A personal account – P. Constantine
The Hull Shape
Paul Handley began with the 3 major drawings of the Sutton Hoo ship from which its form must be calculated.
They are described more fully in Investigations 2 and 4 on this website.
i) Fig 185. Science museum Provisional plan drawn under the guidance of Lt. Cmdr. Hutchison Sept 1939. An authenticated, signed-off copy is in Ipswich Museum where it was discovered by Joe Startin.
ii) Fig 324. Early 1973 drawn and annotated by Colin Mudie, called Drawing 202. A more advanced version of this drawing numbered 202.8 was available to the research team.
iii) Fig 325. The Plank and Rivet plan which is a British Museum drawing.
There was also some reference to the publication Das Nydamboot by Angelika Abegg-Wigg.
To understand the following diagrams, it is necessary to know that when a hull is represented on paper the front view, looking at it from the bow, will show a series of shapes at different places, all evenly spaced from each other, along the length of the hull. The first drawing (right) shows how this works. The horizontal lines can be ignored as they are just for measurement. This drawing only shows half of the hull with one of the shapes, number 8, highlighted. The low numbers represent the shape nearer to the bow and as the numbers increase, they represent the shape closer to the middle of the ship. The shapes are actually cross-sections of the ship.
The Sutton Hoo ship has 26 ribs running across it to support the planking, so if the shapes coincide with these ribs the drawings will show roughly what shape each rib would be. This also means that rib 13 would be about in the middle, also called the width, or the beam, or the midship station of the ship. The ribs do not exactly coincide with the shapes (see later explanation under 'Ribs' at the bottom of this page)), especially at the ends, but it helps the imagination to think of them as ribs at this stage.
To compare the different drawings, mentioned at the top, with each other, they have to be drawn to the same scale. It is of interest to see that the naval architect Edwin Gifford had carried out this exercise when he was calculating the shape of his half-length longboat Sae Wylfing. Some of his drawings, could be exactly overlaid on other drawings, as they were produced on transparent material (left) so that he could decide whether the different versions agreed. Bringing drawings to a common scale is now a relatively simple process by using computers. He chose to bring the drawings to 1:20 scale. The one which is pointed at the top is the Colin Mudie 1973 version. the one beneath it on plain paper is the Science Museum 1939 drawing. I have not placed them precisely in alignment so that the two different drawings can be seen.
The two halves of the ship
On the first half-hull drawing at the top-right of this page the front shapes of the ribs are shown, but on the Gifford drawings both sides of the ship are shown. It is very important to realise that although these sides look to be very much the same as each other - they are not. At first glance the total drawing just looks like the ship’s shape - the same both sides, but this is not so; the shapes are different, especially towards the ends of the ship.The way these drawings are made is to show the shapes of the ribs at the front of the ship on the left side, but the shapes of the ribs at the back of the ship on the right side. The cross-sections on the right side will be numbered from about 14 at midships to 26 at the stern. This saves time in drawing and as both sides of the ship should be the same as each other it is just a matter of repeating one shape as a mirror image on the other side when making any rib. This can be illustrated by a model made for the local cub group. The front cross-sections were drawn across to produce both sides the same and then cardboard models of the shapes were made. Here the shapes are loosely stacked upon each other next to their drawing. The complete model can be seen at the bottom of the 'Using Sae Wylfing' page on this website.
Checking the shapes
Paul Handley looked at the drawings and took measurements from each of them to compare them. He had loaded the cross-sections from each drawing and overlaid them to compare them for changes and discrepancies. If he found alterations from one to the other he had to attempt to discover why these changes had been made. One feature immediately jumped out. It could be seen
The 1973 drawing on the right does not have the smooth curvature of the 1939 drawing on the left.
There is a corner in the bottom of the hull
in the shape of the ship in the Colin Mudie 1973 drawing – as published in Volume 1. The earlier 1939 drawing showed the ship as having a rounded shape, but Mudie’s drawing showed it having a sharp knuckle at the turn of the hull. Where had this come from? Mudie was of course, trying to pull the damaged end together (see Investigation 2 this website), but it ought not to have produced a radical change of shape to the bottom of the ship. However, it was known that when in 1974 Mudie continued to work on the shape, the sharp knuckle had become almost non-existent, so it was only a transitional thought. The unfortunate part about this is that many people over the years have accepted this shape as being the ‘official’ hull shape because it is in Volume 1. Precisely-accurate models have been made for museums based upon these drawings. It needs a trained eye to spot this change, but since 1973 nobody had mentioned it. Since Paul Handley saw it, only two experienced shipwrights have seen it without being briefed first.
