PART B

ASSOCIATED FACTS AND

SEPARATE INVESTIGATIONS

 

CHAPTER 33

BOW DOOR FAILURES AND OTHER INCIDENTS OF RO-RO VESSELS

Following the ESTONIA catastrophe the governments of the countries involved - Estonia, Finland and Sweden - initiated a number of investigations concerning

(1) the national organisations with responsibilities for own vessels according to national law and international conventions;

(2) the national organisations with responsibilities for foreign vessels under the Port-State-Control (PSC) regime;

(3) the activities of the international Classification Societies.

First of all, however, all car/passenger ferries with bow doors were closely inspected as soon as possible whenever they arrived in a Swedish or Finnish ports. It is unknown whether this procedure was also carried out in the Estonian ports. From Sweden it is only known that reportedly 28 vessels were found to have serious defects to their bow-door arrangements.
From Finland the following report was received:

»The bow and stern doors of all the Finnish passenger/Ro-Ro cargo/ ferries have been inspected by the maritime inspectors of the Finnish Board of Navigation (F.B.N.).
Two vessels were stopped for repairs. The locking arrangements have been welded into closed position on board the ferries which run between Helsinki-Stockholm, Helsinki-Travemünde and Vaasa-Sundsvall, and also on board the ferries which are cruising in the Northern Baltic.
A special Port State Control Survey has been carried out on board all the passenger ships which are regularly trading to the Finnish ports. Captain Veikki Inkinen, Chief Maritime Inspector of the Gulf of Finland Maritime District, informed that two main categories of damage to the locking devices were found:

1. Fatigue fractures.
2. Wear and tear.

Cracks and fatigue fractures in the lugs of the locking devices and wear and tear failures in the bolts. Some of the bolts were 78 mm undersize (max. acceptable 3 mm) due to extensive wear. Also corrosion damages were found.
The inspectors found damage to almost every ship's bow visor's locking devices.
Maritime Inspector Tom Sommerdal surveyed about 10 vessels in Helsinki. Also he found wear and tear damages in the lugs and bolts of the locking devices of the bow visors. Also leakages in the hydraulic systems were noted.
The vessels under Estonian flag were in worse condition than those under Finnish flag.«

In addition, the Swedish government instructed already on the 03.10.94 the previous Supreme Court Judge Magnus Sjöberg to carry out an internal investigation of the National Maritime Administration for Shipping and Navigation - Sjöfartsverket - and, in particular, its Sjöfartsinspektionen - The Shipping Inspection (Shipinspec). This investigation covered among other things the subjects 1) to 3) stated on the previous page and was performed on basis of a Government decision defined as follows:

»On 28 September 1994, the passenger ferry Estonia foundered south-west of Utö in Finland en route from Tallinn to Stockholm. The vessel was flying the Estonian flag. On 29 September 1994, the Government commissioned the Board of Accident Investigation to assist in the accident inquiry. A joint accident board was then formed, with representatives from the Estonian, Finnish and Swedish authorities. One of the main functions of this board is to establish the reason for the accident.
Survivors of this maritime disaster have stated that the vessel's bow "visor" was torn away, which is supposed to have been an important cause of the sinking.
Information has since emerged to the effect that incidents and problems with bow doors have occurred previously with ferries of the same type as the Estonia. A high standard of safety at sea demands that the necessary conclusions are regularly drawn from incidents and accidents that occur. In addition to the investigation now being carried out by the joint accident board (JAIC), other matters should be dealt with including the reporting routines within the Shipinspec.
The Government commissions former Judge of the Supreme Court Magnus Sjöberg to carry out such a survey. As part of this, the investigator is to examine how information relevant to maritime safety is analysed and promulgated within the department, and where necessary put forward proposals for improvements. The investigator is also to look at the options for and the frequency of exchange of relevant maritime safety information between maritime safety authorities and other bodies that have important functions in the field of maritime safety, such as classification societies, insurers and shipowners' associations. Finally, the investigator should examine how such information exchange takes place between Shipinspec or corresponding bodies within our local region, and where necessary put forward proposals as to how this might be improved.
The investigator shall carry out his work in contact with the JAIC and in co-operation with the special committee on increased maritime safety for ferry services that the Government will be setting up.
The Government authorises the Minister of Communications to appoint a secretary or other assistant to the investigator.
The work of the investigation shall be completed with the utmost dispatch.«

