers and sleeves between the flanges and fasteners joining pipes. Almost redundant, but worthwhile to have a catch-all clause here to mention such things as pipe supports. It is also necessary here to bring in the concept of bonding of similar materials. Reword as, 4.1.11.3(a) Care should be taken to ensure that galvanic corrosion effects from dissimilar metals are prevented, by such means as isolation packing, washers and sleeves between the flanges and fasteners joining pipes and in way of pipe supports. 4.1.11.3(b) Care should be taken that where similar materials are separated by a non-conductive material there is electrical bonding across the junction. ___________ 4.1.11.4 Recommendations may also be found in MGN 190 (F): Fishing Vessels – The Premature Failure of Copper Pipes in Engine Cooling Water Systems. No change to this. ___________ 4.1.11.5

Sea water pipes, wherever practicable, should be connected by means of bolted flanges, visible and readily accessible for maintenance and inspection purposes. Existing vessels should be fitted with such arrangements whenever seawater pipework is renewed. E

The phrase wherever practicable is weak and should be removed. It can be replaced by a dispensation for other arrangements that meet the requirement for ease of access and integrity assessment. The phrase visible and readily accessible is also unclear. Visible from where? The requirement is better expressed as a functional one that says The Flooding of Fishing Boats www.banff-buchan.ac.uk

Page 73 of 78

Banff and Buchan College of Further Education

The Universities of Glasgow & Strathclyde

the connection must be capable of close inspection and repair without removal of items of equipment other than floor plates. The rule then becomes, 4.1.11.6(a) Sea water pipes should be connected by means of bolted flanges. Alternative connections will be considered, but only if they provide the same ease of integrity assessment and replacement as a bolted flange. . Existing vessels should be fitted with such arrangements whenever seawater pipework is renewed. E

4.1.11.6(b) Pipe connections should be capable of close inspection and repair without removal of items of equipment other than floor plates ___________ 4.1.11.6 Where cooling water services are essential for the cooling of the propelling machinery, alternative means of circulating water should be provided in the event of failure of the primary source. Such alternative means should be demonstrated to the satisfaction of the Certifying Authority. And 4.1.11.8 New vessels should be fitted with at least two main seawater cooling inlets, with one inlet fitted on each side of the vessel (except when fitted with ‘keel cooling’ arrangements). N These rules requires redundancy of cooling water supply to propulsion machinery – and presumably to any other machinery critical to the boat’s survival. 4.1.11.8 is primary as it deals with the main cooling system and asks for sea inlets port and starboard on the vessel. The intention is to provide inlets on either side to avoid fouling of any one side. This can be generalised to allow for high and low sea inlets – rather than only P&S. Rule 4.1.11.6 then says that “alternative means of circulating water should be provided in the event of failure of the primary source.” This is not clear – it may mean loss of the means of circulating (pumping) or loss of supply – for example, by fouling of the inlet. It should cover loss of supply, distribution (pipes) and circulation (pump). This means that the port and starboard (or high and low) main inlets cannot, themselves function as the alternative supply. The implication is that at least The Flooding of Fishing Boats www.banff-buchan.ac.uk

Page 74 of 78

Banff and Buchan College of Further Education

The Universities of Glasgow & Strathclyde

three inlets are require – the two primary inlets and another that may be already used for some other purpose. This is also discussed in 2.2.6.1 where any overall requirement is for the minimum number of inlets that meets redundancy needs. Reword as, 4.1.11.6(a) There should be two main sea water inlets – suitably separated to take account of fouling and to provide redundancy in location of supply through the vessel’s hull. 4.1.11.6(b) In addition, where cooling water services are essential for the operation of the propelling machinery and for the operation of any other machinery critical to the vessel’s survival, alternative cooling water supplies should be provided to the machinery. These alternative supplies should be from a sea inlet that is independent of the usual cooling water supply and should employ independent pumps and piping to the machinery – for example, by a cross over from another sea water system. Such alternative means should be demonstrated to the satisfaction of the Certifying Authority. 4.1.11.6(c) Where keel cooling arrangements are fitted, the alternative cooling supply in 4.1.11.6(b) is not required. ___________ 4.1.11.7 Sea water suctions of cooling systems essential for internal combustion machinery should be provided with strainers suitably arranged so that they may be cleaned without interrupting the supply. This should be more general in that all sea water inlets should have easily accessible strainers. In addition, when a sea inlet comes out of the water the pipes can drain to cause an air inlet in the system. The sea inlet valves, therefore, should be screw down non return valves. It may be that this rule would be better placed under inlets, but for the moment it is retained in this section as, 4.1.11.7(a) All sea water inlets should be provided with strainers suitably arranged so that they may be cleaned without interrupting the supply. 4.1.11.7(b) Where there is a possibility of the inlet coming out of the water due to seas and motion of the vessel, the sea inlet valve should be a non return valve to avoid draining of the line and formation of an air lock. ___________ The Flooding of Fishing Boats www.banff-buchan.ac.uk

