ÐÓ°ÉÔ­´´

WHERE CAN be very few British men over 30 who have never built models
using an Airfix construction kit. In 1947, Nicholas Kove solved the tax
problems of his plastic comb company by producing a plastic model kit of
a Fergusson tractor. The company, Airfix, had been the first British firm
to deploy a machine for injection-moulding plastic materials. It now blazed
the trail in injected polystyrene model kits. By the mid-1960s, Airfix was
a part of every schoolboy’s life.

Most former enthusiasts have long since taken down their 1/72 scale
aircraft from the bedroom ceiling and thrown them out. Despite the popularity
of the kits of yesteryear, comparatively few schoolboy modellers now design
and build real aircraft or ships, but at least one naval historian – myself
– is an unashamed product of the Airfix company. Though kit companies have
yet to satisfy my interest in early submersibles, my mildly obsessive interest
in hull dynamics was originally triggered by a sequence of Christmases and
birthdays when I was presented with large Airfix kits of sailing warships.

The Airfix Classic Ships series had two big drawbacks: they were rather
expensive and almost ludicrously fragile once completed. They did, however,
teach the attentive modeller the basics of ship layout and the accompanying
jargon. Until the early 1970s, when distribution throughout Europe forced
Airfix to supply instructions to modellers in the shape of wordless diagrams,
an Airfix instruction leaflet would contain, in addition to the assembly
diagram, such written instructions as, ‘locate and cement riding bitts (76-7)
to forecastle deck (2)’. After buildinga couple of ships, an enthusiastic12-year-old
could swap nautical waffle with the best.

When gluing 17th-century vessels such as the Royal Sovereign, the Wasa
and the Saint Louis, modellers could learn the rudiments of ship stability
and hull design as well. The original ships were built when scientific method
was just beginning to creep into the tradition-bound world of shipbuilding.
In the late 1960s and early 1970s, the kit engineers at Airfix were at the
cutting edge of research into historical nautical science.

It is easy to overlook the depth and scope of the research that went
into producing these kits. Unlike a Spitfire or a modern warship, there
were no originals (with the exception of the Swedish warship, the Wasa)
or comprehensive drawings to refer to; the Royal Navy, for example, did
not file drawings of ships until after 1706. Preparation of these kits required
considerable skill and historical insight on the part of Airfix’s toolmakers,
far in excess of what one might expect from the employees of a toy company.

In the case of the Royal Sovereign, launched as the Sovereign of the
Seas in 1637, Airfix’s main sources were two contemporary pictures: one,
a colour engraving by John Payne, was a good likeness of the ship’s port
side and overall rig; the other was the National Maritime Museum’s magnificent
painting (by Peter Lely) of the ship’s exuberantly ornate stern, a fantastic
challenge to the injection-moulding process.

The ship’s lantern dominated the stern of the Royal Sovereign. The lantern
was one of the wonders of the age: it was said that 10 men could stand inside
it without pressing against each other. In the foreground of the Lely painting
stands Peter Pett, who designed the ship with his father Phineas: he holds
a pair of dividers as a modern doctor might hold a stethoscope – they are
his scientific trademark. Shipbuilding was, after all, just beginning to
have pretensions to being a science.

Peter Pett had reasons for wanting to be regarded as a scientist. Though
he was a respected member of England’s leading shipbuilding family, his
father’s previous ship, the Prince Royal (another Airfix kit) had proved
a poorly built menace. Phineas Pett’s decision to double-bolt the bulkheads
with iron pins may have been as much an attempt to keep the hull in one
piece as a design innovation, while his odd decision not to sheathe the
hull below the water line with an extra layer of planks (then a standard
for warships) made the ship a popular residence for barnacles and shipworms.
In service, it was slow, uncomfortable and of dubious stability.

After this fiasco, it seems surprising that the Pett family was allowed
near a royal dockyard again, although it is not difficult to think of parallels
today. The jobbing commercial shipbuilder is still very much with us: the
Type 21 frigates, built to a commercial design that the Royal Navy bought
‘off the shelf’ in the early 1970s (and another Airfix kit), badly held
up the Falklands task force in 1982 with their need for regular topping
up with fuel, purely to keep them stable.