The complete "Report on Maritime Safety Work by The Shipping Inspection" by Magnus Sjöberg is attached in the Swedish Original and in English translation as Enclosures 6.3.116 and 6.3.116.1. It comprises 58 pages and deals exclusively with the conditions existing before and at the time of the ESTONIA accident and among other things the investigator has closely looked into the history of bow door failures and associated incidents as they were reflected in the various files of the different Sjöfartinspektionen offices - not at a central register - of which some shall be explained below:

(a) The car/passenger ferry "VISBY"
The car/passenger ferry VISBY was built in 1972 and had a gross tonnage of 6665 registered tons. On 12 November 1973 on its ordinary run between Nynäshamn and Visby, the forebody of the ship was struck by heavy seas. The result was that the bow door opened and water was shipped on the car deck. The master therefore decided to turn the vessel back to Nynäshamn. This event took place after the vessel had come out into the open sea. There was a high wind and heavy sea at the time. On survey of the damage it was found that both of the locking hooks of the bow door had been broken off and indentations were left on the door. As a temporary measure the bow door and cargo ramp were welded to the hull. After the repair had been water-tested and surveyed by the relevant Classification Society, Lloyds Register of Shipping, the vessel received temporary permission to continue trading until the next yard inspection. In February 1974 permanent repairs were carried out, in which the bow door received new locking hooks of heavier dimensions than previously, and the locking device was further reinforced by shroud screws with lashing attachments.
The Shipinspec's conclusion about the cause of the accident, as stated in the investigation report dated 1 November 1974, was that the bow door locking devices were of too weak construction.

(b) "STENA SAILER"
The Ro-Ro cargo ship STENA SAILER was built in 1973 and had a gross registered tonnage of 2872. On a voyage between Zeebrügge and Dover on 16 January 1974 with a cargo of trailers and transport lorries, the vessel encountered heavy weather with increasing opposing seas. The vessel reduced its speed, but the strong sea broke apart the locking devices of the forward bow door, and lifted it out of closed position. The space between the bow door and the bow ramp was filled with water. The ramp was therefore secured using six chains. In order to avoid additional damage, the vessel was hove to and subsequently Rotterdam was called on as port of refuge where a temporary repairs were carried out. The locking mechanism and hoisting arms of the bow door had been damaged, and the port side ramp hydraulic actuators had become unusable after the failure of the operating rod attachment. The investigation report of Shipinspec dated 12 December 1974 found that the cause of the accident was the took weak constructions of the locking devices of ramp and bow door.

(c) "SVEA STAR"
SVEA STAR, a car/passenger ferry built in 1968 with a gross registered tonnage of 9963, was registered with Lloyd's Register of Shipping. Between Travemünde and Helsingborg on 5 May 1974, the vessel encountered heavy weather and rising seas. A particularly heavy sea struck the vessel, breaking the forward locking devices and extra securings and lifting up the bow door. Water collected between the forward door and the ramp. The vessel returned to Travemünde for damage inspection.
Upon survey it was found that all three hydraulic locking devices were completely damaged, and even the extra securing with two 2-inch turnbuckles had been torn apart. No damage was evident to the actual door or ramp. The locking devices were temporarily repaired and surveyed two days later in Helsingborg by both the Classification Society and the Department.
The Shipinspec investigation report dated 12 December 1974 states that in the opinion of the master, of the Classification Society and of the Department the accident was due to the too weak construction of the locking devices of the bow door.

(d) "SAGA STAR"
The car/passenger ferry SAGA STAR, with a gross registered tonnage of 8226, was built in 1981. On 6 May 1982 the vessel was about to sail from Travemünde to Helsingborg. The weather was calm. When the bow visor was closed the port hinge failed which caused the port side of the visor to fall down four meters. The hydraulic hinges broke and oil leaked out. Immediately afterwards the starboard hinge failed too and the entire visor fell down . After having been surveyed by Lloyd's Register of Shipping, the vessel was given a permit to proceed to Helsingborg - via Malmö - without its bow visor.
Upon survey damage to the visor and its attachments to the hull, and hull damage between ramp and visor was ascertained. It was stated in the Shipinspec investigation report that the accident was probably due to the visor hinges having been of insufficient dimensions.

