Dear Friends
You will have seen that
@RDN1954 Johan was asking about how the size of the Willem Barentsz was determined. Like everything associated with this ship, it is a fascinating story and one really worth telling. So without further ado let's get on with it.
DETERMINING THE SIZE OF THE WILLEM BARENTSZ - Het Schip van Willem Barents - Ab Hoving
The size of ships from approximately circa 1600 was determined by load capacity – tonnage (lasten) - and not via length in feet. This presented a problem as tonnage referred not only to weight (400 Amsterdam pounds = 197.5 kilogram), but also to volume. Compounding the issue was also how the tonnage was determined - measuring the load capacities of wood and grain (just to cite two examples) would result in different tonnages. Also, tonnage did not refer to actual water displacement volume, but only to load capacity. Load capacity, in turn, was defined as the available cargo space after the ship had been fully rigged and equipped with all amenities required for its operation. The more food, water and armament that was taken onboard, the smaller the loading capacity.
A way of measuring the tonnage was to multiply the length, width and depth of the hold (in feet) – the sum of which was then divided by a pre-determined factor based on the type of ship – 250 for men-o-war and VOC ships, 242 for wood-carriers and 242 for the flute ships (Fluyten). The highest factor (400) was allocated to war jachts “oorlogsjachten” of which the Willem Barentsz was a prime example*. This makes sense, because once a (small) jacht was equipped with all armament and military-related equipment, there wouldn’t have been much space left for carrying an additional load.
The classification of the Willem Barentsz as an "oorlogsjacht" brings with it, its own story.
From De Veer’s journal we know that the two ships that were equipped for the WB expedition were of 30 and 50 last respectively. A natural assumption would be to assume that the most prominent men (Van Heemskerck and Barentsz) sailed on the bigger ship. However, the expedition of Abel Tasman proved this was not necessarily so. On that expedition, the Fluyt
Zeehaen measured 100 last while the “oorlogsjacht”
Heemskerk weighed in at 60 – yet the
Heemskerk was six feet longer than the
Zeehaen! The reason? The Fluyt’s small crew and very limited armament accounted for a high lasten number because it was much more “economically” equipped and could therefore carry a bigger load.
Analogous to this, we know that during the Barentsz expedition, the bigger ship (higher lasten) was commanded by Rijp while Barentsz traveled on the ship which would have been the better sailing vessel and also the one with the most prestige – the smaller oorlogjsjacht captained by Van Heemskerck. So even though Barentsz’s ship at 30 lasten may have APPEARED the smaller of the two, it may well have been longer!
What would have been great was if we could have had a table which equates the lasten to the length of a ship, but unfortunately such a table does not exist. According to Ab Hoving, even if such a table did exist there would have been too many variables (between ship types, the year of construction and the region where the shipyard was situated) for such a list to have been of any use. So, in the final analysis, the fact that we know that Barentsz traveled on the 30 lasten ship is a “nice-to-know”, but does not provide us with any indications to the ships’s size or length.
So, what about the size of the crew – would that serve as an indication to the size of the ship? What about a list that equates to the number of men to the lasten or size of a ship? In fact, there were two such lists “Manninge en Monture” by Van Dam and “De Evenredige Toerusting” of the Admiralty. The former allocated one man per 1.5 last and the latter one, 2 lasten per man. But as far as ship’s size goes, this also serves as no indication.
What about the number of cannons onboard? According to Hoving (he bases his theory on the drawings of De Veer) the WB had SEVEN gunports on each side. (This is interesting as both De Weerdt and Hans’s research shows six). When we analyze the Zeven Provincien, we see that 7 cannons require a ship’s length of 80 feet. Bear in mind though that these were heavy artillery of 24-and 36 pounder cannons, while on the Barentsz the cannons would certainly have been of much smaller caliber. Needless to say that the smaller armament would also result in different measurements. Taking Witsen’s pinas as an example, we see that for seven 12-pounders, a ship’s length of 75 feet was needed. It was almost certain that the WB had even smaller armament so that is not of much help either. The best that we can get from this, is the WB would have been smaller than 75 feet but that is not much.
Next up, what about the number of deadeyes and shrouds? According to Van Yk, a ship of 100 feet in length was equipped with 6 shrouds for the main mast. Whilst this is a great way of estimating the size of a ship on a painting or drawing, it is unknown whether the same parameters of 15 or 16 feet between each deadeye also held true for ships of less than 100 feet. And even if we could use that rule, the drawings of De Veers show different numbers of shrouds on different drawings (one picture shows 5, another shows 7.)
Another method was counting the number of “banen” in the main mast and foremast sails. We know that the distance between banen was approximately 75mm but here as well the pictures show too many variances. Ratlines at 40cm apart may also have offered a clue, but again there is no consistency in the drawings.
With all the above of little value, what about the size of different ship’s components? Do they have any relevance to the overall length of a ship? Van Yk tells us that for each 10 feet of ship’s length the keel at the stem would be one-inch thick. From that measurement, various others are derived: the hull planking would be a quarter of that thickness, nine-eights of that for the deck beams and half of that for the wales.
