The Joint Accident Investigation Commission
/o Statens Havarie Kommission
Attn. Börje Stenström
Box 12538 S-102 29 Stockholm
Ihr Zeichen/Your Ref. Unser Zeichen/Our Ref. Hamburg/Date WH/PH/hz July 22nd, 1996 "ESTONIA"
Dear Börje,
Several months have passed since our last contact in writing and although we have recently spoken a few times on the phone, we would like to put down on paper a number of items which we consider to be of relevance before your respective investigations are finalised and the Final Report is completed.
I. Our letter dated 22.01.96 in reply to your letter of 21.12.95 has not been commented on by you at all. Consequently, we do assume that all the items raised in your letter have been answered respectively commented by us to your satisfaction, as has been confirmed by you recently during one of our telephone conversations. Nevertheless, we shall revert to some of the items further down as new information/evidence has been obtained in the meantime.
II. We have read the interview which Olof Forsberg gave the magazine "Stern" some weeks ago with interest, in particular his remarks concerning the quality of our work found our attention. In order to clarify our position with regard to the so-called FELIX Report please accept the following explanations:
1. We had received a translation of 5 pages (out of 85) of the FELIX Report from Spiegel TV with the request to find out whether the results of our investigation would confirm or disprove the allegation of the FELIX Report "that crew members had attempted to open the visor at sea". Before doing so we have cross-checked certain other allegations made in this Report in Tallinn and found them to be basically confirmed.
2. When reviewing our information/documentation it became obvious to us that certain observations of survivors - all of which were told to the criminal police during the interrogations on the first days after the casualty - but also particular circumstances which, so far, remained unexplained, could make sense if indeed crew members had tried to open the visor some time before the casualty sequence.
3. We conveyed our findings to Spiegel TV and "Der Spiegel" where they were published in distorted form.
4. In anticipation of the reaction of the media following the publication of Spiegel TV/Spiegel Magazine on 14th/15th April, we had prepared a press release (enclosure no. 1) which, we believe, makes our position quite clear, i.e. we have never said or written that we believe the allegations of the FELIX Report in connection with "Estonia" to be true, but that assuming that they should be true the stated observations of survivors and particular circumstances - as lined out in our press release - could make sense. So much for the FELIX Report.
III. Evidence from the Wreck
We were most surprised to learn from the Swedish
media that your Commission has stated it needed no more evidence from the
wreck. During our last meeting in Stockholm you had told us that you would
make sure that the interesting items would be brought up respectively taken
on video film before the covering commences. How is this discrepancy to be
understood? Why are you no longer interested in bringing light into certain
matters which do need clarification, if for no other reason than to avoid
the creation of legends? We have been requested from various sides during
the last weeks to state what we consider necessary to be brought up respectively
to be examined/filmed on site and have summarised this in a letter, which
is attached as enclosure no. 2. The letter was initially written upon request
of Henning Witte, the lawyer representing the DIS-Group of Relatives (Bendreus)
in their efforts to persuade the Swedish Government to postpone the covering
until at least your final report has been published. The letter was subsequently
also sent to Swedish TV stations, Svenska Dagbladet and Dagens Nyheter (Anders
Hellberg). It has recently been amended by the necessity to have the condition
of the watertight doors examined, which certainly had influenced the sinking
speed! We have read your interview in Dagens Nyheter in this respect and might
remark that the control light arrangement for the WT doors on the bridge must
have been changed after the vessel was renamed "Estonia", because as long
as she was sailing under Finnish flag the lights were "green" when the doors
were "open" (which was the normal situation) and "red" when the doors were
"closed". To sum it up, in our opinion, it is mandatory to have the accessible
objects brought up and to examine the other items on site respectively tape
them on video.
IV. The book "Katastrofen Kurs" by Anders Jörle and Anders
Hellberg
We have read this book with great interest. We gather therefrom that it is
now public and undisputed knowledge in Sweden that the ramp was untight and
that the technical managers and part owners of the vessel knew this positively.
After they had allowed the visor to become untight they also allowed the vessel
to sail with the only forward closure of the cardeck being untight as well.
As will be shown later this is the root of the catastrophe.
V. Finally, we would like to bring to your attention further investigation results in addition to those reported to you already in our previous letters. These are:
A. Visor 1.
Deformation / Misalignment / Non-Watertightness /
No Firm Connection Vessel-Visor1.1. Hinge plates
1.2. Stempost
1.3. Missing rubber packings
1.4. Old damages2. Locking devices
2.1. Atlantic lock
2.2. Hydraulic sidelocks
2.3. Manual sidelocks3. Hinges
4. Visor lugs of Hydraulic Lifting Cylinders (Actuators)
5. Inner Bulkheads
6. Conclusion 1-5B. Bow Ramp
7. Deformation/Misalignment/Non-Watertightness
7.1. Port outer hinge / bearing
7.2. Locking bolts
7.1. Leakages/Gaps
7.4. ConclusionsC. Cardeck / 1st Deck
8. Penetration of Water
8.1. Scuppers
8.2. Lifts / Doors / SillsD. Summary of Observations made by Passengers and Crew on different Decks
E. Sequence of Events
In detail:
A. Visor
1. Deformation / Misalignment / Non-Watertightness / No Firm Connection Visor/Vessel
1.1. Hinge plates Examination of the remains of the visor hinge plates at KTH, Stockholm revealed a/o that - the starboard hinge plates have deep burning marks from which numerous cracks (fatigue) had started and penetrated the material to varying extent. - the port hinge plates have a smooth surface with slight indications of burning marks in some restricted areas only. - measurements of the closest distances between the vertical plates to which the hinge plates are welded and the hinge plates revealed that this distance is 10-12 mm smaller at Starboard side compared to port side. This leads to the conclusion that the burning action has destroyed 10-12 mm of the hinge plate material of the Starboard hinges which, however, remained at the port side hinge plates. Consequently the visor was misaligned by the distance of 10-12 mm. Evidence: Report of Prof. Hoffmeister, Photos.1.2. Stempost The stempost was originally the strongest part of the visor - 200 mm wide and 60 mm thick - and formed the forward lower support, taking the major part of the design load of 55 ts, thereby resting on the upper extension of the ice-stem being the forward part of A-Deck, thus there was steel to steel contact. According to the initial design the stempost was supposed to carry at least 80% of the weight, whilst the 2 compression bars fitted on A-Deck and the (intact) rubber packings took the remaining weight. Due to this distribution the bottom of the visor was pressed into the rubber packings some 8-10 mm - as long as they were in place - thereby assuring watertightness. Between the top of pyramid on A-Deck and the pocket into which it fitted there was a design play of 50 mm, to the side just of 3 mm. As long as the visor had its original shape and geometry there was not steel to steel contact between the inside of pocket/pyramid nor between bottom of visor/A-Deck. Upon investigation in Hangö it was noted that the stempost was broken practically in 4 parts the lowest of which had disappeared well before the casualty, leaving a gap in the bottom of the visor. Due to the missing part of the stempost as well as the cracks sustained therein the stempost could no longer carry the main weight of the visor which was now taken up by the pyramid/pocket and the forepart of the visor bottom, which in turn led to cracks in the welding seams already some time before the casualty. Furthermore, the changed weight distribution caused additional stress on hinges and locking devices and added to the misalignment of the visor as a whole.
Evidence: Report Prof. Hoffmeister, Photos.1.3. Missing rubber packings It is evident from the underwater videos and demonstrated by several photos that the rubber packings on A-Deck were either missing or damaged with very few exceptions, whilst the flat bar and holding bolts, between which the rubber packings were fitted, are obviously undamaged and in place. The same refers to the holding bolts fitted by the crew when the vessel was still "Viking Sally". Consequently, there was nothing to prevent seawater from entering the visor inside once the outboard level was rising above A-Deck. The rubber packings were not just gone on the A-Deck but also in the most crucial areas in front of the 2 bulkheads at both sides of the ramp opening, i.e. of the vessel, where they are missing about 2 m upwards of the bulkheads. This is exactly the area where they are most required, i.e. on A-Deck in front and about 1 m up the bulkheads, because intact rubber packings, in particular in the above-mentioned locations, not only seal the visor inside against water ingress but are also a relevant part of the visor's closing system, because only the counter pressure of intact rubber packings against the closed locking devices assures a firm connection between visor/vessel and prevents vibration of the visor at sea. The lack of rubber packings in most relevant areas did not permit this firm connection between visor and vessel and the visor therefore vibrated within the existing play, which created the 'banging noises' heard by many passengers not only on the casualty voyage but also before. Needless to say that these movements of the visor created undue further stress on the locking and holding devices.
1.4. Old Damages There are 2 severe damages having existed before the casualty, viz. - in front of the visor between bottom and up the 2nd stringer level, and - in way of the 3rd stringer port side, which is bent upwards with bent and buckled frames/brackets. These damages have doubtlessly affected the geometry of the visor and led - in addition to all the other explained circumstances - to additional misalignment. Details have been explained in previous letters.
Evidence: Report Prof. Hoffmeister, Photos.2. Locking Devices
2.1. Atlantic Lock Examination of the remains of the 3 lugs fixed to A-Deck, now at KTH, Stockholm, revealed that u to 90% of the welds between lugs/bushings were fatigue cracked and the load carrying capacity reduced respectively. The starboard lug with the support bracket is not original, the welds between lug/bushing, bracket/bushing are repair welds made when the starboard lug/bracket was fitted to A-Deck.
Evidence: Report Prof. Hoffmeister, Statements of Yard welders, Photos.To test the load carrying capacity of the Atlantic Lock a mok-up was built by Meyer Werft with original dimensions, except for the welding seams between lugs/bushings which were dimensioned just 3 mm - as found at the lugs recovered from the vessel (original 7-8 mm) - upon request of Klaus Rahka/VTT. The breaking test performed by the Institut für Schiffbau of the University, Hamburg - Professor Petershagen - in the presence of Tuomo Karppinen and Klaus Rahka/VTT-Helsinki revealed that the visor lug broke at a load of 210 ts. Subsequent examination by Prof. Hoffmeister revealed that the bolt (ST 52 steel) was slightly bent, that Starboard and centre lugs were slightly bent apart but the welding seams - although just 3 mm - showed no cracks at all.