(e) "STENA JUTLANDICA"
The car/passenger ferry STENA JUTLANDICA was built in 1983, with a gross registered tonnage of 15 811. On 12 October 1984, STENA JUTLANDICA was berthed at Fredrikshamn. When the bow door was opened, the port hinge of the bow door failed and the bow door fell down a bout 2 metres. After having been lifted back into position, the door was welded in place. A permanent repair was carried out on a later occasion. When the bow door was inspected, cracks were found in the hinge attachment welds. The attachment plate was also partly cracked. According to the Shipinspec investigation report the cause for the hinge failure was the under-dimensioned welding seams of the hinge attachments which led to crack propagation.

(f) Certain other accidents to ro-ro vessels
The accident on 16 March 1987 involving the HARALD OF FREE ENTERPRISE led the Shipinspec immediately to carry out a review of all Swedish flag ferries from the stability point of view. Letters were also written to the owners with questions about such matters as their routines for operation and checking of bow and stern doors.
During this investigation details were also obtained of previous bow door damage to Finnish car/passenger ferries. For example, it was stated that the Finnish vessel FINLANDIA sustained damage to her bow door in 1981 which also led to a modification of her sister ship SILVIA REGINA, later STENA SAGA. It was also revealed that the Finnish car-passenger ferry MARIELLA suffered very severe visor damage in 1986 - see also further in this chapter - which led to structural changes of her and the visor of her sister ship OLYMPIA, then still under construction. During the investigation of the above events it turned out that the bows were designed with too much flare, which had resulted in extreme bow stresses.

As a result of the disaster involving the ESTONIA, one of the actions taken in autumn 1994 was a specific inspection of passenger ferry bow visors by the Gothenburg Shipinspec office. Of the twelve ships examined within the Gothenburg area, eleven showed deficiencies. These ranged from the formation of cracks in the locking devices to operating system deficiencies. One vessel, the LION PRINCE, was not permitted to sail, while heavy weather restrictions were imposed on certain other vessels.

Finally, the following should be noted in this context. The mass media reported that the ESTONIA's sister ship, DIANA II, nearly lost her bow visor between Trelleborg and Rostock on the same stormy night in January 1993 when the Polish ferry JAN HEWELIUSZ sank. However, no so-called Section 70 report was submitted to the Department's Investigation Section as required by the existing regulations. Nor was this event, as far as it is known, brought to the attention of the central authority in any other way until after the loss of the ESTONA. The Department subsequently submitted the question of the master's conduct for possible prosecution, but the prosecutor decided not to bring an action.

The DIANA II as well as the also above-mentioned MARIELLA incident shall be explained in more detail later on in this chapter. Before, however, the reaction of the administration on the SAGA STAR accident shall be outlined as follows:

After survey the chief inspector of the Malmö Shipinspec office, Åke Sjöblom, decided on 17 May 1982 that SAGA STAR could sail without the bow visor until the end of May, provided that the weather was good, that the voyages were undertaken at reduced speed, that due attention was paid to the other factors which might affect the seaworthiness of the vessel and that the log extracts were submitted to the Shipinspec after each voyage.
It was further decided that complete drawings, together with calculation material for "bow door", hinges and hydraulic arrangements should immediately be submitted to the Shipinspec. Following a survey on 17 May 1982 Lloyd's Register of Shipping awarded an Interim Classification Certificate in accordance with the Shipinspec's decision.

The new bow visor was surveyed and approved on 14 June 1982.
The Classification Society submitted the requested calculation material, and the then Safety Section of Sjöfartsverket drew up its own highly comprehensive memorandum with computation sheets. This memoran-dum is dated 14 July 1982 and was intended to form the basis of discussions at a safety meeting. The memorandum proposed that the Administration should determine how various forces are to be calculated. There are also references to the accidents referred to above, and the fact that these led to changes in the rules of the Classification Societies. It was alleged also that the calculations submitted by the Societies indicated inadequate engineering input on the calculation and design side, and it was alleged that the Classification Societies did not monitor this aspect sufficiently. It was mentioned for example that the bow doors of the VISBY and a vessel named WASA STAR subsequently had to be strengthened. The author of the memorandum expressed a desire for further Classification Society and yard calculation material, and proposed that the Administration should take measures to prevent further accidents of the kind in question.

