THE CONDITION OF "ESTONIA" AND THE STATUS OF HER MAIN CLASS AND SAFETY CERTIFICATES BEFORE COMMENCEMENT OF HER LAST VOYAGE
The vessel participated in the "Continuous Class Survey System" for Hull and Machinery (CSH/CSM) of Bureau Veritas and was inspected in regular intervals.
The 3rd period of class for hull and machinery was running from 3 July 1990. Therefore, the next due date for completion of this Class Renewal III would have been 15 July 1995. The last drydocking was in Turku/Finland in January 1994, when inter alia the periodical bottom survey was carried out and the stabilisers were installed as per Hull Survey Report STK/94/3 A - see Enclosure 16.1.205. The last annual survey was carried out from 23rd to 25th August 1994 while the ferry was in service. No remarks about deficiencies affecting the class were noted in any of the reports by the attending surveyor Anders Wirstam - see Enclosure 16.1.206 - Survey Statements STK/94/77A-F dated 25.08.94 - see also the previous Chapter 15.
As part of this survey the following items were checked and apparently found to be to the surveyor's satisfaction:
Since the ferry had been taken over by her new owners, Estline AB - c/o Nordström & Thulin, Stockholm - in January 1993, Bureau Veritas was also authorised by the Government of Estonia to issue Safety Certificates pursuant to the following conventions (specified abbreviated):
LOAD LINE 1966
Thus B.V. issued also the following certificates on behalf of the Estonian National Maritime Administration allegedly after having performed the respective surveys according to the "Survey Report for Issue of PSSC" dated 27.01.94 - attached as Enclosure 16.1.207.
|Passenger Ship Safety Certificate||issued||Stockholm||27.01.94|
|Interim- Enclosure 16.1.208||valid until||27.06.94|
|Passenger Ship Safety Certificate||issued||Copenhagen||23.06.94|
|- Enclosure 16.1.209||valid until||26.01.95|
Note: Although it should have been issued "Interim" it is a normal unlimited/ unrestricted Certificate issued by the Copenhagen office of B.V. and a completely different form from the one used by the Stockholm office before.
|International Load Line Certificate||issued||Stockholm||11.04.94|
|Interim - Enclosure 16.1.210||valid until||11.09.94|
Note: According to the Report of the JAIC a new "Interim" Certificate had been issued by B.V. on 09.09.94 which, however, is not available and apparently was also not available to N&T in October/November 1994, because it is not mentioned in the letter of S.C. Forsberg to the criminal police dated 01.11.94 - see Enclosure 5.2.110.
|International Oil Pollution Prevention||issued||Stockholm||15.03.93|
From the available certificates it has to be concluded that
Note: The reasons for the L.L. Certificate having been issued "interim" only were
(a) the "temporary" stability booklet which was not approved by B.V. (on behalf of the Estonian Board of Navigation);
(b) the too small cross-flooding valve between the heeling tanks.
The reason why the PSSC was issued wrongly was the fact that the vessel did not comply with the requirements of SOLAS 1974 for the "intended trade", i.e. proceeded more than 20 nm from the nearest land, as stated annually in the "Survey Report for the Issue for PSSC - Clause 1 - General Information" and, thus, the bow ramp plus front bulkheads, as upper extension of the collision bulkhead above bulkhead deck = car deck, were ca. 5.8 m too much forward (see Subchapter 2.4.4 above).
The last issued L.L. Certificate had already expired on 11.09.94 and apparently a further interim one had not been issued by B.V. for reasons unknown to this 'Group of Experts'. In any event, according to the certificates issued after the surveys carried out between the 23-25 August 1994 by Anders Wirstam, the Load Line Survey was apparently carried out and Annex A - to Report No. STK/94/77E - see Enclosure 16.1.206 - was filled out although, as it looks, no new Interim L.L. Certificate had been issued by the time of the casualty.
This was, at least, the situation in October/November 1994, since it has to be assumed that N&T would have submitted the L.L. Certificate valid at the time of sinking, had they had it. However, when the Final Report of the Joint Accident Investigation Commission was published, it was most surprising to note that they stated on page 46, item 3.6.2 that there had been a valid L.L. Certificate issued on 9 September 1994. In any event, also this certificate was issued falsely because visor and bow ramp had been leaking for many months before the casualty.
Based on the status of the class and safety certificates only, it could be assumed that the vessel ESTONIA was in an acceptable condition when she departed from Tallinn on 27 September 1994. This, however, was definitely not the case as will be demonstrated in the following subchapters.
Condition of the Bow Area
Based on the information contained in the previous chapters 3.5 and 12 regarding the vessel's performance until 27 September 1994, in particular the various observations of passengers, crew members and other persons on or from the car deck or other decks of the ferry, it has to be concluded that
(1) the visor, its locking devices (hinges) and rubber packings;
(2) the bow ramp, its locking devices, hinges and rubber packings had been in a very bad condition respectively were non-existent when the ferry commenced her last voyage on the evening of 27 September 1994.