Evidence: Report, Video film, Photos from Institut für Schiffbau, Hamburg, Report Prof. Hoffmeister2.2. Hydraulic Sidelocks
We have nothing new to report, but would like to have the parts available at KTH to be examined by Prof. Hoffmeister for fatique cracks.2.3. Manual Sidelocks
These additional securings - called "heavy weather securings" by Bureau Veritas - were not - as we all know - engaged. We are aware of the statement of Ulf Hobro at our first joint meeting on 16.2.95, namely that these locks were only engaged when other parts, e.g. the hydraulic cylinders, would be repaired in port. Therefore it might be helpful to draw your attention to the statement of the 2nd engineer Peeter Tüür taken on 3.10.94 by the Tallinn police, in which he testifies that he was the responsible engineer for watertight doors and the hydraulic arrangements of the visor. He further stated that the visor has 5 fixing points and that there were 2 manual bolts which the boatswain has to engage. Evidently Peeter Tüür knew that these bolts had to be engaged, thus it follows logic to assume that at least the nautical offiers and the boatswain - if not the technical inspector - knew it.
Evidence: Statement of Peeter Tüür taken on 3.10.94 by Tallinn Police, Report about Joint Meeting International Commission/ German Group of Experts - 15/16.02.953. Hinges at Stockholm
(a) Examination of the recovered steel bushing with part of the visor hinge plate still attached to it revealed a/o that - the gap between bushing and hinge plate is some cm wide; - the material accumulated at one side and extending into the gap is apparently welding material; - in the remains of the welding seams between bushing/visor hinge plate small burning marks can be seen. This leads to the conclusion that the gap must have existed prior to the casualty as otherwise the material could not have gotten into it.
(b) A video film made a couple of days before the casualty by a passenger shows the Starboard hinge in detail. It is obvious that - there is a wide gap between the outer hinge plates of vessel/visor in a location where normally the inner part of the steel bushing and the distance ring should be; - the outer securing plate is missing, although 3 bolts can be seen sticking out of the bushing; - the outer part of the bushing appears to be disproportionately large; - the lower part of the bushing at the outer side appears to be missing completely. Note: The relevant part of the video film will be analysed by a professional analyst and the result will be submitted to you in due course.
(d) Examination of the remains of the 4 visor hinge plates at KTH, Stockholm revealed a/o that - the 2 Starboard and the inner port hinge plates are straight; - the lower part of the port outer hinge plate is bent to port . According to Prof. Hoffmeister this leads to the conclusion that the port outer hinge broke last when the visor was in a position to the port side of the centre line of the vessel.
Evidence: Report of Prof. Hoffmeister Report of Brian Roberts Photos / Computer drawing4. Visor lugs of Hydraulic Lifting Cylinders (Actuators)
After the failure of the hinges the visor moved forward and the lugs underneath the visor arms were cutting through the 8 mm deck plating, but only for 120 mm, then they hit the rather strong transverse deck-beam. According to the estimate of metallurgists it took any time up to 10 minutes for the lugs to cut through this strong beam. The condition can best be demonstrated by the photo below showing the void space on B-Deck Starboard side with closed visor.
B-Deck starboard side pre-casualty condition When the lugs were working their way through deck and deck beam whilst the visor was moving forward/aft naturally deep scorings were sustained by the lugs. Assuming that the vessel was still in an upright condition - which is part of your scenario - these scoring marks would be more or less of similar intensity at the outer as well as at the inner lug plates. This, however, is not the case as can be clearly seen by the photos below.
The above photo shows the outside of the port lug with evidently no or only minor scratches.
The port inner side, however, which is shown on the above picture, evidently sustained very deep and sharp scorings indicating severe pressure from port to starboard and straight as drawn by a ruler. The following picture shows the inner side of the starboard lug - practically the counter-part to the port outer side - with some deep scorings at the upper part indicating more of a circular movement.
The picture below finally shows the outer side of the starboard lug - practically the counterpart of the severely scored inner side of the port lug - indicating also severe pressure from port to starboard side, though apparently not as heavy as at the port lug inner side. The cut area through the green painted forecastle deck is clearly visible and straight as if drawn with a ruler.
The above pictures lead to the following conclusions:
- The scorings at the lugs can only have been caused when the lugs were cutting through deck/deck beams.
- Contact between lugs and deck/deck beams is only possible after the hinges have failed.
- The much deeper scorings at the port inner side respectively starboard outer side compared to the port outer side and the starboard inner side indicate severe pressure from port to starboard.
- This leads to the conclusion that the vessel listed already to starboard when the hinges broke and the lugs commenced their cutting work.
- Since it took some 10 mintues for the lugs to cut through the deck beam and thereafter another 360 mm of deck plating had to be cut before the aft side of the ramp-house could have come into contact with the ramp, however, there are only very minor scratches at the outer side of the port lug plate, it has to be concluded that the vessel must have already had a considerable starboard list while this cutting work was in progress.
- Furthermore, bearing in mind that there had obviously only been sufficient contact between these frames (at port side of the inner ramp house in way of the first 4 frames) and the port side of the ramp to overcome the ca. 40 ts holding power of the port ramp hook, the question arises what force opened the starboard ramp hook (if it had ever been closed), because the minor indentation and mainly only light scorings decreasing from port to starboard at the inside of the ramp house do demonstratea.) that there was only very minor contact between the larger part of the upper ramp and the ramp-house between the 5th frame counting from port side and then decreasing to the starboard side, and
b.) that the visor at the time of contact, i.e. when the port frames were bent and the starboard ones only scored, must have been in a condition shifted to port as it had already been when the last hinge plate, i.e. the port outer hinge plate, broke.- In other words, it can be concluded that when the ramp-house finally came into contact with the upper part of the ramp the vessel had had a very substantial starboard list for some time already and the visor had been in a condition shifted to port of vessel's centre line.