The Royal Sovereign, like the Prince Royal, cost a huge amount of money.
Decoration alone soaked up Pounds sterling 6691, 10 per cent of the total
cost. The vast quantity of gold leaf involved soon earned it the nickname
‘Golden Devil’ from the Dutch, its usual adversaries. Nor would this have
been the wispy gold leaf we are familiar with today: traces of the gilding
remaining on the wreck of the near-contemporary Swedish ship, the Wasa,
were found to be almost 0.09 millimetres thick, many times as thick as modern
gold leaf.

To produce the kit of the Royal Sovereign, engineers at Airfix made
a series of large wooden patterns from measurements and drawings taken from
the two pictures. This was no mean feat in view of the extreme intricacy
of the carvings. The patterns were then ‘die-sunk’and a pantograph reduced
the size of the parts to the desired scale of 1/168. This generated the
shape of the final injection mould into which plastic was then forced under
high pressure at 250 Degree C. The resulting model, once assembled, gave
a convincing image of the ship, although the small scale and the delicacy
of the plastic yards made it virtually impossible to fit all the running
rigging: rather impressionistic standing rigging had to suffice for all
but a handful of zealots.

To modern eyes, the model’s stern makes the ship seem top heavy – and
modern eyes are right. While there was no formal concept of hull stabilty
in 1637, practical experience with three decks of heavy cannon soon proved
Pett’s many critics right when they said: ‘The art or wit of man cannot
build a ship well-conditioned and fit for service with three tier of ordnance.’
Eventually, a whole tier of guns was removed, and an appropriate amount
of rock shovelled into the ship’s bowels to hold it reasonably upright.

The design of the Royal Sovereign was not a total failure, however:
its rounded stern, a departure from the square stern that characterised
English warships, marked a step forward in the design of hulls by trial
and error. In 1585, the shipwright Matthew Baker made a drawing of his ideal
hull, rather fancifully superimposing a cod’s body on the underwater section
to suggest the form desired, but he offered no mathematical basis for shipbuilding
practice, and failed to consider stability at all.

A century later, Anthony Deane’s Doctrine of Naval Architecture is awash
with equations and figures for building hulls, but still has nothing to
say about stability. Doctrine of Naval Architecture also turns out to be
an all-too-apt title for the work: Deane’s numbers turn out to be an arithmetical
petrification of the English version of the European empirical shipbuilding
tradition. As late as 1750, the Swedish naval architect Fredrik af Chapman
was solemnly taught that a ship’s bowsprit ‘must’ be 9/16ths of the length
of the foremast by English naval architects, who shrugged when he asked
why.

The Royal Sovereign’s hull has a classic 17th-century form, as codified
by Deane. Unlike modern hulls, which are based on roughly parabolic curves,
17th-century hulls were generated by a series of circular arcs, known as
‘sweeps’, each with a radius defined by little more than years of trial
and error tempered with more than a little conservatism:

radius of the first sweep, 1/4 of the beam;

radius of the second, 20/36 of the beam;

radius of the third, 7/9 of the first;

radius of the fourth and fifth, 17/36 of the beam.

This is Deane’s formula for the shape of a vessel at its broadest point:
the Royal Sovereign’s hull, built 50 years before Deane was writing, is
an excellent example of its application. The broadest section was not necessarily
in the centre of the vessel: experience, and comparisons with ducks and
fish, led to this being about two-fifths of the ship’s length from the bow.

The Petts could have done worse than seek Swedish advice on the stability
of hulls. In 1628, nine years before the launch of the Royal Sovereign,
Stockholm harbour had seen arguably the most spectacular demonstration ever
of elementary principles of stability when the Regalskepp (Royal Warship)
Wasa capsized after a maiden voyage of 1500 metres.

In 1961, historians raised the wreck of the Wasa. Its restoration is
almost complete: the ship was moved to a new purpose-built museum earlier
this year. A team of Airfix engineers visited the Wasa, with a view to preparing
a kit of the ship. The result, with one or two provisos, was a classic of
the plastic-moulder’s art.

The restoration of Wasa’s stern was still far from complete when the
Airfix team visited Stockholm, so there are some discrepancies between the
stern of the final mould and that of the ship as it now looks, but the model
otherwise reflects a 17th-century warship exactly.

