A few observations about of Bismarck's failed stern section

[Dave Saxton]

Since I don't want to get involved in the exchanges taking place on that other thread, I thought I would post some mainly technical observations about the weld designs and weld execution to the detached stern section.

The implied conclusion of much of the recent posts, and many secondary, and tertianery writings is that the German welded joints on Bismarck were compartively poorly designed and excuted, compared to the practices of other (allied) warships from the period. This doesn't seem to be the case, at least not in comparision to then comtemporary practices. Of course, compared to today it was probably poorly designed and executed.

I recently found a USN ship fitters blue book from 1941, and here found many of the same welding design practices that we find so shocking on the Bismarck, being reccomended to USN ship builders. One of the most alarming US practices was the designation of many welds as "sealing welds". These sealing welds were not meant to perform a significant structural strength mission, but to simply seal a joint or gap from air or liquid. It actually reads that such joints don't need to be of high quality, as the main structrual joining is to be accomplished by some additional joining system. Usually a riveted joint or a hybrid riveted/welded joint design. Did German ship builders of the time hold similar views? I don't know, but it would explain much, if so.

The stern section of Bismarck was joined by a system of lap joints, with one side rivetted and the other side welded on by end fillets. This was the primary system in my opinion. Some of the outer seams of the joint, such as the upper deck and the shell plating had welded butt joints(that clearly failed). This was another system employed.

British weld engineers of the period understood the pitfalls of such hybrid joint design and worked to convince their own designers to not employ such joint designs in new designs:

"Overlap welded joints should be eliminated, except were required for assembly purposes. The experiance of builders familar with welded construction has shown that the provision of over lap joints as a safeguard against inaccurate plating and fit up is not generally necessary. Butt joints have greater fatigue strength than fillets.."(Shepard 1943)

The Germans understood this too (Kirschner 1938) and the materials being employed had such weldability, as to not require the older unwritten welding rules and conventions such as using lap joints with side fillets instead of butt joints, and never fillet welding cross ways on a strength member. These ingrained conventions and welding practices had been used to deal with the poor weldabilty of older materials. The joint design of the stern section on the Bismarck class however used rivet plates so this required fillet welds on the other side, if welding was to be employed as well.

The adoption of a hybrid joint design probably stemmed from several factors. One of course, was fact, that the joint location represented a structural discontinuity. Additionally, it also represented a point were dissimilar metals had to be joined. Moreover, it was near the stern so vibrations, and alternating loads represented a compounding set of factors to consider. End fillets were used on the welded side of these rivet plates. By employing end fillets, the fillet weld ran length ways on the framing, not cross ways, and it was mistakenly thought by many engineers during the 30's, that end fillets gave about the same strength under alternating loads as butt welds(Strength of Arc Welded Joints. 1931) Set rules (even unwritten) and conventions are not easily altered. Although it was well known that riveted joints had not near the tensile, or shear strength of a butt weld (The tensile strength of butt weld of St52 steel is more than twice that of a riveted joint) weld engineers had to admit that a riveted joint had overall more "give", than the heat effected zone in high strength steels. This may have been one rational in utilizing a hybrid approach at this location. We have only learned more recently that end or side fillet welds will usually fail when exposed to plastic deformation during high speed shock loading. It is no surprise that it is reported the riveted side of the joints remain intact and the welded side is gone, on the wreck today. In my opinion the Germans may have been too carefull to avoid adopting unproven welding practices too early, so they went only part way, still clinging to riveting, but this required the use of lap joints, and fillet welds.

There is a British memo from 1943 on how to effect quality welds in shipyards titled: " Welding Rules for the Executive". 9 points are layed out and all nine were followed by the constructers of the Bismarck as well. I havn't the time to go into this now, but some of these 9 points may have resulted in what observations we find today at the bottom of the Atlantic. In light of modern knowlege we can look back as engineers, or welders, or simply interested individuals, and cringe at the practices of those who were the poineers in generations past, but they didn't have the advantage of hindsight at the time.

[RF]

It shouldn't be forgotten that the stern failure came as a result of battle damage - the stern torpedo hit - not as a maritime accident or shoddy engineering.

Similary Hood was lost due to battle damage and not specifically to inferior design.

[Antonio Bonomi]

Ciao all,

when Bismarck sunk, she went down like an 'elevator' stern fist ,..bow the last,.. vertically.

Now the stern was still perfectly on her own position, all 3 propellers perfectly slowly running,... this according to a Bismarck survivors that told me this 2 years ago, .. and he was only 50 meters far away,.. just looking at her and at the damages if ever possible to be seen, ... and he did not notice any damage on that area,..propellers Ok,..rudders OK !

Now we only have 2 possibilities here, ... .if we trust the guy,.. and I had the impression he was still very sharp and his memories very good indeed on this particular case, .... either the water pressure going down, .. or an hit on the ground down below,.. on the hill as suggested by Ballard book sinking process drawing.

The huge damage on the area of the rudder smashed into the propellers seems to confirm a very strong hit on that area down there,... confirmed by the damages on the A-Anton turret barbette area were the ship hull collapsed completely,..as a result of the bow forced to touch the ground before the hull did,..so forcing the 'nose' bow of Bismarck to touch down before the hull and crashing the ship structure there,.. at A turret barbette level.

I had a very interesting discussion with many good friends at B&V in Hamburg, were Bismarck was built 2 years ago about this sinking process.

My personal opinion as far as today is that the hit on the ground is the most probable cause of that stern detachment and the other damages,...

