The Dutch 72-gun ship ca. 1690 – the apogee of Dutch ship design of the Classical Age

Waldemar · June 17

It should be clarified at once that this plan, tentatively dated by me to the end of the 17th century, is one of the last of the era before the widespread adoption of design diagonals, which in the Netherlands occurred in the third decade of the 18th century at the latest (see as to this Ab Hoving & Alan Lemmers, In Tekening Gebracht. De achttiende-eeuwse scheepsbouwers en hun ontwerpmethoden, 2001). In historical terms, this is the period of the Third Anglo-Dutch War (1672–1674), the War of the League of Augsburg (1688–1697) and the War of the Spanish Succession (1701–1714).

As it seems, this very plan was not previously widely known and was kindly provided to me for examination, in terms of the design method used, by Ab Hoving. This is an exceptionally fortunate circumstance for several reasons, and I am most pleased that it demonstrates classical prediagonal design methods in their latest, most advanced form and for the most demanding projects, that is of a dedicated warships.

It is apparent at first glance that conceptually this is a more sophisticated design compared to the civilian designs presented so far (Samuel 1650, Witsen’s pinas 1671, Rålamb's boyer & fluit 1691), while it is more similar in this sense to the Dutch capital ship ca. 1665 (Hohenzollern model), the French (Atlantic) heavy frigate design by Chaillé 1686 and the designs of 1679 by Hubacs of Dutch origin, boasting dozens of designed and built ships to their credit.

* * *

The primary aim of the project will not even be to search for all the (main) proportions of the design parameters, as many of them were quite arbitrary (as they still are today), but rather to search for more universal procedures applied for shaping the body of the hull, quite independent of these proportions.

The plan (Archives of the Nederlandsche Marine):

The drawing is originally captioned: Frigate ship long between posts 165 feet for 72 guns [seen] from inside.

Contrary to this caption, this is a design for a ship 156 feet long, not 165 feet, as indicated by the size of linear scale (30 feet long; a foot divided into 11 inches), and according to the metric scale added nowadays, the plan was drawn at a scale of 1 inch = 5 feet 7 inches, or 1:62.

At this scale, the drawn ship measures 156 x 42 x 15 feet (length x breadth x depth in hold), which corresponds fairly well to the dimensions of almost the entire, long series of 28 2nd rate ships built under the 1685 programme.

As one would expect from a dedicated warship, the hull has quite a sharp entry, as well as overall shape compared to merchantmen designs. In design practice, this is achieved primarily by increasing the height of the deadrise, or in other words, the line of the floor. According to the design method and customs, only the leading (conceptual) frames are pre-designed. The shape of the remaining, filling frames, could be or had to be (especially at both extremities of the hull) determined according to the ribbands, already installed during the actual construction.

However, not to rule out any possibility, it is also possible that the drawn linear scale is out of sync with the rest of the drawing and, in addition, slightly longer than it should be, and that the ship length of 165 feet, given in the caption, is correct. In that case, the scale of the drawing would be 1:66 and the hull breadth would be 44 1/2 feet (giving in this way, for example, the dimensions of Gouda 1719, built in Amsterdam by van Reenen).

This does not really change the essence of the project (recognition of the design method), and in the context of the dating and design practices of the time, it may be indicative of the repeated use of the same designs to build ships of different sizes. To put it another way, the plan may have been drawn at a scale of about 1:62 as for a 156-foot ship, and after the scale change, it became the design for a 165-foot ship.

Keel assembly & longitudinal division

Note: some of the dimensions below are actually redundant, but have nevertheless been left on the diagram for better readability.

* * *

The keel has been drawn quite realistically – on the plan it has a deflection with the greatest value at the forward master frame, amounting to 1/3 of the keel height, and is tapered towards the stern to 4/5 of the total height. The stempost rake is very small, being only 4 x sternpost rake (measured to the supposed rabbet line). As late as 1737, the French shipwright-designer Blaise Ollivier reported that on Dutch ships it was usually larger, 6 x sternpost rake.

To determine the position of the leading (conceptual) frames, the ship was divided into nine equal parts. For the fore part of the ship an additional division of four parts has been set up.

The contours of the 'virtual' single master frame are not even defined on the body plan, nevertheless the position of this frame is essential in this design to establish the nominal (largest) values of all narrowing lines (breadths/widths) on the plan view. To put it another way, it is this actually unbuilt in real construction, single 'virtual' master frame that is widest at all levels – at the 'flat' level, at the level of the line of greatest breadth and at the level of the toplines that define the upperworks. Its lengthwise placement have been chosen in such a way that it falls at 1/3 of the keel length.

For the construction of the physical ship in real scale, two 'twin' master frames were defined. The fore master frame falls at 1/3 of the length of the entire hull, and the aft master frame at an equal distance from the single 'virtual' master frame (see diagram). By design, the contours of these twin master frames are not quite identical, the largest difference being about 1/4 foot, nevertheless similar enough that in carpentry shipbuilding practice they could be treated as identical without much detriment to the quality of the end result in terms of its conformity with the design.

The diagram also indicates the likely placement of the two quarter frames. From these positions, going towards both extremities of the hull, the contours of the frames already have a different geometric structure compared to the central frames between these quarter frames.

Main design lines (rising & narrowing curves)

It must be said at the outset that the shape of all the elliptical lines in this design, obtained by means of the popular geometrical transformation called mezzaluna, and employed in its numerous variations, is so closely coincident with the ideal shape of the arc of a circle that it is almost certain that the use of these elliptical curves was merely the result of the designer's lack of a compass of sufficiently huge size, rather than a deliberate effort to obtain some particular curvature different from the ideal circular arc. In drawing practice, it was in this way that a number of points lying on the desired curve were determined, and these points were then connected using drawing instruments – either templates of fixed curvature or battens of variable curvature adjustable by a screw or string.