5. Inner Bulkheads
The same as can be observed from the lugs underneath the visor arms is confirmed by the damage/scorings to the port inner bulkhead, whilst the opposite starboard side is completely untouched with absolutely no contact marks. The 2 pictures below show the inner side of the visor bulkhead with deep scorings/indentation and the holes partly torn.
The ramp is at that stage already partly pushed open whilst the visor is resting in way of the obvious contact area on the ramp. The partly open ramp is thus preventing the visor from falling off to starboard. The sketch below shows the approximate situation at this particular moment. The foundation of the port actuator is still inside of the vessel's front bulkhead, although already broken loose from B-Deck.
Of course, the side view cannot demonstrate the severe list which the vessel must have already had at that time.
The following 2 pictures show the starboard inner bulkhead totally without damage or even scorings/scratches.
6. Conclusion
1-5 In summary of item A.1-5 the following further conclusions can be drawn:
- The visor was also misaligned due to improper burning at the starboard hinge plates.
- The stempost was severely damaged and had lost its main weight carrying function nearly totally. This led to additional misalignment of the visor in combination with additional stress on hinges/locking devices. The gap in the stempost area in connection with cracked welding seams has led to flooding of the visor inside up to outboard level.
- The missing rubber packings had, in addition to the a/m gap in the stempost and cracked open welding seams, also resulted in a rapid flooding of the visor inside due to the so-called pumping effect.
- The firm connection between visor/hull of vessel normally established by the combination rubber packings/visor weight/tension of locking devices was lost due to the missing rubber packings in relevant locations which led to movements of the visor at sea within the existing play and additional stress on hinges/locking devices. The movements of the visor created the low banging noises heard by passengers on previous voyages as well as on the casualty voyage.
- At least the 2 most severe old damages in the visor had led to additional deformation and misalignment.
- The Atlantic Lock arrangement at vessel's side was weakened due to improper repair and fatique cracks in the welds between lugs/ bushings of up to 90% of its initial load carrying capacity.
- The 1st breaking test performed by the "Institut für Schiffbau", Hamburg revealed that the visor lug broke at 210 ts load whilst deformations of the bolt and the starboard and centre lugs were minor. The welds, however, showed no cracks at all, although just 3 mm strong. This leads to the following conclusions:(a) The arrangement failed at 210 ts which is more than double the design load of 100 ts.
(b) The arrangement at vessel's side with even weaker welds than having existed at the initially installed Atlantic Lock - withstood 210 ts, i.e. is able to carry more than 210 ts.
(c) According to the damage picture found on "Estonia" the lugs at vessel's side broke, whilst the visor lug withstood, although stretched and bent. This leads to the conclusion that the Atlantic Lock arrangement at vessel's side must have been substantially weakened - as found by Prof. Hoffmeister - and thus failed at some load below 210 ts.
(d) The Atlantic Lock arrangement which was installed on "Estonia" at the time of its failure can in no way be compared with what had been installed by Meyer Werft on the newbuilding "Viking Sally". Its load carrying capacity was substantially reduced.Note: Three further breaking tests were performed by the "Institut für Schiffbau", viz.
(a) bushings/lugs at vessel's side without any welding connection at all = result : 100 ts;
(b) bushings/lugs at vessel's side connected with spot welds only (in order to simulate as closely as possible the condition existing at time of casualty) = result : 142 ts;
(c) bushing/lugs at vessel's side with 3 mm welds connected as found at the lugs having been installed at the time of casualty/visor lug over-dimensioned to make sure that the lugs at vessel's side break, bolt ST90 (as has been at the previous tests already = result : 220 ts.As the breaking occurred with no welds at all at 100 ts and with 3 mm welds at 220 ts it can be concluded that 1 mm welding seams corresponds to a load carrying capacity of 40 ts, i.e. the initially installed lugs/bushing with 7-8 mm welds had an LCC of 7 x 40 ts = 280 ts + 100 ts = 380 ts. Based on the test results of VTT, Helsinki the break-load of the starboard side lock was found to be 178 ts, whilst the port one was found to be only 131 ts. In addition, the 2 manual sidelocks would have produced about 50 ts each, i.e. 100 ts. In summary it has to be concluded that at the time the vessel was delivered to her owners there was an LCC of in total 789 ts~800 ts available which fulfilled all requirements having then been in force.
- The manual side locks were not engaged, although they should have been according to 2nd Engineer Peeter Tüür and Bureau Veritas.
- The starboard visor hinges were substantially predamaged. This is not only demonstrated by the remaining parts available at KTH, Stockholm, but also by the video film made by a Swedish passenger a couple of days before the catastrophe.