.. so Bismarck sunk, .... levelled sinking,.. hit the stern on the ground first on the hill,... consequently unbalanced the hull level forcing the bow to hit soon after,.. than was completely on the ground and run down hill horizontally to reach the place she is now down there.

Just my opinion,... of course,...

Ciao Antonio

[Dave Saxton]

I guess it's been over two years ago that I wrote that intial post. A few key peices of information have come to light since then. One is a 1942 welding study of Stahl 52 welded joints, done in response to Prinz Eugen almost having it's stern blown off by a Mk38 torpedo. This study is a primary document. Another important document, is the welding protocols of Blohm and Voss put in effect in 1939, and still essentially the basis of their standards today. Another is the failure mode of the ober deck at the point of failure. Yet another is the rational behind the use of hybrid joint design for the internal connections.

The German 1942 study determined the stress's and levered loads required to break butt welds and fillet welds of St52 shipbuilding steel. The specs for ST52 are revealed to be:

UTS 53.3kg/mm2 (77,000psi)
YTS 37kg/mm2 (53ksi)
elongation 31%
reduction of area 61%
toughness 111J (~80lb-ft)

The welds were done with E52K low hydrogen shield arc electrodes (E7015-18). No welding faults of significance were in the Xrays of the welds. The weld zone UTS was an average of 53kg/mm2, but the YTS was 45kg/mm2, indicating the weld zones were harder and more brittle, which is always the case.

It required loads of 53kg/mm2 to break the butt welds to a leverage factor of three. When the leverage factor exceeded three, it required less loading. With a leverage factor of 10, the loads required to break the welds was only about 20kg/mm2. The butt welds always broke in the heat effected zones hard spots, adjacent to the centerline of the weld metal.

It required loads of 45kg/mm2 to break the fillet welds of T joints, to a leverage factor of two. Beyond a leverage factor of two, loads required dropped to about 35kg/mm2. The fillet welds always broke through the throats of the welds.

A key peice of information is where the upper deck is torn away on the Bismarck wreck. It's in the ST52 parent metal just aft of the butt weld to the Wh material forward. The remaining deck material aft is bent sharply downward at the weld seam. This particular weld didn't fail outright, but the forces exerted streched apart the St 52 plating for the most part, a short distance from the weld it's self.

This information allows us to quantify the loads that broke off the stern section, however it may have happened. It would be from <53kg/mm2, depending on leverage distance, to <65kg/mm22, the UTS of the Fox A7 weld metal, used to join to the 50mm Wh material forward.

[Dave Saxton]

I thought I had better post the above before I became de-logged.

Prior to 1939 the German Navy prohibited welding directly to armoured bulkheads. The rational was, that the heat would cause hard and brittle spots in the armour. By 1939, ways were found, through careful stress (pre and post heat) relieving, that armoured bulkheads could be welded. However, most of the German ships had their hulls completed by that time.

On Bismarck and the other German warships of the period, there's a transverse armoured bulkhead protecting the steering gear at the point were the distal stern was joined to the main hull. The welding ban created a problem of joint design, compounded by the fact that the joint would come at a rather abrupt discontiniuty in the overall structure. They had to rivet to the armoured bulkhead. They probably decided to use fillet welds on the other side to improve strength vs alternating loads. The shear strength of rivets would be significantly less, than a series of 400mm long fillet welds.

Normally such fillet welds would be enormously strong vs alternating loads, but it wasn't really know to engineers untill the 1992 Kobe earth quake, that rapid shock loading would tear this type of fillet weld apart relatively easily. The problem is, that a shock wave would slightly deform the parent metal surface under the fillet weld fusion areas, causing the weld to break through laminar tearing.

With the breaking of the internal connections by the shock wave of the torpedo blast under the stern, and additional battle damage, the remaining few butt welds of the outer shell, as well as the upper deck plating, could have been overwhelmed by the loads exterted during the sinking, however they may have come.

A major factor may have been rapidly osculating alternating stresses, caused by hydrodynamic forces during the sinking. B&V has determined the allowable stresses of such osculating loads, to St52 welded joints, in post war studies. They have determined several "klassifications" of allowable stresses of alternating loads, and the frequancy of the vibrations, for welded ST52 constructions. The allowable loads, according to B&V, under such conditions, are usually somewhat less than average tensile strength of St52 welded joints, but it depends on the joint design, any welding faults (particularly mis-alignments), and any possible damage to the weld .

[RF]

when Bismarck sunk, she went down like an 'elevator' stern fist ,..bow the last,.. vertically.

I was under the impression from Ballard that Bismarck was in more or less horizontal position as it ship progressed from the surface to its first contact with the underwater hill, rolling back to its upright position...

If it had hit the hill in a vertical position stern first, the stern presumably would have been smashed deep into the mud and obliterated completely before Bismarck slid downhill onto its final resting position, would it not?

[paul mercer]

According to Ballard, the stern broke off at the surface, but in his drawings of how he thinks she sank, BS left the surface stern first and upside down, then rolled back upright after the turrets had fallen out and went down with the bow slightly higher than the stern. He then shows the ship hitting the mud STERN first, with the bow hitting shortly after, but he gives no reason why it should not have been the initial impact that broke off the already weakened stern, or even if it broke off during the slide. Are there any reasons why it could not have happened that way?

[RF]

If it came off on initial impact with the hill, even with Bismarck more or less upright, I think the stern would have been smashed to pieces/driven into the thick mud; I believe the stern end section was found some distance from the main hull, more or less intact?