Mathew Baker's early concept of ship hull design, ca. 1570

[Waldemar]

 

A slightly different interpretation is also possible, closer even to the Mediterranean ways. The difference in the formation of the frame shape would be that the futtock template was moved and tilted first, and only then the toptimber mould (or the actual toptimber during skeleton assembly) was applied. This interpretation also explains well the slightly larger width of the actual ship compared to its design breadth. More clearly, this is the result of the Mediterranean method of frame forming combined with the adoption of a convenient round value for the futtock sweep radius.

 

In addition, it may be related to the anomaly already mentioned that the position of the midship frame does not coincide with the greatest width of the hull. Normally, the Mediterranean method was used in a non-graphical way, i.e. without the use of a plan, so it is likely that Baker may have tried to sort it out in this way while making his drawing.

 

It is now rather impossible to decide which of the two interpretations is correct (dimensional differences are rather negligible in this particular case). If one accepts the second, Baker's drawing would probably be the only drawing from the era that shows graphically the pure Mediterranean method. However, if one accepts the first interpretation, it could be considered to represent an intermediary between the Mediterranean method and the methods known from the slightly later other English manuscripts on shipbuilding.

 

 

 

Edited October 22, 2022 by Waldemar

[Waldemar]

 

By a happy twist of fate, I found in my home resources a better copy of Baker's plan. This was decisive. After a few minor changes and amendments, the pure Mediterranean method clearly emerged. Taking into account inaccuracies of the original plan and to a lesser extent its later distortions, I find the resulting reconstruction of the lines satisfactory, finally revealing the method used by Baker. 

 

Two further detailed comments need to be added to the accompanying drawing. Baker deliberately did not draw the narrowing line of the floor for the stern section of the hull, because it was not necessary: he simply took the relevant coordinates for the stern quarter frame from the bow quarter frame. Of minor importance from the conceptual point of view, the radius of the breadth arc for the bow quarter frame is smaller than the rest.

 

 

 

[Mark P]

Good Evening Waldemar;

 

Thank you for the new post. Interesting to follow! 

 

To return to the subject of the dead-rise, mentioned earlier in this thread: the deadrise is specified in some early contracts as 4". 

 

More interestingly, though, the Salisbury MS of around 1620 describes the deadrise as being necessary to avoid the floors being weakened by cutting the limber holes in their underside; the limber holes instead being cut out of the chocks which were set each side of the keel to form the deadrise. Perfectly sensible when you think about it. Presumably someone soon realised that the bonus was that ships built with a deadrise were more weatherly. 

 

All the best,

 

Mark P

[Waldemar]

 

Hi Mark,

 

Yes, this is all correct. Thanks to the improved quality of the reproduction of the Baker's plan, I was able to update the midship mould (shown below). As it happened, this only affects the deadrise, which was raised from 3 inches to 4 inches, at the expense of the keel height, reduced from 12 inches to 11 inches. Here I would also add that in the Harriot manuscript deadrise is specified for small ships at two inches and for large ships at three inches.

 

I have also made other similar minor adjustments here and there, but these are without major impact on the overall reconstruction. For example, the keel length was increased from 60 to 60.5 feet.

 

 

 

Edited October 17, 2022 by Waldemar

[Waldemar]

 

It will also not be out of place to show some examples found in shipwrecks and other written sources. It is clear that the deadrise curves for the main frame could be either concave or convex (the latter even mentioned in the Newton manuscript).

 

Mary Rose (from The Tudor Warship Mary Rose by Douglas McElvogue):

 

 

 

Basque whaling ship (from The Underwater Archaeology of Red Bay. Basque Shipbuilding and Whaling in the 16th Century, 2007). In this case the weakening of the floor timbers was to some extent compensated for by the T-shaped keel:

 

 

 

 

Two examples of midship moulds from Baker's manuscript, the one on the left with the deadrise drawn in (from The Gresham Ship Project. A 16th-Century Merchantman Wrecked in the Princes Channel, Thames Estuary. Volume I: Excavation and Hull Studies, 2014):

 

 

An instructive example also comes from the manuscript by the professional shipbuilder Manoel Fernandez, Livro da Traças de Carpintaria (1616). As can be seen, he designed the midship frame without the deadrise. It was then added later as a kind of supplement in a ready frame (compare two frames shown in the center of the draught).