- The different damage pictures of the lug plates underneath the visor arms, i.e. the scratches/scorings at the inner and outer sides can only have been developed when those lug plates were cutting through deck-plating/deck-beams. Such cutting was only possible when the hinges were broken and the visor moved forward. The port outerside plate just shows some minor scratches whilst the port innerside plate is very deeply scored, indicating load from port to starboard. A similar picture is shown on the starboard lug plates where the outerside is much more deeply scored than the inner side. This is clear evidence that the visor's weight was lasting from port to starboard and this cannot be explained with the waves coming from port, because when the hinges broke wind/sea were already coming from straight ahead (statement V.Kikuts), whilst the vessel was turning further to port (statements Hillerström and Jonsson). The deep indentations/scorings at the port inner longitudinal bulkhead of the visor which were created when the visor was resting on the ramp, support this assumption as at the opposite side - the starboard inner longitudinal bulkhead - there is not even a scratch. There are more marks/indentations/ scorings and scratches pointing in the same direction, but the a/m should be enough for now. From the above it has to be concluded that the hinges broke when the vessel already had a considerable starboard list which triggered off the catastrophe sequence-of-events and must have been caused by something other than the bow ramp being pulled open by the visor. As will be explained in the following water entered the cardeck not only from the full visor through the damaged bow ramp, but also through the ventilation openings and subsequently through the scuppers with non-functioning non-return valves.
- At the time of the hinges breaking and contact between ramp/ramp-house the visor was in a condition shifted to port of vessel's centre line.B. Bow Ramp
We would like to to begin this section with some statements made by crew members to journalists on the day of their survival, the 28th September, 1994, after they had just been brought to safety:
(a) On 28.9.94 at 12h33 the following was published: "The 2nd engineer of the ship has been rescued and in an interview he said that the front "car gate" was defective and had let in water."
(b) On 28.9.94 at 15h57 the following was published: "According to an Estonian machine operator, Henrik Sillaste, who managed to escape from the ship through a funnel, the gate of the cardeck was for some reason slightly open and water came to the deck through it. The body of water, weighing many tons, that got inside the ferry made it very unstable, and the wind and the size of the waves (according to the reports as high as 10 metres) were enough to push the ship to its side."
(c) On 29.9.94 Reuter published the following: "The cause of the accident has not been determined. Henrik Sillaste, a surviving member of the crew, said he saw water pouring in through the bow gate shortly before the vessel capsized off the Finnish coast Wednesday morning. When the water was knee-high on the cardeck the engines stopped. The boat listed to its side and sank. Sillaste said he thought it was rain and wave water entering the ship because one of the two bow gates was not properly locked."As the above statements were made only a few hours after the crew members had survived the catastrophe and were still under the impression of what they had gone through during that night, and as they were made totally without the later influence from shore people, these statements have to be considered true, no matter what the respective person might have stated subsequently. As a matter of fact, there is more evidence in respect of "the front gate was defective and had let in water" and "the gate of the cardeck was for some reason left slightly open and water came to the deck through it", because 2nd engineer Peeter Tüür stated on 3rd October 1994 to the Tallinn Police among other things, "... behind the visor there is the ramp. It shall also be hermetically tight! You find just a few spots where the ramp is not hermetically tight." Meant is certainly the gap or gaps created by the twisted port outer hinge with broken bearing which had been put by the crew repeatedly (your words) on repair lists, however, repairs of which were obviously deferred by the management possibly until the next scheduled time in shipyard. In the book "Katastrofen Kurs" Anders Hellberg and Anders Jörle mention in this respect on page 76, top a/o: ".. But let us first turn to the damage to the ramp which according to N&T was taken up in the list of repairs to be carried out. Already here a fact is found which no normal classification society should accept: There was practically a damage to which in practice was the collision bulkhead of the vessel, but which was just put on the list for repairs to be carried out." As stated above this is obviously undisputed public knowledge in Sweden!
7. Deformation / Misalignment / Non-Watertightness
7.1. Port outer hinge/bearing It has to be assumed that this hinge became twisted when the vessel had a list to starboard in excess of 2°, which was not adjusted before the loading or discharging of trucks commenced and the damage occurred when one or more heavy trucks drove over the ramp. This seems to be a risk well known to ferry operators. The hinge became twisted and the bearing broke, which was most probably only noted by the crew when the ramp could not be closed before departure because the port side had sunk down somewhat. To rectify this some piece of steel was hammered underneath the bearing to keep the ramp closest possible to its upper normal position. Obviously it was thereafter possible to open/close the ramp, however, some gaps remained, i.e. the bow ramp = upper extension of collision bulkhead above freeboard deck = was not watertight anymore, or to use the words of 2nd Engineer Peeter Tüür: "no more hermetically tight", because it was deformed and misaligned.
The temporary repairs can best be seen from the photo below made from one of the underwater video films. The steel piece underneath the bearing can be clearly seen. The ramp was kept in approximately normal position as far as up/down is concerned, however the forward deformation could not be corrected by that.
7.2. Locking bolt
In addition to the 2 hooks at the upper part of the ramp, there are at each side 2 hydraulically operated and conically shaped locking bolts which moved from vessel's side into pockets at the ramp side, thereby increasing the pressure of the ramp on the rubber packings. The 2 starboard bolts and the upper port bolts apparently still had been able to move into their mating pockets, because these pockets were found to be torn open after the casualty. The port lower bolt, however, could not move into its pocket due to the deformation of the ramp in forward direction.7.3. Leakages / Gaps
As a result of the above there were substantial leakages around the ramp to the large cardeck.
- at the port lower corner about 1,5 m upwards and about 1,5 m athwartships;
- at some locations at the opposite starboard side of the ramp. As at sea there was always water inside the visor up to the outboard level - the picture below shows the normal height - something had to be done to avoid major water ingress on the cardeck.As the owners/technical managers were obviously not prepared to take the vessel out of service for the required time to permanently repair this damage and on the other hand major water ingress to the cardeck had to be avoided by all means, the crew sealed the existing leakages both from visor side - see picture below - but also from cardeck side. This was frequently observed by truck drivers and other passengers.