* * *

Line of the floor (green)

This is essentially the most important design line for any ship of the period. At the main frame, the deadrise is large, appropriate for warships, 1.5 feet measured from a realistically curved keel, giving an inclination of 1:8.75 in relation to half the floor width. The total width of the floor almost matches the 'standard' value of 2/3 of the hull breadth, and in this particular case is 5/8 of the breadth. At the stern, the line of the floor terminates normally at the height of tuck (where the fashion pieces join the sternpost), At the bow, at the intersection with the rabbet line, the height of this line is not great, however, the sharp gripe is still obtained by the specific way in which the bow frames are formed, different from the central frames.

In the vertical plane (sheer view), the lowest point of the floor line is at the fore master frame, while in the horizontal plane (plan view), the greatest width of this line falls already in a different place – at the single "virtual" master frame. The section between the single "virtual" master frame and the fore master frame is a mirror image of the corresponding section of this line from the aft side. Both observations also apply to all other design lines.

After examining several cases from the era, I can probably say that the curvature type of the line of the floor is very characteristic of Dutch designs, i.e. a logarithmic curve for the fore part of the hull and a (supposed) circular arc for the aft part.

Line of the greatest breadth (blue)

Apart from the (quite accurate) approximation of this line to the arc of a circle using a mezzaluna, this line does not differ in any way from the standard in terms of the way it is drawn. However, it is possible to point out the relatively quite significant height of this line at the master frame (almost 4 feet), with the effect of increasing the transverse stiffness of the ship and thus enabling the lowering of the gun ports (and therefore the rest of the above-water part of the hull for better weatherliness), and also the noticeably sharper shapes of this line at the bow (in top view) compared to merchantmen, for greater speed.

Lines relating to the upperworks (violet)

The same concern for improving weatherliness by lowering the height of the upperworks is also visible in the case of the toplines – at the fore part of the ship they are lowered by about half a foot compared to the aft part of the hull. In the top view, the designer partially spoiled these lines (or just ignored some of them while drawing the frame contours), especially in the stern part, which is not surprising considering that these were the least important elements of the design. As a consequence, their run in the top view on the below diagram was partially corrected based on the frame contours. Also, in these circumstances, the lowest of the three toplines is not necessary at all to define the shapes of the upperworks (and even harmful), because the designer, when drawing the frame contours, used a different method to determine the radii of the upper breadth sweeps anyway, i.e. based on the respective frame widths.

Perfect shapes in the sense of geometric smoothness. And without any correction with diagonals and waterlines whatsoever! Just the correct formation of the frames. Yet, there is something else incomparably more important, about which later.

Now, just that this plan is a real, unique gem with probably no equivalent, because it shows the true, advanced engineering face of Dutch shipbuilding of the Ruyter era, and hitherto hidden behind the façade of Witsen's and van Yk's essentially carpentry-oriented works, both touching hardly on conceptual design issues, unlike, for example, their English contemporaries in the field.

Before continuing properly, a bit of fun with the renderings. I am constantly amazed at how perfectly smooth the hull lines have been achieved by the designer through the skilful selection of the individual sweeps radii for all the consecutive frames.

Gun ports are not evenly spaced, most likely for structural reasons (knees, beams, hatches etc.). In some of the renders, the ship has been trimmed four feet aft.

Well, now there's bound to be a Nobel Prize 🙂. It was enough to look at something else than the works of Witsen and van Yk themselves (these, although unrivalled in structural and carpentry aspects, are nevertheless extremely poor in terms of design methods). It was also enough to finally break with the uncritical endorsement of modern studies by academic fellowship, displaying a notorious tendency to conformistically rewrite increasingly inadequate and anachronistic theses from each other's 'official' publications, and to look directly at the sources without such mediation.

But to the point.

Analysing the shapes of the frames on this plan, I finally uncovered that not one but two lines of the floor were used in this design, nevertheless, for some reason, the second one was not drawn at all on the sheer view (or was later erased). This undrawn/erased line of the floor (in black on the diagram) was used for fixing the lower edges of bilge sweeps. In contrast, the one left on the original drawing (green in the diagram) shapes the bottom of the hull and also materialises the edge of the 'flat', as the ship was intended to be built using the bottom-first method. They have both the same deadrise at the fore master frame.

Among other source indications, this plan is one of the most important pieces of evidence that the leading frames (could) have been pre-designed in their entirety also in this very method (i.e. bottom-first), especially for more complex warship designs. Both of these lines are used in a different way to their counterparts in the Mediterranean/English tradition.

The way in which the contours of the frames are formed is shown in the graphic below. Apart from the frames at the hulls's extremities, i.e. #1 and #9, the order is as follows:

– first a line of the „flat” connecting the keel to a point on the line of the floor of the "flat" (green colour),
– then a futtock sweep with a radius of 3/4 x the respective frame breadth,
– then a bilge sweep tangent to the futtock sweep and tangent to the level of the second line of the floor (black colour),
– then a reconciling sweep with a fixed radius of 2/3 x max. breath, joining the bilge sweep and the line of the „flat”, tangentially on both sides,
– finally an upper breadth sweep with a radius of 1/4 x resp. frame breadth and a toptimber sweep in accordance with the longitudinal lines defining the upperworks (violet).