The shape change was not the end of the story, because when the other cross-section shapes were examined, in many cases their shapes had changed even more. The clearly rounded curves have all-but disappeared towards the ends of the ship.
Mr Handley took a copy of the 1939 Science Museum drawing with the numbered, curving shapes on the left-hand side. He copied these curves across to the right-hand side (ignoring the numbers 15 -25 above), so that he could see the completely recorded rib shapes in 1939. Then next to them, he drew the same ribs as found in the Mudie drawing which, by comparison, are much straighter.
Mudie was asked to pull the apex of the stern together. He has done this, but in so doing has lost the curvature, particularly on the higher part of the hull. Some of his cross-section lines are almost straight. Towards midships his sections return to being close to the original curved form. It was thought unlikely that the straightness of his sections on the upper hull could have been the shape of the original craft which are also represented by very curved lines in the Science Museum’s Fig. 135 lines.
All lines and sections are interlinked
It is important to realise that it is not possible to pull some parts of the hull together without affecting other parts of the craft. Mr Handley also had the benefit of referencing the photographic evidence of the planking and rivet patterns. He was able to take drawings from photographs to check the gently curving section lines and then use his computer on the upper part of the ship where the stern has to be pulled together. Pulling curved, hull components back together does not straighten them. However, it must not be thought that all of the sections of the badly damaged hull would produce a fair hull shape, some parts will be more reliable than others. Whilst checking photographs Mr. Handley noticed that comparison of the profile shape (viewing them from the side of the ship) of the bow and stern suggested that the stern end sections may have dropped down as well as spreading as the soil settled. They were checked by scanning the images and overlaying them, to compare the profiles.
Developing the hull form
The following is a slightly technical description of the way that decisions have to be made by anyone working with computers and how those decisions can affect the final outcome. This relates to the second paragraph under Computers? in Investigation 4.
By detailed examination of all three sets of drawings certain aspects can be selected as being ‘most probably’ accurate representations of the parts of the ship where they are found. This is not true of every section, in every drawing. By taking dimensions only from the representative ‘probably accurate’ sections, a series of offsets-measurement markers can be recorded for key frames. Initially these markers are points in space measured vertically and horizontally relative to the keel position. When a curved section line is superimposed on them it can be seen whether the markers lie on the line or are away from it on either side. Using control points the computer can manipulate the curved line in an attempt to explore whether a better fit to the offset-markers can be achieved. It is not simply a matter of making the markers coincide with a single curved line, because all sections and all lines are interlinked. The control points need to gently pull or push all offsets and all sections together, at one time. The curved lines must produce a smoothly faired form. When a preliminary solution is arrived at, further dimensions can be taken from it to compare with the original 3 sets of drawings to see, for example, that the beam dimension at different points is still in agreement with them. More intermediate cross-section, offset markers for other parts of the craft can be taken and checked to see if they too agree with the 3 core drawings that are being cross-referenced. Eventually, a close fit to many points on most drawings can be arrived at and then tested for proximity to being an agreeable solution. The form produced should be a smooth shape without humps and hollows
In the computer the faired form of the hull can be checked by shining ‘virtual’ light along its surface whilst rotating it. Any imperfections will be visible in a similar way to looking along any shiny surface at the reflected light pattern. The hull shell resembles a long canoe or rowing shell. It is only the 'skin' of the craft that is being modelled at this stage.
Extracted from Paul Handley's comments at the time of his investigation (in italics)
They drew what was there.
The 1939 Science Museum team drew what was there as shown by the rounded end. Some things were not properly understood and the suggestions and speculations made hurriedly at the time may subsequently have proved to have been incorrect, but this does not necessarily reflect on the accuracy of the drawings that were made. We have no raw data from the dig, no recording notebooks from which the 1939 drawings were made.
The next comment is of particular significance for it draws attention to another vitally important observation. Since the publication of Volume 1 the two drawings called Fig 324 & 325 at the top of this page have always been casually regarded by almost everyone, as being the work of Colin Mudie. They are printed together. His name appears on one of the pages. The one on the right with all the planks and rivets is the iconic image that everyone understands. Haven't the rivets on the right just been added to the lines plan on the left? The answer is 'No!' Two totally different drawings from different sources have been 'lumped' together.
The two sets of images at page 434.
1. The Colin Mudie lines are faired. He was following advice from Rupert Bruce-Mitford to bring the bow & stern shape together as they would have been in the original ship, but in fairing the lines the drawing introduces a hard turn to the bilge. There is no original evidence for this.
2. The accompanying right-hand drawing based on the photographic record and the rivet pattern has irregular lines and is not faired. These drawings do not have the hard turn to the bilge of the Mudie drawings.