According to the file the matter was finalised by the then acting Sea Safety Direction of Sjöfartsverket who wrote to Lloyd's Register of Shipping on 21 September 1982, criticising the Society's calculations and superinten-dence. The letter concludes as follows: "The Administration requests that you test the suggestions made regarding the present version and indicate the reinforcements that you consider necessary. In view of the fact that the locking devices of bow visors broke open in bad weather on a number of vessels, the Administration further requests you to account for the dimensioning of the locking devices of this vessel."

Apparently the SAGA STAR accident had triggered off even more activities within the Administration as reflected in the letter that the then Safety Director, Per Eriksson, wrote on 15.03.84 to Lorenzo Spinelli of the Italian Classification Society RINA - see Enclosure 33.419. It reveals that the strength expert, Gunnar Hjertstedt, of the National Maritime Administration (Sjöfartsverket) had written a memo after the cause investigation following the SAGA STAR accident and after the subsequent discussions with Lloyd's Register, of which page 1 is available and where among other things it is stated:

»The correspondence and discussions with the classification society involved in this problem have shown that they consider the opening and closing arrangement of doors as lifting appliances exempted from their surveys, except on special request of the owner, while the other scantlings of the door and the locking arrangement are subject to their control.
To this procedure the Swedish Administration of Shipping and Navigation (Note: Sjöfartsverket) has the following objection. Details involved in the opening and closing of bows are integrated parts of the doors, which we find difficult to separate from the regulations concerning securing and the problem to calculate and establish the scantlings of other details of the door. If the control in this case is separated to different authorities the hazard of missing important control objects is great, which has been demonstrated in this case. The Administration of Shipping and Navigation would suggest that this problem is discussed within IACS.«

The memo had apparently been written sometime in 1982, but in March 1984 the responsible safety director was obviously not even certain whether he had passed it on to "dear Lorenzo" when being reminded by his strength expert before his retirement, which is another demonstration of the attitude of the responsible authorities in those years. The only page available of this memo is attached together with another page from another letter or memo signed by Gunnar Hjertstedt but without date referring to the "closing and securing of doors" as Enclosure 33.420.

After each of the above-mentioned accidents exclusively concerning bow doors of car/passenger ferries or Ro-Ro cargo vessels which were only 1 year or less old - with the exception of the SVEA STAR having been 8 years old - Shipinspec and/or Sjöfartsverket only investigated against the Classification Societies, mainly Lloyd's Register but also Bureau Veritas and Norske Veritas. The Building Yards are not even mentioned and the Classification Societies were blamed for took weak construction and/or under-dimensioned locking devices and hinges.

This is, of course, remarkable to note bearing in mind that the head of the Investigation Section of the National Maritime Administration (Sjöfarts-verket) attended all the meetings of the JAIC investigating the ESTONIA casualty, be it the official full Commission meetings or the internal meetings of the Swedish part of the JAIC where it had obviously been decided already shortly after the catastrophe to put most of the blame on the Building Yard.

Finally the report of the Gothenburg office of the Shipinspec concerning the STENA SAILER accident shall be quoted as follows:

»On 2 May 1974, a report was received from the Gothenburg Shipinspec office of a meeting dealing with examination of bow door locking devices. From this report it emerged that STENA SAILER's bow door had come open during a previous voyage. It also appeared that the bow door of STENA SAILER's sister ship, the UNION WELLINGTON under the New Zealand flag, despite having been reinforced, had been opened during heavy weather. Another sister ship, the SEA TRADER, had crossed the Atlantic in bad weather, but on this vessel the bow door had been welded closed. The following may also be found in the report. "During the meeting, it emerged that very little is known about the forces that affect the bow door. However, through the trials conducted by G.L. (Note: by Germanischer Lloyd), some slight under-standing of this has been gained. It appeared that more or less all the locking devices on existing vessels with bow doors are of too weak construction. That matters have gone as well as they have so far is due to the fact that vessels with bow doors are mostly used in restricted trades where they do not normally meet such heavy weather. I propose that the Administration investigates how locking devices on bow doors ought to be placed, designed and constructed, and that there should then be a check on the locking devices of existing vessels. This presumably will not be done by the Classification Societies, since after all these have already approved the existing locking devices."«

Note: Special attention has to be drawn to the obvious fact that this proposal was made already in May 1974 by the chief inspector of the Gothenburg office to the National Maritime Administration - Sjöfartsverket.

