How the classic hulls were built

fred henderson
15th September 2005, 22:31
In the thread “How the classic hulls were created” I explained how the Mould Loft translated the Drawing Office lines plans into full size lines, from which patterns were produced. The frame patterns produced in the Mould Loft were used in the Frame Bending Shop to shape the individual frames for a ship.
Each of the ship’s frames was made by bending a long flanged bar to the exact shape required to enable it run in single piece from the keel to the edge of the weather deck. The Frame Bending Shop contained one or more large, very thick slabs of steel, with a regular pattern of holes, rather like a giant peg board. Strong steel pegs were fitted in to the board to pick up the curve of the unique wooden pattern for each individual frame. In early days, the frame was heated in a furnace, one end was then clamped to the board and the other end hammered across the surface of the board until the required shape was obtained. Later the frames were cold bent using very powerful hydraulic rams. Either way it was slow, hard but very precise work.
After each matching pair of port and starboard frames was bent the pegs were adjusted to the shape of the next pair of frames. The frame spacing along the length of the hull would usually be between 18 inches to 2 feet, so there were a lot of frames to be bent. After bending, the frames would be carefully marked and rivet holes drilled. A similar procedure would be used to produce the beams. Meanwhile in the Plater’s Shop all of the individually numbered plates would be cut to size, rivet holes drilled and the plate rolled to achieve the required curvature.
Out on the building berth preparations would be made to begin construction.
The building berth was merely a graded river bank. In the old days it was simply compacted earth, and remained that way in some yards. Later a reinforced concrete strip was placed down the middle of the berth and in some yards the entire area was concreted.
On the centre line of the building area a row of square platforms were set up to support the ship, high enough for the builders to work underneath and spaced far enough apart for workmen to pass from one side of the ship to the other. The platforms were constructed from heavy wood blocks rather like railway sleepers. Packing pieces were placed on the top of the blocks so that there was a constant angle of declivity for all the towers, which were called “keel blocks”.
The first part of the ship, the keel plate was laid on these blocks. It was in fact a single line of plates, riveted together along the centre of the flat part of the bottom of the ship. In most traditional shipbuilding contracts, laying the keel triggered an instalment payment from the future owner to the builder.
Once the keel was in place, a web of makeshift wooden scaffolding was constructed close to the future outline of the ship. Individual frames were then lifted, held in position with wooden props and secured to the keel plate with service bolts. The double bottom tanks were then constructed, the first transverse beams attached to steady everything and the line of frames faired. As soon as the shape of the ship was confirmed, permanent riveted connections replaced the service bolts. I have posted in my gallery an 1895/6 Armstrong’s photo of a ship at this stage in production.
When all of the frames had been erected and secured the builder claimed another instalment in the price. Plating then began. The plates were attached to the frames in fore and aft strakes, which were vertically overlapped alternately over and under. So two complete under strakes would be fitted then joined by the over strake. These strakes were assembled, one plate at a time and attached with service bolts until riveted. Beams were attached and decks laid to allow the workers access to the inside of the structure as construction progressed. I have posted a Harland & Wolff photo of Aragon at this stage of construction in 1959. When all of the plates were riveted in place the builders claimed a further instalment and set about preparing the ship for launch.
The entire riveting process was a major problem for shipbuilders. All frames and plates needed to be precision drilled before the ship was assembled. All plating needed to be overlapped adding significantly to the structural weight. The frames, beams and plates were individually attached to the ships structure, often high in the air with minimal facilities. The riveting process required a lot of workers and was very dangerous. The first worker in the process was a Heater, who heated the rivets in a special coke brazier until they were almost white hot. (It follows that no riveting took place at the start of a shift.) Once work had started the brazier could not be moved. When a rivet was up to temperature, the Heater removed it from the brazier with special long handled tongs and threw it towards the Catcher with the rest of the gang. The Catcher caught the red hot rivet in special leather gloves and inserted it into the next hole. The Holder-On placed a special tool over the head of the rivet and used brute strength to keep it hard against the plate, until the Riveter hammered over the other end of the rivet. As the rivet cooled it contracted and sealed the connection. That seal was not always watertight however, and the Caulkers gang were employed to seal the joint.
The astonishing thing is that although shipbuilding moved from wood, to iron, to steel as its construction material, the basis of the building process did not change, even though the structural properties of the materials used were very different. The classic shipbuilding methods produced aesthetically pleasing hull lines in a hugely inefficient manner.