 

 

And some other archaeological examples or taken from shipbuilding period manuals, as drawn by me a few years ago:

 

 

 

Edited October 22, 2022 by Waldemar

[Waldemar]

 

The following reproduction of a circa 1600 plan (from the Russian archives), possibly made by David Balfour, also confirms that contemporary designers did not treat the deadrise as an integral part of the designed frame shape, but rather as a necessary add-on. Again, the deadrise is outside the design grid, unlike all the rest of the frame profile.

 

It is fair to say that a similar approach is taken by modern researchers, forgetting or deliberately omitting deadrise curves from their graphical reconstructions. Therefore, when conducting one's own research, one cannot rely only on modern interpretations of historical material.

 

 

 

Edited October 17, 2022 by Waldemar

[Waldemar]

 

How exactly was this done? Many works have been written about the Mediterranean method, but by far the best single modern work is Le maître-gabarit, la tablette et le trébuchet. Essai sur la conception non-graphique des carènes du Moyen Âge au XXe siècle by Éric Rieth (1996).

 

Below are some renders of a schematic 3D model of the ship drawn by Baker, made using the Mediterranean method. The pre-designed frames are located only between the quarter frames. The shape of the other frames, close to the ends of the ship, was determined by longitudinal ribbands. These elements can be seen in the graphics below.

 

The overall shape and specificity of the lines obtained bear a striking resemblance to the hull shape of an Iberian whaling ship, the famous Red Bay vessel.

 

 

 

 

 

 

Edited October 18, 2022 by Waldemar

[Waldemar]

 

To give an idea of how the pre-designed frames were shaped in full scale in this method, below an illustrative diagram. No (detailed) plans drawn on paper were needed, just few simple geometric scales/devices used on the mould loft.

 

 

 

[Waldemar]

 

While rotating a wooden pattern, even of the most complicated shape, is extremely easy and convenient in full scale on the mould loft, the opposite is true when it comes to draw such a shape on paper by conventional methods in multiple copies, and each copy at a slightly different angle (using a paper template in scale would be not quite practical either).

 

Therefore, when design plans on paper came into use in England, the Mediterranean method of rotating/tilting the futtock mould had to be substituted by a more suitable way. There is also one telling comment by Baker himself about his fellow shipbuilders, that they did not understand the purpose of rotating the moulds. Even if this remark may be a little unfair, it quite clearly suggests that English shipwrights quickly developed or used other methods.

 

 

 

 

Edited October 19, 2022 by Waldemar

[Waldemar]

 

Thank you for watching so far. Should anyone wish to mention in their publications or refer to this interpretation of Baker's drawing, my name is Waldemar Gurgul.

 

 

[Waldemar]

 

In the Danish archives there is another interesting 'complete' plan of a ship with a keel length of 68 feet, design breadth of 21 feet and the depth in hold of 9.5 feet. Its draft of 8 feet, written and drawn on the plan, is well suited for shallow Danish and Baltic waters. Contemporary descriptions on the drawing are in English, and, as in Baker's drawings taken from his Fragments, here too the ship is actually wider than the design grid. For this reason, I carried out a reconstruction of its hull shape using the Mediterranean method. 

 

 

Yet, interestingly, in this case the run of both narrowing lines was chosen in such a way that rotating the futtock moulds was not needed at all, only sliding vertically and horizontally is enough. Either way, the actual line of greatest breadth (of the reconstructed shape) does not coincide with the designed narrowing line of greatest breadth all along the length of the hull. I have shown both lines in different colours in one of the renders below.

 

 

 

 

 

It is difficult to say who created this plan, but it is known that there were a few shipwrights of Scottish and English origin working in Denmark during this period. As can also be seen from both examples (i.e. from this one and Baker's), the Mediterranean method allowed for nice, smooth hull lines to be easily achieved, which certainly made it very popular.

 

 

 

 

 

 

Edited October 20, 2022 by Waldemar

[Waldemar]

 

It is possible that one more element of the utmost conceptual importance can be better explained (I am trying to understand this drawing as I go along), and that is the exact location of the geometric device used to limit the rotation of the futtock template.

 

This is shown on Baker's draught as a short horizontal line, which I interpret as being placed on the level of the first, undrawn deck. The measured distance between this first deck and the (drawn) main deck above is 5.5 feet.  And perhaps more importantly, the height of the first deck above the base line is exactly 8 feet.