This conclusion is in direct contradiction to the statements made by Bureau Veritas in the many certificates issued by them in respect of all the safety relevant parts of the vessel and it is indeed the question on what basis these certificates have been issued.
Before the observed deficiencies shall be explained in detail the relevant parts of the foreship, which played a crucial role in the casualty sequence-of-events, shall be shown below and on the following page again for easy reference.
In addition, these deficiencies are shown on a drawing of the foreship with open visor which is integrated into this report and forms the following page.
(1) The visor, its hinges, locking devices and rubber packings:
The problems with the visor were initiated by two different circumstances, viz.
- severe structural damage caused by proceeding through heavy ice at excessive speed.
- the decision of the Technical Managers, N&T, not to renew the rubber packings, assuring weather(water-)tightness of the visor and preventing the visor from vibrating and thereby avoiding cracks emanating therefrom.
The shell plating as well as the internals are very severely damaged in two places, viz. there is a deep vertical indentation in the bow leading upwards from the stempost for about 2 m, as can be seen on the picture below from the outside,
The arrow points to the deep indentation which is completely painted and the paint does not show contact marks. The following photos showing the inside of the same area do confirm the deep indentation in a very strong part of the visor to be of old origin.
Furthermore, the shell plating as well as the 3rd stringer at the port side are severely indented and set up with the frames buckled which had been sustained already some time before the casualty. - See the pictures above.
The condition of the visor bottom with stempost is evident from the picture below taken on 30.08.94.
Arrow 1 - Indentation in the foremost part of the visor bottom indicating that the visor's initial position was changed to forward most probably due to misalignment of the whole structure. This led to contact between the visor bottom and the stempost on the forepeak deck. Since the visor bottom in this area is not constructed to take up load, the bottom became indented and cracked.
Arrow 2 - Indicates a cracked welding seam through which water penetrated and created the rust stripes.
Arrow 3 - Indicates transverse cracks in the visor bottom in way of the contact area with the starboard steel pad welded to the forepeak deck and which should take up about 10% of the visor's weight in closed condition. Since the corresponding port side is not affected at all, it has to be assumed that the starboard side absorbs all the weight which is another indication for the complete misalignment of the visor.
Arrow 4 - Indicates a further probably cracked area.
In summary of the above it has to be concluded that the visor was completely misaligned in longitudinal, transverse and vertical direction which led to considerable problems during opening and closing. These problems increased all the more as time went by as confirmed by the many observations, e.g. of previous passengers, in particular the truck drivers having spent hours on the car deck waiting for the crew to be able to open the visor (see Subchapter 12.4.3 above).
The difficulties with the visor are also confirmed by previous crew members and pilots having observed the frequent opening and violent closing until the locating horn on the forepeak deck (pyramid) finally fitted into the mating pocket in the visor bottom (see Subchapter 12.4.3 above).
Consequences of the misaligned visor
Due to the increasing misalignment it was at first difficult to close the locking devices and subsequently to open them.
Reportedly it had not been possible to open and close the Atlantic lock hydraulically for several months prior to the casualty and the crew members had to go inside the visor before arrival and after departure from Tallinn and hammer open respectively closed the bolt of the Atlantic lock which, among other things affected the load-carrying capacity of the lugs. Also the hydraulic side locks were difficult to close and subsequently to open. Reportedly on many occasions after arrival in Tallinn the lugs had to be cut off the visor by flame-cutting to be able to open the visor. Subsequently the lugs were rewelded to the visor bulkheads which continuously reduced the load-carrying capacity of these lugs (see Subchapter 12.4.3. and Chapter 30).
Obviously also the visor hinges had become affected because on 17 September 1994, i.e. 10 days before the last departure, a passenger filmed the visor and, in particular, the starboard visor hinges in Tallinn during the closing procedure.
The passenger was standing in front of deck 7 and filmed the closing of the visor, finally zooming in on the starboard hinges. The video-prints below are part of the film. Picture 2 is an enlargement of Picture 1.
On Picture 2 there are obviously several major deficiencies visible, viz.
(a) the gap between the starboard vessel hinge plate and the outer bushing is much too large (arrow 1).
(b) the outer bushing is sticking much too far out of the outer visor hinge plate (arrow 2) [which explains the large gap mentioned above].
(c) the lower part of outer bushing is missing.
(d) the securing plate is missing and the outer end of the bolt is not visible.
(e) the outer bushing appears to be twisted anti-clockwise.
(f) the inner bushing appears to be twisted clockwise.
See also Subchapters 12.4.3, 12.5 and Chapter 30.
All this indicates that the steel- and bronze-bushings of the starboard visor hinge arrangement had been replaced some time ago in a very unpro-fessional way with an extremely poor standard of workmanship.
Also the member of a working team, the nautical student Rain Oolmets, who stayed on board from 15 July to 11 September 1994, had discovered cracks in the welding seams up to 10 cm long between the steel bushings and the visor hinge plates of the port and starboard visor hinge arrangements which underlines the generally poor condition of these safety relevant components. See Subchapter 12.5.