The report of Magnus Sjöberg on the STENA SAILER continues:

»On 8 May 1974, the Department wrote to Bureau Veritas, with which STENA SAILER was classified, and sent the society the same memorandum and questions as had formerly been submitted to Lloyd's Register of Shipping in connection with VISBY and SVEA STAR. Bureau Veritas was given the chance to respond to the questions in the memorandum and to give its views on the accident in question. It was then asked what maritime safety measures Bureau Veritas intended to undertake in connection with the bow door accidents that had happened.
Bureau Veritas replied at the end of July 1974, and this response indicated inter alia that the society intended to make its requirements for bow door locking devices stricter.«

Note: When DIANA II and VIKING SALLY (ESTONIA) were designed and built in 1978-1980 there were no requirements at all in the Bureau Veritas Rules then in force (1977 Rules). This reveals from a letter that the BV head office wrote to the JAIC on 10.01.95 where it is stated among other things: »The bow door drawings of the MV ESTONIA were reviewed by the Hamburg local office of Bureau Veritas which was in charge of the review. List of reviewed drawings is given as Annex 4. « »The drawings were checked against the Bureau Veritas 1977 rules which did not contain formula for the rule strength of securing devices of shell doors opening outwards.«

The complete letter is attached as Enclosure 2.4.3.37. See also comments in Subchapter 2.4.3.

So much for the Swedish investigations publicly known. The results were certainly one of the reasons why Safety Director Bengt-Erik Stenmark had to resign from his position and was replaced by Johan Fransson.
It is unknown whether a similar investigation was carried out in Finland with the Finnish Board of Navigation - then Sjöfartsstyrelsen and now renamed Sjöfartsverket. Certainly on first view the Finns would even have had more reasons to look into the performance of this organisation which, after all, had issued wrong "Passenger Ship Safety Certificates" for VIKING SALLY, SILJA STAR, WASA KING - the three previous names of the ESTONIA while trading under the Finnish flag. It is also unknown whether a similar investigation was carried out in Estonia.
In any event, in the course of the investigation by this 'Group of Experts' some further bow door incidents concerning car/passenger ferries in the Baltic have come to light which are worth being explained, viz.

VIKING SAGA now REGINA BALTICA
Built for AB Sally - simultaneously with VIKING SALLY - at the Turku Shipyard in 1980. On 20.10.84 the vessel was on her scheduled voyage from Helsinki to Stockholm and encountered a south-westerly wind of 14 m/s, i.e. Bft. 6-7, when at around 23.00 hours she was shaken hard by the sea, coming from ca. 3 point on port side. The alarm of the locking device system was not activated. In Stockholm, however, heavy structural damages in way of the port visor side were found and temporary repairs were carried out. During the permanent repairs, performed later, it was recommended to reinforce the visor. For particulars see the survey reports attached in original and in office translation as Enclosure 33.421. Classification Society: Det norske Veritas (DnV).