It is when one turns to the stern of the model that two important features
of the ship show best. On the one hand, the ship bristles with intricate
gilded carvings. All the carvings are lovingly reflected by Airfix in 1/144
scale. On the other hand, the Wasa was a truly atrocious design from the
standpoint of stability. Though a view of the stern rather exaggerates the
position, it now seems almost comically obvious that the ship was seriously
top heavy if observers note the position of the water line in relation to
the height of the sterncastle.

The underlying reason why the ship sank so soon is apparent on the model,
though many of the actual causes of the catastrophe were still uncertain
when Airfix first produced the kit. The Wasa was originally scheduled as
a comparatively small ship by its Dutch designer, Henrik Hybertsson. However,
under pressure from the Swedish king, Gustavus Adolphus, it was enlarged
on the stocks. Hybertsson’s original design was for a single gun-deck, but,
again under pressure from the king, the completed ship had two tiers of
guns with more guns on the weather deck.

To bear the weight of all these guns, and the recoil stresses they created,
the hull was built for strength rather than stability, with huge longitudinal
wooden girders and vast amounts of heavy strengthening between the frames.
All of this took up so much space that there was little room left for the
huge amount of ballast needed to make it stable. Like most Dutch-built ships
of this period, its bottom was all but flat, so there was little space for
ballast to begin with; tinkering with the ship’s size and armoury turned
it into a death trap.

As the ship crossed Stockholm harbour on 10 August 1628, it encountered
a squall. The ship heeled over, then, very stiffly, righted itself. It was
no sooner upright than it began to go over again: this time the heel turned
into a capsize, and the ship’s career (and many lives) came to a premature
end.

Fittingly, it was largely in Sweden, a century later, through the work
of Frederik af Chapman, that mathematics and formal physics found a role
in determining the stability of a ship. Using his methods and concepts,
we can reproduce the failings of the unfortunate Wasa .

In the case of the Wasa, the centre of gravity was 6.16 metres above
the keel; and the metacentric height, the distance between the centre of
gravity and the metacentre, was 14 centimetres at an angle of list of 10
degrees. Simplifying drastically and allowing for the squall, this spells
doom for the Wasa. A more usual metacentric height for the period would
have been anything between 50 centimetres and 1 metre at a 10-degree list:
the ship never stood a chance. Hybertsson was spared the embarrassment of
seeing his masterpiece keel over: he died a year before the ship was completed.
As so many of the alterations had been made at the king’s insistence, nobody
was held to blame: the ship’s loss was put down to act of God. A more robust
approach might have advanced naval architecture by a century.

Some naval historians think the Wasa was based in design on a slightly
earlier vessel, the Saint Louis, built by Dutch shipwrights for Cardinal
Richelieu’s emergent French navy in 1626. Following its success with the
Wasa kit, Airfix decided to produce a model of the Saint Louis, but soon
confronted a problem that required fairly serious speculation in the field
of maritime history. The only image of the ship was a contemporary broadside-view
engraving by Henrik Hondrius. While this gave the company’s engineers a
good starting point, they had to theorise the details of the ship backwards
from the Wasa, a process requiring considerable historical skill. The result,
logically enough, was in some ways a smaller version of the Wasa. This gives
the modeller the opportunity to compare the unfortunate Swedish ship with
something very like the form it would have taken if Gustavus Adolphus had
not insisted on so many guns. The difference between the displacements of
the two ships is only about 200 tons (1400 tons as opposed to 1200 tons),
but the fact that the French ship was never forced to carry the 64-gun ensemble
that caused the Wasa so many problems left it comparatively free of hydrostatic
horrors; while the hulls are very similar, the centre of gravity of the
Saint Louis would have been comfortably low. The Saint Louis was also free
of the over-elaborate two-storey galleries in the stern that were (and are)
the Swedish ship’s glory, but also part of its nemesis: the similarity in
the respective designs of the turreted galleries is, however, unmistakable.

The Saint Louis not only emerged from its home port in one piece but
had a long and successful career, showing that 17th-century Dutch-built
warships were capable of competent flotation when undisturbed by royal whims.