Note: According to a previous crew member it was known to those responsible since March 1994 that there were substantial leakages in the bow ramp and that there was always some water on the cardeck.
Needless to say that such method to secure watertightness of the upper extension of the collision bulkhead of a passenger vessel is unacceptable and totally useless against a water column of about 6 m in the visor
7.4. Conclusion
In summary of this Section B it has to be concluded that
- the bow ramp - being part of the upper extension of the collision bulkhead - was damaged to such an extent that there were substantial gaps/leakages between the inside of the visor and the cardeck;
- the port lower securing bolt could not engange its mating pocket at the rampside incl. the sensor for the indicator lights (in consequence neither on the bridge nor on the cardeck panel could be "green" indicating ramp closed and secured).
- the cardeck was open to the water quantities inside the visor which would have been disastrous for the stability of the vessel already long before the catastrophe. To avoid this the crew plugged the leakages from visor as well as from cardeck side, which might have been partly successful during normal trading, but was ineffective against the pressure built up in the visor on the night of 27/28.9.94.
C. Cardeck / 1st Deck8. Penetration of Water
8.1. Scuppers
We assume that there had always been some water on the cardeck since March 1994, the quantities of which certainly depended on the height of the water level inside the visor, which, in turn, depended on the sea condition. In other words, the worse the weather and seastate were, the higher the water level was inside the visor (and the more violent was the water movement inside the visor) and the heavier was the water pressure on the "seals" in the leakages, the more water penetrated these "seals" onto the cardeck. All this was well known to a number of crew members and the technical managers. In order to cope with the water on cardeck which, due to the more or less permanent stern trim, quickly ran to the aft part, some or all scuppers were constantly open. We assume that the remark made by Äke Sjöblom to a journalist when he was interviewed on the 28th September, 1994 about his findings onboard the "Estonia" on the 27th September, 1994 has to be understood in this respect. He said: "We went down to the cardeck and experienced a lack of respect for loadline matters, and when we say this and we discussed it with the Chief mate, we said we don't understand how come that this hatch cover is open here? It has always to be closed and it has not been closed for the last 1 or 2 years, and so on. I think it was in 2 or 3 places." In the subsequently drawn up "Report of Inspection in Accordance with the Memorandum of Understandiing on Port State Control" signed by him in Tallinn on 27th September, 1994 Äke Sjöblom has a/o noted:
"1250 Covers on bulkhead deck to be closed."
Convention reference: LL66, C II-1
Action taken: 17 = to be rectified before departure"You know certainly much more from him as you have no doubt spoken to him in detail. In any event he certainly does not know in what condition the scuppers and other openings on bulkhead deck = cardeck were, when the vessel departed from Tallinn.
8.2. Lifts / Doors / Sills
There are 4 lifts inside the centre casing with openings to the cardeck
- in the forward part a person-lift which goes via 1st Deck down to 0-Deck (swimming pool/sauna);
- in the middle part a person-lift which ends on 1st Deck. Both lifts have only 100 mm sills on cardeck level (the cardeck doors have 230 mm sills). In addition there are 2 big store lifts further aft down to the provision/ store area on the 1st Deck which have no sills at all (suitable for forklift). It is thus possible for water to run from the cardeck through the lift casings into 0-Deck (most forward lift) and straight into the 1st Deck (deck below cardeck) into the accommodation area as well as in the provision area. Whilst we have no evidence that there was water in way of the provision area (although we assume that there was), there is a lot of evidence that there was water in the accommodation area of the deck below cardeck and on the cardeck at a time when the bow ramp was still only slightly open, i.e. before the major starboard list commenced. The quotations below are part of the statements of passengers having been in cabins below the cardeck. These statements were taken by the criminal police soon after the rescue.Antti Arak: cabin 1056 - 21 years "At first I heard screaming and then I saw water running in the alleyway."
Ain-Alar Juhanson: cabin 1056 - 18 years (at Turku 29.9.94). "Whilst I slept I heard a heavier bang which I had heard before already, but not so heavy. After that bang the vessel suddenly listed to starboard. All 4 of us woke up, grabbed our clothes and jumped into the alleyway, where we noted water."
Martin Nilsson - cabin 1027 - 22 years. - he left the cabin and jumped up one flight of stairs to the next deck (cardeck), there was some water on the bottom - some cms.
Bengt Nilsson - cabin 1026 - 35 years. - He opened his door and saw Martin Nilsson jumping past him.
- When he reached the deck above (cardeck) he saw water coming from wall and streaming over the bottom and walls. It was rising.
- He stated also that he saw something along the walls which looked like some powder.Carl Övberg - cabin 1049 - 42 years.
- he knew his way up and when he passed the cardeck, he saw water penetrating.
Note: This was at a time when the first - slight - starboard list had just occurred, i.e. before the sudden port list at 01h02.Carl-Erik Reintamm - cabin 1094. - ca. 23h00/23h30 in bed;
- suddenly he heard 2 heavy, strange scraping noises which apparently came from below;
- he felt also that the cabin was tilting, but just a little bit;
- he opened the door and saw a small stream of water behind one alleyway wall on the floor;
- vessel had only a small list at that time;
- he rushed up and reported to 2 crew members that he had seen water in the alleyway downstairs, whereafter both rushed down;
- the vessel listed more and more.Taavi Raba - cabin 1070 - 21 years.