MARIELLA
Built for Rederi AB Slite at Turku Shipyard and delivered in March 1985. On 07.11.85 the ferry was on her scheduled voyage from Helsinki to Stockholm, when at around 23.00 hours she was hit in the front by an uncommonly high sea from port which shook the vessel hard. Locking devices and hinges of the visor broke, the visor was lifted up and remained connected to the ship only by the visor actuators. Due to the enormous pressure, exerted on the pistons during thrusting up the visor, the pistons were widened and the pressure lessened. As a result thereof the visor came down again and the ingress of a large quantity of water onto the car deck was only prevented by the immediately initiated full-astern manoeuvre. The ship continued to Stockholm on very slow speed. The Classification Society of MARIELLA was and is Det norske Veritas. At the time of this incident the vessel was still under guarantee. It was considered to exchange the heavily damaged visor with the identical visor of her sister ship OLYMPIA which was still under construction at the Turku Shipyard, but it was subsequently decided to repair and convert the visor on site while the ferry kept trading and during which time the loading/discharging operations were carried out exclusively via the stern ramps. The repairs, which also included the locking devices being considerably reinforced, required about 5 months.
This incident, which was probably the most serious visor damage of a car/passenger ferry in the Baltic area before the ESTONIA catastrophe, led to considerable discussions in the shipping industry. The fact that DnV participated in the inspections/meetings with three Finnish representa-tives and one Swedish representative underlines that the problems were realized and that it was clearly considered to be a Class matter, the F.B.N. was not involved. After analysing the events it was agreed that the immediate reduction of the speed and the turning of the stern into the sea had finally prevented the loss of the visor and the subsequent probably unavoidable catastrophe. - See the Inspection Report Enclosure 33.422.
The MARIELLA - same as the ESTONIA - is equipped with a visor/bow ramp combination, i.e. the upper part of the ramp extends into a deck house which is part of the visor. In addition to this the vessel is equipped with a 2.50 m high and moveable collision bulkhead on the car deck.
As a consequence of the MARIELLA accident the locking devices for the bow visor of her sister ship OLYMPIA, delivered in April 1986, were totally changed upon the instructions of the Swedish Sjöfartsverket - inspector Lennart Ahlberg. The owner, Gustav Myrsten, was reportedly personally handling the changes and reinforcements of the locking devices.
Subsequently the OLYMPIA was chartered by P&O and used in liner service between Southampton and Bilbao. At the same time the Classification Society was changed from DnV to Bureau Veritas (B.V.) and P&O ordered the installation of eight additional locking devices in view of the regular crossing of the Bay of Biscay.

SAINT PATRICK II
Built for SF-Line, the third Viking Line partner, in 1973 as TURELLA by Sietas Werft, Hamburg. She is a passenger/Ro-Ro cargo/ferry of 7984 gross tons and 1325 dwt, built to Lloyd's Register Class Rules and supervision and is classed + 100 A1, Ice Class I, Ice Class IA, which corresponds to Finnish/Swedish Ice Class Rules 1971. The vessel is fitted with a bow visor and ramp for vehicle access at the forward end. At the aft end there is a stern door/ramp. The mean draught is 5.35 m, while ESTONIA's draught was about 5.5 m. Information was received from Helsinki about damage to the visor of this ferry having been in bare-boat charter for Tallink (ESCO) from September 1993 to May 1994 between Helsinki and Tallinn. During this time the ferry experienced the following:

- on 11 January 1994 - The bow visor was frozen in position above and below the Atlantic lock, which was the modification fitted in 1982 consisting of a hydraulic locking arrangement between the forepeak tank top below the bow ramp position and the bottom of the visor. The visor could not be opened in Tallinn and the vessel was discharged by the stern. The visor's lower area was heated with steam hoses to clear the ice and then released.

- on 18 February 1994 - The starboard side bow plating, below the visor and in way of the Atlantic lock space, was found to be heavily set in and the visor could not be opened. After discharge of vehicles, on the early morning, the vessel went to the Baltic Shipyard, where repairs were effected to the starboard bow. See the attached photo documentation. After this incident on 18 February, extreme care was taken when navigating in ice conditions, with the vessel being trimmed by the stern to keep the visor higher above the waterline.

- on 29 March 1994 - The bow visor was found to be jammed and upon examination it was noted that the port side bow shell plating, in way of the Atlantic lock position, had been heavily set in, similar in nature to the damage noted on the starboard side on 18 February 1994. Immediate temporary repairs were carried out to allow the vessel to continue her service. On 4 April 1994 it became necessary to carry out some additional welding to the temporary repair on the port side. Upon subsequent dry-docking it was found that extensive damage to the shell plating on both sides had been sustained whilst the bow visor was damaged as follows:

- Bow lower edge plating, port and starboard buckled and set in and visor void tank possibly breached.

- Visor locking arrangement open position, port side fractured and starboard side missing. Visor displaced slightly to port.

A photo documentation with explanation is attached as Enclosure 33.423.
The experience of the SAINT PATRICK II as explained above confirms the observations by passengers on board the ESTONIA during the ice winter 1994 and the respective damage to visor and locking devices found subsequently.