It is obvious from the above description of the visible deficiencies that the condition of the visor hinges rendered the vessel already initially unseaworthy.
Consequences of missing rubber packings
The last Finnish crew had requested the renewal of 15 m of rubber packing in way of the visor for the then planned yard time in January 1993, because the rubber packing had last been renewed in January 1991. This meant a renewal of the entire packing on the forepeak deck plus about 2 m up each front bulkhead as demonstrated by the following drawing.
Since the new technical managers N&T were of the opinion that visors are full of water at sea anyway, the rubber packings were not renewed in January 1993 and also not later. This means that the visor became quickly filled up with water at sea up to the height of the outside water level. In calm sea it was the height of the bow wave which filled the visor up to the 3rd stringer, however, the worse the weather was and the deeper the bow was diving into the waves, the higher the water rose inside the visor.
The following picture shows the ESTONIA in Tallinn, with open visor but closed bow ramp so that the otherwise invisible underside of the bow ramp is shown with the lowest section being dark - see arrow. Doubtless this was caused by the water inside the visor with hydraulic oil from the leaking cylinders swimming on top of the water surface. When the vessel reduced speed and the bow wave became smaller and finally disappeared completely when the vessel was alongside, the water streamed out of the visor and the oil settled everywhere inside the visor, also on the bow ramp which forms the aft inside of the visor.
The water streaming out of the visor was observed by many passengers when the vessel was alongside in Stockholm with closed visor and, among others, also by the skipper of the Sandhamn pilot boat as well as by several pilots when the ferry was slowing down during the approach to the pilot boat.
The bow wave became smaller and the water streamed at once out of the visor. (See Subchapter 12.4.4.)
It is obvious that at sea inside the visor the conditions in a partly filled ballast water tank did exist with resultant sloshing effects causing additional loads on hinges and locking devices and vibrations to the whole visor, the consequences of which are described in Subchapter 34.3.
(2) The bow ramp, its locking devices, hinges and rubber packings:
The bow ramp was part of the upper extension of the collision bulkhead above bulkhead deck and as such had to be mandatorily watertight based on the stringent requirements of SOLAS 1974. This, however, was obviously not the case because back to December 1993 there is reliable evidence that frequently there had been a lot of water on the car deck - see Subchapter 12.4.4. There is also a lot of evidence that the bow ramp was severely misaligned and - in fact - was no more connected to the port outer and probably also inner hinges for some time - see Subchapter 12.4.3. It has thus to be assumed that the water penetrated onto the car deck from the visor, which was filled up to the outside water level, as soon as the vessel was on full speed and the bow wave had risen to ca. 2.5 m height. Since even in calm sea the visor was filled above the level of the 3rd stringer, which is above the car deck level, it is apparent that even in good weather conditions water was streaming onto the car deck. Most of it was most probably absorbed by the 12 scuppers at each side, some of which, however, were reportedly always blocked by stones and other debris.
Consequently it has to be assumed that there was already a deadly connection between the always water-filled and -refilled visor and the large, open car deck via the open port lower edge of the bow ramp since some time before the last departure.
The locking devices of the bow ramp were difficult to lock and, once locked, even more difficult to unlock. There is lots of evidence from truck drivers and other passengers having waited for hours on the car deck watching the fruitless attempts to open respectively close the securing bolts hydraulically and, finally, watching them being opened by flame cutting of the upper parts of the mating boxes, which should have been subsequently rewelded but frequently were not. When the crew did not succeed in closing the bolts hydraulically, they tried it by hammering, even by using sledge-hammers and very often gave up, i.e. the ramp remained - at least partly - unsecured! (See Subchapters 12.4.3 / 12.4.5.)
All this led, of course, in time to a considerable reduction of the load-carrying capacity similar as with the hydraulic side locks of the visor. Also the hinges of the bow ramp were severely affected, in particular the port outer hinge. This very heavy hinge arrangement was already damaged before the vessel was taken over by her new owners in January 1993, because the old Finnish crew had already requested renewal at the planned yard time in January 1993 - see Subchapter 6.5.1. Nothing was done by N&T and since the mates of the ESTONIA performed the loading and discharging of the cars and trucks reportedly rather carelessly, the vessel very often had a list during loading or unloading which finally broke the port hinge completely according to the report of a truck driver being an eye witness.
The crew had severe problems to lift up the ramp at all which was only possible after some temporary measurements had been carried out. Ever since, the misalignment of the ramp increased visibly and the problems were growing daily.
Also the rubber packings around the car deck opening, which should be sealed tight by a properly closed and locked bow ramp, apparently did no more seal watertight since some considerable time already, because the last Finnish crew had ordered 10 m of rubber packing to be renewed at the next yard time which was planned for January 1993 - see Subchapter 3.4. - as mentioned before. This was more than half of the total length of the rubber packings, but also these packings were never renewed.