From a number of shipwrecks or parts thereof discovered and recovered, a large number of measurements are available of various ship’s components. These would offer us the best opportunity so far of determining a ship’s length except for two provisos: One, it does not allow for any deviation or personal interpretation of Van Yk’s rule by the individual shipwrights - something we know the Dutch shipbuilders were renowned for. Two, we need to have recovered artefacts from the ship in question.
As far as the latter goes, the researchers of the WB struck gold when a large piece of the stem and associated planking of the ship was discovered not far from the geographical site where Het Behouden Huys stood by Russian archeologists and which now resides in a museum in Moscow.
This great BUT There is a caveat! Examining the hull planking at the stem of the ship, it was discovered that it was DOUBLE-PLANKED!!! Which immediately raised the question – how should the two 4cm-thick hull planking at the stem be calculated. Viewed as a single layer it gives us a length of 40 feet and, combined, a length of between 80 and 100 feet.
Next up the hull frames (spanten). According to Van Yk, a one-inch thickness was used for each three feet of ship’s length. The 12 and 13cm (5-inch) thick frames recovered, would suggest a length of 60 feet. Nicoleas Witsen adopts another formula: he measures the thickness of the hull-planking (4cm) with a factor of 436 to result in a ship’s length of 17 44 m or 61.5 feet. Seeing that there is only a one-and-a-half-foot variance between van Yk’s and Witsens’s applied formulae, it would seem that on the surface, this is a case of game, set and match, but hold onto your horses. Above the wales the planking is only a single layer which again questions the validity of the other “conclusions”.
At this point you may be well wondering: How is it possible that there can be absolutely no record/specifications of a ship built in more or less the same era as the Willem Barentsz? Well – there is one – an 85.5-feet pinas built according to the Zeeuws bestek (list of specifications) relating to the pinas.
Converting all dimensions, measurements and information of the pinas into a drawing resulted in a layout which looks remarkably like the 1671 pinas that Witsen describes. However, it bears little no resemblance to the Willem Barentsz. The internal layout and deck configurations of the Willem Barentsz, in particular and as described by De Veer, are vastly different to that of the 85.5 feet pinas.
The pinas featured two full-length decks of which the lower at the stern makes a pronounced step to compensate for the natural zeeg (curvature or sway) of the hull. The upper deck houses the forecastle and a cabin at the rear with a hut on top of that. In contrast the WB featured a single deck which follows the zeeg of the hull all the way from through stem to stern. On top of that deck are the front and midship canopies with the helmsman’s hut on top of the latter.
Thus: according to the specifications of the pinas, the last 18 feet of the deck was lowered and flat because if the deck had followed the zeeg of the hull, it would have negatively influenced the position of the gun ports. That means that we can subtract 18 feet from the length of the pinas to arrive at a gun deck that does not need to be stepped in order to fulfil its purpose or to compensate for gunport positions.,
Which then gives us our first real indication of the size of the Willem Barentsz:
Provisional Specifications:
Length: 67.5 feet
Width: 21 feet
Depth of the Hold: 9 feet
Now, one question remained. How seaworthy would a vessel of that size be?
The sketch of the pinas was transferred onto a line plan with a realistic distance between the center of gravity and the metacenter. The center of gravity is regarded as the point at which the total weight of the ship’s structure and everything on board culminates while the metacenter (calculated from the hull frames) is the point of intersection between a vertical line through the center of buoyancy of a floating body such as a ship and a vertical line through the new center of buoyancy when the body is tilted. Obviously the metacenter needs to be above the center of gravity to ensure stability
The distance between these two is crucial. If it is too small the ship would capsize easily; too big and the ship would be so stable that it would be impossible to steer or refuse to react to inputs. The crew would call the latter a “wrede” (cruel) ship, The ideal distance between the center of gravity and the metacenter is regarded as three feet for ships in the 80 feet length category or somewhat smaller.
With this information applied to our provisional dimensions of 67.5 x 21 x 9 feet., only a single alteration needed to be made (Van Harpen):
The width of the ship had to be increased by half a foot – thus to 21.5 feet. With these parameters applied, the ship would have had a water displacement of 120 tons. If we take into account that it was classified as 30 lasten (60 tons), it would leave 60 tons for the actual ship plus its essential equipment. This all gels perfectly with everything we know about the load capacities evident in Dutch shipbuilding of that era.
With the final measurements pinned at 67.5 x 29.5 x 9 feet how would the WB have fared as a sailing vessel. The answer: REMARKABLY WELL, Thank You! Under full sail she would have been fully manageable in winds up to 5 on the Beaufort scale. Beyond that, it would have called for a reduction in sail area. And what about speed? At a wind speed of 10 meter per second, she would have been capable of 8 knots which compares very favorably to the 4-6 knots of the big Spiegelretourschepen of 50 years later.
And there you have it Johan - that is how we know what the size of the Willem Barentsz was. In a later posting, I will apply this information to the drawings of Hoving, Kolderstok and De Weerdt and we can see how they measure up.
I hope you guys enjoy it.
The honest answer is I have no idea. The fact that the position of the windlass has been a bone of contention for so long, makes me think, probably not. If it had been portable, it would certainly have solved all issues.