- when he left his cabin (after the first list) he had to jump through a water curtain (wall) to reach the stairs;
- when he passed the cardeck doors he saw water penetrating.Jaan Stern - cabin 1122 - 53 years (Estonian)
- most forward cabin directly adjacent to the water ballast tanks;
- was in cabin which he had to leave, because water was already entering the cabin.Holger Wachtmeister - cabin 1047 - 41 years. -
he was awakened by a scraping noise and a hard bang - the scraping noise continued;
- he got up and dressed and left the cabin when the list started;
- he noted water when he passed cardeck possibly already below in front of his cabin.Jasmina Weidinger - cabin 1027 - 19 years
. - she was also hearing the noise of the stabilisers which held the vessel steady when she suddenly heard a murmuring noise. The noise, which she had not heard before, was low in its intensity. It appeared to her that the murmuring sound came from above and it appeared to her as if water was running very slow
- she looked at her watch; - this was at 24h00 (Swedish time);
- a little later she heard a heavy sound above the cabin ceiling;
- she got the impression that something large and heavy was moving from one point to the other;
- after these heavy scraping noises the vessel took the first list;
- Daniel jumped out of the cabin and shouted that there was water in the alleyway;
- she also saw that water trickled out on the floor before her cabin.All cabins are located on the 1st Deck at starboard side, i.e. the deck below cardeck, with the exception of cabin 1056 - which is located at port side but directly at the midship alleyway next to the 3rd stairway (from forward) leading down from the cardeck - all other cabins are located at the starboard side. In the same connection has to be seen the conversation between Security Officer Ervin Rhoden and an unknown man in Estonian language held in Södersjukhuset Hospital on 28.9.94, which was overheard by Paula Liikamaa (casino employee). Paula testified to the police: "The watchman whose 1st name is Ervin and surname is Rhoden said that he had switched on his walkie-talkie when he was in his cabin, when he heard another watchman - Silver Linde - say over his walkie-talkie that he was on the cardeck and that there was a lot of water on the cardeck and that they must leave the vessel." As Ervin Rhoden with absolute certainty did not listen to any walkie-talkie communication after 01h02 when the abrupt and severe port listing occurred it must have been before that time, i.e. this is another indication that Silver Linde was on cardeck before 01h00 when there was already a lot of water.
In summary of the above quoted extracts from statements taken by the criminal police of Finland, Sweden or Estonia shortly after the casualty it can be concluded with sufficient certainty that at the time when the vessel took the first major starboard list, there had already been so much water on the cardeck that
(a) it was able to flow over the 100 mm sills of the person-lift, run down the casing to the 1st Deck, fill the deepening below the lift and fill part of the alleyways between the cabins at starboard side, partly already penetrating into the cabins;
(b) it was able to flow over the 230 mm high sills of the cardeck doors and was rising inside the centre casing on cardeck level, simultaneously flowing down the stairs to the deck below (water curtain).D. Summary of Observations made by Passengers and Crew on different Decks
In order to get as close as possible to the most probably sequence of events we have carefully analysed the statements of passengers and crew members made to the police shortly after their rescue. The various statements of Silver Linde have not been taken into account.
Deck 8 : In the fore part were the accommodations of Master and Chief engineer as well as the owners cabin, which was occupied by one of the Stockholm pilots with his family. The Master and Chief engineer and the Stockholm pilot with his family are missing. In the aft part are some crew cabins and the day-rooms of officers and crew. After the first heavy list to starboard one crew member looked out of one of the windows at starboard side of the dayroom and saw "another ship in the sky ".
Note: This must have been "Silja Europa" then already abeam.Deck 7 : This is the main accommodation deck for the crew with 4 cabins in front of the superstructure which were used as follows: (from starboard to port)
1st cabin starboard side: Economy superintendent Jan Bergendahl (did not survive)
2nd cabin starboard side: Technical superintendent Tomas Rasmusson (not onboard)
3rd cabin starboard side: Chief mate Juhan Herma (did not survive)
4th cabin starboard side: 1st Engineer Arvo Tulvik (did not survive).The windows of these cabins were overlooking the foreship and it has to be assumed with certainty that the Chief mate and the 1st Engineer did look out of their windows probably already after the first severe bangs at 00h45/50 and at that time apparently saw nothing, therefore it has to be assumed that the visor hinges and locking devices were not yet broken at that time and that the visor was still in place, because otherwise they would have alarmed the bridge already at that time and would have commenced action much earlier. Consequently the bangs heard at 00h45/50 must be connected to something different breaking, e.g. the bow ramp. About 10-15 minutes later, i.e. shortly before 01h00 or at 01h00, also the visor hinges and side locks broke and the visor moved forward/aft and somewhat up and down within the 120 mm space between the fore part of the actuator-lugs underneath the visor arms and the strong athwartships deck beam at frame 159. Each time the visor had moved forward a gap of 100-120 mm opened athwartships and water from the inside of the more or less full visor splashed out in cascades and subsequently surged in depending on the movements of the vessel. This was observed by Valter Kikuts, the deck passenger sitting with his friend behind the windows at port side in the stairway on 7th Deck, i.e. directly next to the cabin of 1st Eng. Arvo Tulvik and before the cabin of 2nd Eng. Peeter Tüür. Kikuts has stated to have seen this gap opening/closing already at 00h36 which appears to be impossible, 00h56 is the more likely time. In any event it was obviously seen by 1st Eng. Arvo Tulvik already before or immediately after the 1st heavy listing to starboard, i.e. at about 01h02. He had apparently left his cabin and walked aft when 2nd Eng. Peeter Tüür opened his cabin door and Arvo Tulvik said to Peeter Tüür: "Apparently the visor broke up, would be good if the vessel would be beached." This was also heard by the Reefer engineer Andres Verro, who left his cabin at the same time. Based on the apparent fact (see statement Mikael Öun and many others) that the 1st severe starboard list occurred at 01h02, Arvo Tulvik must have made his observation before that time, i.e. probably at some minutes before or at 01h00, as it is most unlikely that the 1st Engineer would have stayed in his cabin much longer than a couple of minutes (to phone the bridge knowing that they can see only the tip of the visor and his boss, the Chief engineer one deck above him, who as well as the master can see from their cabin windows nothing at all of the visor, finally to get dressed and leave the cabin). That's when he met Peeter Tüür and Andres Verro.