 

It would be another hallmark of the Mediterranean methods, which were originally developed for the single-decked vessels like galleys and other low board craft. In this very case, however, it is a rather confusing application, as the narrowing line of the greatest breath (visible on the plan view) is roughly on the level of the main deck, and not on the level of the first deck. No wonder it was soon replaced in graphical methods by a more lucid means, and at the same time more adapted for larger vessels.

 

This is illustrated in the graphic below.

 

 

 

Edited November 21, 2022 by Waldemar

[Waldemar]

 

And to conclude this thread, below is a reproduction of a ship plan drawn by Venetian shipwright Stefano Conti (original in the Archivio di Stato, Venice), to be built according to the Mediterranean method. Just one comment here: in its place of origin, the Mediterranean method of frame moulding (i.e. non-graphical!) was apparently still in use even for large ships as late as around 1700.

 

 

 

Edited November 21, 2022 by Waldemar

[druxey]

Fascinating topic/discussion! What relationship is there - if any - between the span across the supposed first (lower) deck and the keel length and moulded breadth?

[Waldemar]

 

Thank you very much Druxey!

 

You have drawn my attention to the fact that, unlike many other examples, here the moulded breadth (taken as the width of the master frame design grid) equals the breadth at the main deck level. And Baker is known to have experimented with different proportions or configurations, so perhaps that was his intention in this case.

 

Either way, here are the desired dimensions:

 

keel length: 60.7 (~60) feet
breadth at the 1st deck: 21.567 feet
breadth at the 2nd (main) deck: 24 feet

 

The only round ratio I have been able to spot is that of the breadth at the 2nd deck to the keel length (and even this only after rounding the keel length to 60 feet), which is 1 : 2.5.

 

 

[druxey]

Thank you, Waldemar. My impression is that ratios were very common - often decimal rather than duodecimal.

[Waldemar]

 

Bingo! A small update.

 

moulded breadth: 24.3 feet

 

24.3 x 2.5 = 60.75 feet

 

The calculated value of 60.75 feet may be the true length of the keel, which is quite elusive to measure very accurately on the drawing.

 

 

 

Edited November 21, 2022 by Waldemar

[Waldemar]

 

1 hour ago, druxey said:

My impression is that ratios were very common - often decimal rather than duodecimal.

 

True. It is usually a mix of both, more or less. Now I should perhaps make an update of all drawings, at least where the keel length and its derivates are given, but that is much work and dimensional differences rather small, quite within working tolerances. 

 

Either way, I am satisfied that the conceptual idea of Baker's ship has been hopefully found and its components identified on his plan.

 

 

Edited November 22, 2022 by Waldemar

[Waldemar]

 

This is hopefully my final interpretation of the master frame. It is essentially the same as the previous one, just supplemented with decks and some additional dimensions. Note the width of the main deck camber.

 

 

 

 

Edited November 22, 2022 by Waldemar

[druxey]

I note some other near ratios on your latest iteration, Waldemar! Might they be exact, but for distortion over time and reproduction, I wonder? That 11.5 is sooo close to 12, for instance, as is 24.3 close to twice 12!

[Waldemar]

 

🙂

 

Druxey, please feel free to judge this for yourself in the attached drawing. I just added a blue line half a foot away from its original location.

 

It may not make much conceptual difference, but the deck camber is most likely 1/30th of the ship breadth. I look forward to your comments, which may again have the chance to provoke some further discoveries.

 

 

 

[Waldemar]

 

2 hours ago, druxey said:

I note some other near ratios on your latest iteration, Waldemar! Might they be exact, but for distortion over time and reproduction, I wonder? That 11.5 is sooo close to 12, for instance, as is 24.3 close to twice 12!

 

I agree with Richard Barker's conjecture that originally the greater ship's breadth from the master frame design grid was the result of the addition of another deck to single deck ships, without modifying the method (sequence) of construction.

 

However, in this very case the difference in these dimensions is too small. Therefore, I believe that the moulded breadth of 24.3 feet is simply due to the selection of round values for the master frame sweeps radii. In other words, Baker was actually aiming to get the moulded breadth as close as possible to the width of design grid (24 feet), but at the same time he preferred to use round values for the master frame sweeps radii. Baker then measured and multiplied these 24.3 feet by 2.5 and thus obtained the keel length, which is 60.75 feet.

 

This could be a sign of evolutionary changes in the way ships were proportioned and designed.

 

 

Edited November 22, 2022 by Waldemar

[B-Ram]

Hello everyone,

 

I am new at all of this, but I just wanted to say that I am enjoying the conversation very much and learning a great deal of information.