Page 75 of 78

Banff and Buchan College of Further Education

The Universities of Glasgow & Strathclyde

4.1.11.8 New vessels should be fitted with at least two main seawater cooling inlets, with one inlet fitted on each side of the vessel (except when fitted with ‘keel cooling’ arrangements). N See above under 4.1.11.6. Delete. ___________ Add new Rule to account for MGN 49 that says do not fix flexible sections of piping in cooling water or other systems unless necessary to withstand movement or vibration. 4.1.11.8 Flexible hose and other flexible fittings should not be used in cooling water or other fluid systems) unless necessary to withstand movement or vibration. The fittings must be suitable for the conditions of use and be clearly marked with a part number or other description and the date at which they must be replaced. ___________ 4.1.11.9 Refer also to Section 2.2.6 (Scuppers, Inlets and Discharges) Retain ___________________________

The Flooding of Fishing Boats www.banff-buchan.ac.uk

Page 76 of 78

Banff and Buchan College of Further Education

The Universities of Glasgow & Strathclyde

APPENDIX F EXAMPLE OF INFORMATION BOOKLET TYPICAL LIST OF MATERIALS Item Nº 1 2 7 8 9 10 11

Description

Details

Material

N. Bore

Port Bilge Suction Valve Chest Starboard Bilge Suction Valve Chest Diesel Engine Pump Suction

S.D.N.R – 5 ways S.D.N.R – 5 ways

CI/GM

80NB

CI/GM

80NB

Butterfly Valve

CI

80NB

Diesel Engine/Gilkes Pump Dis Diesel Engine/Gilkes Suction Diesel Engine/Gilkes Suction Diesel Engine/ Gilkes Inlet

Wafer type N.R Valve

CI

80NB

Flexible Joint

80NB

Flexible Joint

80NB

Butterfly Valve

CI

80NB

Comments

Bilge/Fire/ Deckwash Pump Bilge/Fire/ Deckwash Pump Bilge/Fire/ Deckwash Pump Bilge/Fire/ Deckwash Pump Bilge/Fire/ Deckwash Pump Bilge/Fire/ Deckwash Pump

12

Diesel Engine/ Gilkes Outlet

Wafer type N.R Valve

CI

80NB

13

Port Bilge Pump Discharge Starboard Bilge Pump Discharge Fire / Deckwash Isolating Valve Engineroom Emergency Suction Deep Well Emergency Suction Port Engineroom Suction Starboard Engineroom Suction Deep Well Emergency Discharge SW/FW Emergency Cooling

S.D.N.R + test Cert, S.D.N.R + test Cert, Butterfly Valve Whale 4” DK fitting Whale 4” DK fitting Foot Valve C/W Strainer Foot Valve C/W Strainer

GM

80NB

GM

80NB

CI

80NB

Brass

100NB

Hand Pump

Brass

100NB

Hand Pump

GS

50NB

GS

50NB

14 16 21 21 29 30 36 43

Straight Thro’ Storm Valve

50NB

Blanked Off Spectacle Flange

65NB

The Flooding of Fishing Boats www.banff-buchan.ac.uk

Page 77 of 78

Banff and Buchan College of Further Education

The Universities of Glasgow & Strathclyde

Example Vessel Piping Schematic

Note: This vessel has a bow thruster

The Flooding of Fishing Boats www.banff-buchan.ac.uk

Page 78 of 78