Consequently it has to be considered an established fact that the visor hinges and side locks broke at 01h00 or some minutes before.
Next relevant observations made / noises heard by crew-members and passengers shall be summarised as stated below:
2nd Eng. Peeter Tüür:
- woke up by things moving around in his cabin
- had heard before metallic bangs apparently caused by objects hitting against each other
- thought the noises came from badly secured lifeboat or that the visor had gone loose;
- felt that the speed was reduced and that there was a starboard list of about 30°;
- opened the cabin door and met the 1st Engineer Arvo Tulvik who said: "Apparently the visor broke up, would be good if the vessel would be beached."
Note: The cabin of 1st Engineer Arvo Tulvik is on 7th Deck, port side, in front of the accommodation. From the window he can overlook the foreship with visor. The above remark from Arvo Tulvik was also heard by the Reefer Engineer Andres Verro (see his statement).
- Next he heard "Mr. Skylight 1 and 2", which means according to Peeter Tüür's understanding: "Fire Alarm", the 2 fire-fighting teams had to go to their stations
- he was No. 1 of team No. 1 and his station was on the 8th Deck - port side;
- he went back to his cabin, picked up walkie-talkie and flashlight, left via window (due to the list) and went to Deck 8;
- at that time he was confident that there was the risk that the vessel might sink;
- heard on his walkie-talkie that the Information Desk tried to reach the bridge, but got no answer;
- he tried it himself but also got no answer;
- light went out but came back (auxiliary shut off and start of emergency generator);
- after 4-5 minutes also this generator stopped - vessel on the side - and thereafter only emergency lights fed by batteries.
Note: This was 01h31.Andres Verro - Reefer Engineer - cabin 710 - 35 years.
- was in cabin;
- at ca. 01h00 was in bed reading a book when the vessel suddenly listed to starboard;
- heard by radio "Mr. Skylight 1 and 2" which means actually "Fire Alarm" spoken by a male voice;
- believes that he heard before the list some strange metallic sounds;
- when he left his cabin he met the 1st and 2nd Engineers, one of them said: "Visor went away" (or something like that);
- he got the impression that possibly the fastening clips of the bow ramp had failed and through them water came in.Aarne Koppel at Turku - 29.9.94 - cabin 7011:
- 5°-10° starboard list, felt it was an unnormal situation and woke up
- heard a low breaking sound;
- at 01h00 - when we got dressed - the vessel suddenly listed to ca. 30°;
- when I was still in the cabin putting on warm clothes the light went out/on several times, that was when the main engines stopped;
- it took ca. 5 minutes from the moment I woke up from the 5° list to my leaving the cabin through the window, which was at the high port side.Elmer Siegel - motorman - cabin 7007 - 43 years. Statement taken 29.9.94 at Mariehamn:
- was in cabin on 7th Deck aft;
- 3-5 minutes before list started heard noise like somebody beating against vessel's hull with a large hammer which came from cardeck direction.Andres Vihmar - cabin 713 - 7th Deck (cabin is located 20 m aft from front bulkhead)
- 5-7 minutes after 1st listing the list was already 45°;
- when he came to the boat area on 7th Deck port side Aulis Lee and Aarne Koppel were already there;
- people were streaming up to the 7th Deck through the open doors for another ca. 5 minutes, thereafter it was no longer possible due to the list and no one escaped anymore from the inside of the vessel.
Note: This means that already after ca. 01h15 it was no longer possible to get out of the inside of the accommodation.Summary of observations made / noises heard on Deck 7 and 8
a) Before the starboard list: 1st Engineer saw visor moving / loose / broken up and various crew members heard metallic noises and as if somebody was beating with a large hammer against vessel's hull which came from cardeck.
b) After the starboard list:
- there was a vessel about abeam at starboard side
- probably "Silja Europa";
- speed was reduced and subsequently the engines stopped after the list was 30° or more;
- the Information Desk as well as 2nd Eng. Tüür tried to reach the bridge by walkie-talkie - ca. 01h10/01h12 - both got no reply;
- already after ca. 01h15 it was no longer possible for anybody to get out of the vessel due to the excessive list.