 

Thanks, 

Bill

[Waldemar]

 

Hi Bill,

 

Thanks for your post. If you have a specific idea or question, go ahead and post it here. It might even be a chance to clarify some previously unseen problem.

 

 

[druxey]

I've never seen a larger, clearer version of the plan that you are overlaying on, Waldemar. That figure written on the plan; does it read 5.2634? If so, what does this signify?

 

I've been pondering your illustration of deadrise/limber channels. It begs the question as to when (in naval ships) the channels were moved to run above the floor timbers. The variants you show are fascinating, especially version E that runs centrally above the keel.

 

[Waldemar]

 

2 hours ago, druxey said:

That figure written on the plan; does it read 5.2634? If so, what does this signify?

 

It actually reads 263 1/4, and it is the radius of the wale as written on the original plan. Please also take a look at the graphic in post #16. There is some confusion, however, because I initially worked with a rather poor copy of Baker's drawing and some of the figures in my first sketches are not quite accurate.  These need to be updated. Yet still the value given by Baker is not accurate either as he has drawn the wale with an actual radius of about 247.6. To the left of the number is the quadrant marking.

 

 

2 hours ago, druxey said:

I've been pondering your illustration of deadrise/limber channels. It begs the question as to when (in naval ships) the channels were moved to run above the floor timbers. The variants you show are fascinating, especially version E that runs centrally above the keel.

 

These are just samples taken from archaeological finds of the early modern period. From memory: variants A, B and C are quite typical for Dutch ships, D would be suitable for French ships, E for Iberian and English vessels and F as found in one of the Mediterranean wrecks. It should be taken of course only as a general outline. The main purpose was to show how the configuration of the limber channels would have varied at this time.

 

 

Edited November 22, 2022 by Waldemar

[druxey]

Thanks, Waldemar. I did understand that the limber channel drawings were representative, based on findings.

 

Also I appreciate the clarification on the figures written on the cross section. Is the difference between the number and actual radius perhaps due to distortion over time?

[Waldemar]

 

I estimate that drawing inaccuracies have a much greater impact, say, in a relationship like 70:30 or even 80:20 to paper distortion.

 

And in some cases to the point where it is sometimes difficult to decide whether to correct or repeat them, because these inaccurately drawn lines have an impact on subsequent geometric constructions. Paper distortions alone would be fairly easy to overcome....

 

 

[trippwj]

What is the rough scale of the drawing?

 

Is it possible that as a representative as opposed to true "scale" drawing the line of the wale lost fidelity due to the need to extend a greater distance to draw the line than was available on the wirk surface?  If the intended arc was 263" actual scale of the drawing that would be some 22 feet to draw it accurately...

[Waldemar]

 

2 hours ago, trippwj said:

What is the rough scale of the drawing?

 

Is it possible that as a representative as opposed to true "scale" drawing the line of the wale lost fidelity due to the need to extend a greater distance to draw the line than was available on the wirk surface?  If the intended arc was 263" actual scale of the drawing that would be some 22 feet to draw it accurately...

 

This is perhaps an opportunity to clarify issues of size, scale and dimensions on this drawing.

 

The actual size (scale) of a paper plan is irrelevant as long as there is a linear scale on it. And there are several linear scales in Baker's drawing. They are all the same length, but represent 24, 30, 40, 20 and 44 real feet. These give, respectively, a ship with a keel length of 60.75, 75.94, 101.25, 50.62 and 111.37 real feet. Incidentally, this means that the design of this ship was conceived as a reference model, i.e. by selecting or creating any particular linear scale, a ship of the desired size could be built.

 

Now, all dimensions on this drawing are given in feet, which correspond to the linear scale of 24 feet. This includes the value of the wale radius, which of course is not 263 inches but 3159 inches (i.e. equal to 263 1/4 feet).

 

Let us assume that the paper drawing was made at a scale of 1:48. In order to draw an arc with a radius of 3159 inches, it has to be reduced 48 times, which gives 65.81 inches. This is 5.48 feet, not 22 feet.

 

But there's more to it than that. It is highly doubtful that such large arcs would be drawn with compasses in practice. In such situations, special instruments similar to bows, tensioned with a chord or screws, were used, preferably in combination with appropriate geometrical or mathematical methods for determining points on the arc.

 

Thank you.

 

 

Edited November 23, 2022 by Waldemar