Shipshape 8

How to Build an Anglo-Saxon ship

Paul Constantine

One-fifth model turnover

Unruly Underlout, Broad Boards, Rivets and Roves

One-fifth model

Shipshape 7 explained the importance of the 1:5 model in trying to determine the plank runs for the ship. The model was constructed upside down and in order to investigate further it had to be released from its formers to access its interior, the correct way up. The exterior was coated to protect it and then support formers (see padding) were constructed prior to turning it over. The supporting cradle maintains it in the correct shape after turning. In this form it is beautifully shaped and quite lightweight. Lifting one end showed how flexible it was prior to fitting its internal ribs.


Pictures left and right show the one-fifth model's sweeping, elegant form.



Unruly Underlout

The stern underlout, as mentioned in Shipshape 7 had a natural fault (see below) that had not been apparent on the outside of the timber when selected. It gradually became apparent as the wood was cut away close to the pith. The central core of the tree is called the pith. All the branches stem from it and it is known to be a problem area. It is often softer and darker in colour than the surrounding wood and it is prone to splitting. It is a good idea to try to avoid using the pith if possible. This original underlout took many man/person-hours to carve, so when the fault gradually became apparent, the build team were faced with a dilemma. Should they try to repair the timber or should they begin to make it all again?

It would be expensive both in terms of finance and in the time taken to make a replacement – even IF a suitably curve timber could be found again. It might have been possible for this flawed underlout to work within the structure for a while, but then begin to deteriorate and need replacing. To change it, once it was surrounded by and rivetted to a number of other planks would be even more difficult and time-consuming. It was a very tough decision to have to make.

The fault was investigated to discover its depth by chopping a trench along it. Could good timber be set into it? An allied problem was the grain disturbance created by the fault. Instead of the grain running straight along the timber it had been deflected away at one side, so that it ran outwards on one of the slender ‘wings’ to which the next (garboard) strake would be attached. Rivets passing through this area would have little supporting strength from the underlout. Taken together, the damaged section around the pith and the weakened grain in the land (‘wing’) meant that the inevitable judgement had to be made, so the old underlout was removed and another was constructed to go in its place. To write these words takes almost no time at all, but to shape the new piece took well over 400hours of painstaking work.


When working with green timber allowance for shrinkage must always be considered. Wood does not shrink uniformly. It is a complex story, some of which was explained in Shipshape 7. Generally, the length of wood remains much the same. Most shrinkage is in the width, across the grain. There is some small shrinkage along the annual ring affecting the thickness. The timber for the first underlout was worked at the same time as the bow cutwater and the keel on either side of it, so they all shank at about the same rate, together. The new underlout had to be left a little too large, so that the shrinkage could catch up, when it could be trimmed to match.This allowance can be seen in the ends of the new underlout (right). It can also be seen in the lower picture, that the new timber was not perfect, but that calculations had been made to show that the imperfection would be removed once the new underlout was trimmed to size.


Broad Boards

At the very beginning in Ship > Investigation 2 > Plank widths I explained my immediate reaction to finding, for the first time, the image and technical information about the ship. I was sitting in the library of Imperial College, London, anxiously reading Bruce-Mitford’s The Sutton Hoo Ship Burial, seeking the single, most critical fact that I believed could make the reconstruction viable – or not. This is how I described it:

David Turner with a Broad-Board tree

Plank widths

Having found the book my immediate question was: How wide is the widest plank? This is because the most difficult dimension to achieve in solid timber is the plank width. The wider the plank, the bigger and older the tree has to be, in order to reach a size to produce that plank. I understood that planks cannot just be cut right across a tree trunk when it is viewed from the end. The grain pattern, the shape of the annual rings of the individual plank have to be carefully selected. This technicality can be explained in more detail later, because it is essential to understand it: the Anglo-Saxons did understand. I was thrilled to find on page 354 of Volume 1 a table showing plank or strake widths where the widest plank was 12 inches. There are many missing dimensions in the table, but the vast majority of widths were well below this figure. At a stroke I could see that it was achievable to find timber this wide. It may not be easy, but would be possible.

I need to add at this point that the 12ins (300mm) width is from rivet-to-rivet, so to allow for an overlap beyond the rivets, the widest plank needs to be about 14ins (350mm). This is a real challenge, as every fraction of an inch, at this size, becomes harder and harder to achieve. If you have read the numerous sections about timber in both ‘Ship’ and in ‘Shipshape’, you should, by now, have had the ‘explanation’ promised in this early writing.

Towards the end of 2022 timber from the National Trust’s Blickling Hall in Norfolk arrived at both Sutton Hoo and The Longshed. Hopefully, this timber will have the dimensions to produce those challenging widest planks used in the centre of the ship. There remains the dangers of internal faults, but riving the timber will be the only way to answer all the questions.


First Rivets and Roves

On the website Shipwright Reports > December, Tim Kirk proudly reported that a major landmark in the ship’s construction was reached on 21 December 2022 when the first iron rivet and rove linked two pieces of wood together. It had been just over ten years since I first met Martin Atkinson the director/property manager at Sutton Hoo in Tranmer House, on 1 March 2012 to suggest that the people of Woodbridge were contemplating a reconstruction of the Anglo-Saxon ship. He was positive and helpful in offering his assistance and he would have been overjoyed to see this moment, had he not tragically died so early. He, together with archaeologist Angus Wainwright, was one of the people critical to making this whole project possible. 

For information about the qualities of the iron needed see Shipshape 5. The wrought iron came from Topp & Co in North Yorkshire.