16th September 2005, 02:55
As an architect who for many years taught building construction, including steelwork, I have always been intrigued by the lapped connections between adjacent plates in a strake, and with the formation of rebates in the plates where the lap joint meets the strakes above and below. Also with the way in which the strakes become flush at the stem. Does that make sense? I would very much like to hear a detailed explanation of how all that was done, it would conclude years of speculation on my part!

fred henderson
16th September 2005, 22:59
Hi allanc, I will try to answer your questions. Firstly however, I am sure that you will realise that this post addresses the most common shipbuilding methods. There were of course a number of variations.
In most ships the plates were aligned vertically in the same way as they are in a modern, welded ship, so there was no need to rebate the joints. The edges of the ship’s plates where normally left square cut, as there was no machinery capable of changing the profile, until the introduction of automatic burning machines. The plate butts were normally lapped aft, except that in some, high speed or high value vessels they were flush with connecting strips on the inside.
The stem was made from forged steel. I have posted a photo of the launch of Monarch of Bermuda, which shows how the plates were attached. Above the waterline a soft stem was constructed from sharply bent vertical plates. You will notice the half-round bars that were added to protect the plate butts from the anchor.
I hope this helps.


17th September 2005, 07:02
Thanks Fred,
That helps, but being a graphic-minded character I probably need to see a picture! Over the years I've looked hard at the few surviving rivetted ships in our region, such as the restored Minesweeper ex HMAS Castlemaine at Williamstown, and have a fair idea of the principle if not the complete details. I encouraged a former student to photograph the Jeremiah O'Brien and it seems that the Liberties, although welded generally followed rivetted practice, some being substantially rivetted. I remember the Forts etc. were rivetted at vertical joints, welded horizontally and with plenty of variations. My trouble is in trying to recall things that I saw as a 15 year old!
Thanks, and best regards,

30th September 2005, 11:01
I'd like to hear some other comments from more people involved in ship constructon, as Fred points out, there were many variations from yard to yard

30th September 2005, 16:31
I also understand that shipyards used a porthole cutting tool, doing a batch at a time?
Some task on the earlier liners eg Lusitania & co.

30th September 2005, 16:42
I believe more trade unions were involved in cutting /fitting port lights than most other things on newbuilding, is this correct ?.

fred henderson
30th September 2005, 21:57
You have to remember Malcolm, that the shell plates were attached to the frames and adjacent plates one at a time, so the plater was only concerned with the portholes in the plate he was working on before it was issued to the building berth.


Derek Roger
16th October 2005, 23:50
Even after the introduction of welding the shear strake was still rivited due to the high stress concentration and the naval architetcs of the day did still not Have confident in welds .
I had an emergency docking of an old dredger into the Pictou shipyard in Nova Scotia ( she was all rivited ) . The Pictou yard incidentally during the war set the record for tons of ship produced per man per day . They built Victory vesssels ( all welded )
The Dredger had caught her bucket on one of the seams and ripped it apart . Divers plugged her with matresses etc while we cleared the slip to recieve her . The repair was simple enough except there were no riviters ( we would have had to go to the graveyard to dig them up ! ) Also there were no " Huck bolts readilly available ( they are like a rivit except they are fastened bye a hydraulic jack insted of using the contraction of the rivit as it cools to effect the joint . We eneded up in doing a full weld repair which was difficult when it came to mating the new plate with the rivited connection as they were not flush as you suggest but being lapped reqired some forging of the new plate to marry the shape .
The real skill however was not so much in the hull which only saw about 15 PSI ( for a 30 ft draft vessel ) of water pressure but in the skill of the boilermakers who ( in the case of Scotch boilers ) had the same lapped connections to deal with but for a woking pressure of around 250 PSI which meant a test pressure of 2 times that ie 500 PSI. I have some interesting recollections re boilers if anyone is interested .