Here Handley is saying something quite explosive. Drawing 325 has not been faired (smoothed to make all shapes agreeable). The rivet pattern does not work. The diagram is more of an 'artist's impression'. It is just illustrative, not accurate as is drawing 324 that does belong to Mudie on the left. Volume 1 has a kind of chronology, that is to say, in general the story begins at the beginning and progresses in about the correct order to the end. Taking an approximated 1939-based image and placing it adjacent to a precise 1973 image in the part of the book dealing with predominantly post-war investigations has had the effect of misleading most readers.
Joe Startin later pointed out that the two drawings are not even to the same scale.
Handley refers to Fig 325 as follows:
In Figure 325 the lines (the archeological reconstruction) are a lot fuller at section 7 and 8 than the Colin Mudie lines, but very close to the same shape at the mid-ships sections. In this museum plan, lines are much fuller all round, including the mid-ships sections, which would help the boat carry load and make it more stable.
In other words this drawing published opposite to Mudie's drawing is actually based upon Fig 185, the hurried, but unfaired 1939 representation of the ship earlier in Volume 1.
Round and flat bottomed hulls
It is generally true of all craft types from sailing yachts to fishing boats that the more gently round the hull shape the more it is intended for rowing. The more towards the wineglass shape the hull moves, the more it is intended for sailing. The greater the beam, the more stability for sailing. The flatter the bottom, the greater the increase in stability, as seen in barges. The rounder shape in the 1939 drawing indicates a rowed boat. Colin Mudie’s hard turn to the bilge producing a flatter bottom moves towards a sailed-ship form. It is difficult to see the evidence for this interpretation.
Distortion and shrinkage in the ground.
It is worth considering possible distortion. Volume 1 explains how the thickness of the planks was reduced to a single thin line of material. This affects the actual size of the ship when applied to its total dimensions.
Different pressures at depth.
A bucket full of beach sand will settle and become more compacted towards the bottom. This situation might have been possible with the buried ship. There will be less compaction towards the surface. What effect might the pressure of the mound have had on this situation? The spring action of planks in a boat hull will push outward for, say, the first 20-30 years. If this situation is ignored for 1400 years, sideways spread is possible with the topsides being pushed out.
The bow of the ship was different from the stern
Bow section fullness
When the forward sections of the museum plan are compared with the aft sections it can be seen that the forward sections are considerably fuller. This difference in fullness cannot be seen clearly in the photos or as a result of taking measurements off the frame stations in the photos, which was done by measuring the camber of hull segments at frames.Therefore an assumption has been made that fullness in the bow sections was unintentionally exaggerated in the museum plan, probably as a result of expediting the preliminary lines fairing process. Also, it is quite likely that at some sections the keel would have been pushed up relative to the side sections as the boat settled, thus increasing the curvature in these sections. The bow section fullness has therefore been reduced a little in the faired computer hull plan, although the forward sections remain slightly larger than aft sections.
The computer hull model had a small amount of keel rocker (curvature to the underside of the keel when the craft is viewed from the side).This feature was retained throughout the fairing process as it has been suggested by Scandinavian experts that a boat of this type would have had some keel rocker and it is slightly easier to generate a fair hull (or to plank a hull) with a small degree of centreline rocker. This rocker line refers to the line of intersection of the garboard planks (the first planks on either side of the keel) on the centreline rather than any additional projecting keel timber.The rocker is a fair curve. It can only be quantified between longitudinal limits. Between Frame 6 and 20 the maximum rocker is just over 5 inches.
Rocker is perhaps to be expected for a rowing-only craft, as it both reduces the wetted surface area in the ends and improves maneuverability in restricted areas. Some rocker would also enable the bows of the boat to get closer to the shore edge when running up a beach to enable crew to disembark without getting too wet. Further, a small amount of keel rocker would enable the boat to be freed and turned more easily after grounding for this reason. So unless there was a need for deeper ends and keel depth - for lateral resistance for sailing - it seems reasonable that these features would be absent and the hull would have a small amount of rocker with shallow keel timber.
Ribs are not included at this stage. It will be noticed that the ribs are set at 90° to the keel so they become angled as they move towards bow and stern. This means that they are not necessarily the same dimensions and form shape as the vertical cross-sections. In addition Mr Handley pointed out that frames 1, 2, 23, 24 and 25 have non-uniform spacing as measured on the museum drawing for these frames. There will be more information about ribs later.
Are not included at this stage although Mr Handley provided details, They will be covered later.
It's interesting to see the outcome of Paul Handley's investigations. There are many more drawings and explanations, but here in a nutshell is a drawing that shows some 1939 sections in dotted lines with Paul Handley's sections, taking everything into account, in solid lines. The reasonably close correlation can be seen.