If the engineers at Airfix faced an intriguing problem in historical
naval architecture when deducing the Wasa’s structure, they were faced with
a far thornier one when asked to produce a kit of the Pilgrim Fathers’ vessel,
the Mayflower. There was no visual evidence at all – no plans, no drawings,
no pictures, not even a reliable verbal description. A Mayflower kit, however,
would sell well in the US, so commercial pressures and skill in historical
naval architecture produced a model that remained in the Airfix catalogue
for 12 years.

To perfect a miniature version of the Mayflower, the Airfix designers
consulted Matthew Baker’s 1585 manuscript in the library of Magdalene College,
Cambridge, from which they learnt the probable form of the hull. The actual
dimensions of the model were taken from a table of proportions laid down
by William Borough, Comptroller of the Navy from 1589 to 1598. The Airfix
model is arguably a more learned guess at the form of the original ship
than the model in the Science Museum, built in 1926, and hitherto thought
the best representation.

But this unlikely scholarly tradition at Airfix could not continue.
The model kit industry was under a heavy economic cloud throughout the 1970s:
Airfix ceased trading in January 1981, finally giving way under a ban by
the European Community on ‘warlike’ illustrations on kit boxes. This resulted
in a compulsory redesign, costing Pounds sterling 300 000 on Airfix packaging.
An American company, Palitoy, bought Airfix and soon Humbrol, a British
company, became the second owner. Neither attempted to continue the proud
Airfix tradition of fine ship kits.

The last Airfix Classic Ship kit was a model of Captain Bligh’s HMS
Bounty, issued in 1979. We are unlikely to see more. While some of the much-loved
old kits of the 1960s and 1970s have been re-issued, the moulds are beginning
to show signs of age, and often require the modeller to hack away the wall
of ‘flash’, a thick plastic membrane resulting from leakage in the mould.
Some kits are now collectors’ items, sold at collectors’ prices. Anyone
with the kit of the liner, the Southern Cross, should consider keeping it
in a bank.

But the big, rattling box containing an Airfix ship kit will always
hold a special place in the hearts of a few fanatics for whom the words
‘locate and cement’ bring a lump to the throat. No Christmas afternoon was
quite complete without the mildly hallucinogenic stink of polystyrene cement
and the frantic scramble for tiny parts eaten by the carpet. Unlike so many
children’s Christmas presents today, you didn’t have to program an Airfix
kit: it programmed you.

* * *

A SHIP DESIGNED TO TOPPLE IN THE GENTLEST BREEZE

WHEN a ship is at rest, the upward force of its buoyancy acts through
the vessel’s centre of buoyancy (B1). This point is always at the centroid,
or geometric centre, of the immersed part of the hull.

The consequence is that when a ship heels, the centre of buoyancy moves
to a new position (B2). The downward force of a ship’s weight acts through
the vessel’s centre of gravity (G), whose position never changes.

When a ship is vertical, the opposing lines of action of these two forces
coincide and the vessel settles to a depth in the water so that their magnitudes
are equal as well. When the ship tilts through an angle (u) from the vertical,
the lines of action no longer coincide and a moment, or turning force, is
set up.

As long as the upward force of buoyancy is nearer to the sinking side
of the vessel than the downward force of weight, this moment tends to right
the vessel. The magnitude of the moment depends on the length of the perpendicular
from the vessel’s centre of gravity to the vertical line throughB2 (GX).
As u increases, GX becomes shorter; when GX is zero, the vessel is about
to capsize.

Any further increase in u pushes the upward force of buoyancy farther
from the sinking side of the vessel than the downward force of weight, to
B3, and then both forces help the vessel to keel over.

Broadly speaking, the stability of a vesseldepends on the relative positions
of the centre of gravity and the metacentre (M). The metacentre is the point
of intersection of the vertical line through B2 and the vertical line through
B1 when the vessel is at rest. The greater the ‘metacentric height’, GM,
the more stable a ship; and u can be fairly large before GX becomes zero
and the ship capsizes.

In the case of Sweden’s ill-fated warship,the Wasa, GM was 14 centimetres
and GX was 1 centimetre when u equalled 10 degrees. All in all, an almost
perfect design for disaster.

William Scanlan Murphy is a freelance writer, naval historian, broadcaster
and musician. His book Father of the Submarine, a biography of George Garrett,
was published in 1987 by William Kimber.