De 13 Søskende

Plans

While living in Odense, Denmark for 10 months, I was introduced to the book Wooden Boat Designs², which contains plans and descriptions of 41 traditional Danish boat types. At first, I had a photocopy of a photocopy of the original Danish edition. Wanting to go legitimate, I bought a hard copy of the same. I selected this vessel to model, but did not begin immediately, as other projects were ongoing. I then found the book translated to English and bought it. Still the model had not been started. When it became available some years later, I purchased a CD contain high-resolution scans of all the drawings, and still the naissance of the model was in the distant future. By the time I did begin the model, all of the drawings could be downloaded for free from the Maritime Museum of Denmark! I’m glad to have one copy of the book in addition to the scans, but I wasted money on the the original book and the CD. These links are to the three sheets for De 13 Søskende:

• Lines plan
• Sail plan
• Construction drawings

All of the drawings in Wooden Boat Designs were created by Christian Nielsen. Realizing how traditional local craft were disappearing due to the introduction of the motor and the specialized vessels that followed, the Maritime Museum of Denmark wished to obtain a record of the vanishing types. For fifteen years Nielsen visited different parts of Denmark during the summer months, traveling by train and bicycle. In the winter months, or when not busy in his small boatyard on the island of Fejø, he worked up the material he had gathered during the summer. The effort was all financed privately. He also managed to build 15 models between 1939 and 1953, including one of De 13 Søskende.

Mr. Nielsen was the grandson and namesake of the man that built De 13 Søskende. It is coincidence that, of the 40+ drawings within this book, I chose to model the one to which the author had a familial connection. I chose it for its aethetics, the somewhat unusal aft spar, and the very unusual purpose for which it was built (eel fishing).

Other Models

The drawings that Christian Nielsen (the younger) produced were not his first association with the museum. After finishing his apprenticeship, he took a job at Helsingør shipyard and continued his studies through a private tutor. He was introduced to the Maritime Museum of Denmark through this tutor, and the initial interest was for his skill as a model builder. The museum’s first request was for a model of De 13 Søskende. Here are three pictures of the model from the museum. The two black and white images are also available on their website.

Model of De 13 Søskende, built by Christian Nielsen, Fejø.
Model in the custody of the Maritime Museum of Denmark.
Photos: Maritime Museum of Denmark

One interesting piece of information not found elsewhere is the vessels registration number, which is seen on the sail and the bow.

Historic Photographs

The image below is of De 13 Søskende later in life, after the fitting of an engine and a change in owners and hull color. This information comes (via fejoe.info) from Mie Lobedanz, great-granddaughter of the original owner . It is not a direct translation of the original source. Some details repeated elsewhere are omitted and the order rearranged to suit.

Extant Vessels

I was hosted by Morten Gøthche and Johannes Ambjørnsen at the Viking Ship Museum. Johannes showed me the two original vessels Viktoria and Dan. Viktoria belongs to the museum, and Morten owns Dan. Both were built by the same shipwright as De 13 Søskende.

Author’s photos

Known extent vessels:

Name	Home Port	Builder	Year	Length	Beam	Draught	Owner
Viktoria	Roskilde	Christian Nielsen	1904	26.5	8.6	3.3	Viking Ship Museum
Dan	Askø	Christian Nielsen	1908	26.8	9.3	3.3	Morten Gøthche, Roskilde
Mæfikken	Dybvig Haven	Kristian Mortensen	1917	29	10	5	Fejø Drivkvaselaug, Fejø, Finn Sloth Larsen
Karen	Bogø	Anders Andersen	1921	22	8.4	3.4	Bogø Kvase Lag, Bogø
Ellen	Kalundborg	Kristian Mortensen	1922	26	10	3	Henning Pedersen, Kalundborg
Concordia	Udby Vig	Kristian Mortensen	1927	28	10	3	Nis Olsen, Holbæk

Viktoria

Photos: Viking Ship Museum, Roskilde, Denmark

Karen

Photos: Bogø Kvase Lag / Karen af Bogø⁶

Materials

Fasteners

Dimensions

The nails used on a replica built at the Viking Ship Museum are 65mm long with a square shank 4 mm on each side. The head diameter is unknown. I scaled the image from the website and arrived at ~10 mm.

Photos: Viking Ship Museum, Roskilde, Denmark

Jamestown Distributors sells copper boat nails and provides a table of nail gage to rove diameter and nail head diameter. The ratio of rove diameter to nail head diameter is consistent.

Rove OD	Head Dia	Ratio
7.9375	4.191	1.894
9.525	5.08	1.875
11.1125	5.969	1.862
	Average	1.877

The diameter of the roves used for the replica was 19 mm. One might therefore presume that the nail head corresponding to a 19 mm rove would be 19 / 1.877 = 10.123 ≈  10 mm, which is consistent with the dimension scaled from the photograph.

The Crown Nail Company offers Danish Boat Spike Nails made of steel. These have a rectangular shank, the closest size to the copper nails used in the replica being 4mm x 5mm with a 10mm head, also consistent with the scaled image.

The web page describing the construction of the replica eel drifter states that iron nails would have been used at the time De 13 Søskende was constructed. It says that copper was used for leisure craft, but it was not until the 1950’s that traditional boatbuilders also began to build with copper nails. In the pictures I took of Viktoria, built in 1904, one can clearly see that the fasteners are copper. However, it is quite likely that the fasteners were all replaced as part of a restoration.

Despite iron having a yield strength 1.5 that of copper, I’ll be presuming the original iron and modern copper fasteners are of approximately the same dimensions.

Hull Planking

Butt Seams

The lower strakes were made of 2 planks, and the upper strakes of 3 planks^{1, p16}. Of the 10 strakes, how many are considered upper and lower is not clear. I choose based on:

• total length (There is a limit to the length of timber that could be obtained)
• variation in width (more of a board is discarded as this increases)
• required board width to accommodate curvature (wider boards are more expensive and much of the board would be wasted)

The butt joints would traditionally be placed at the frames^{1, pp16,54}, though joining planks between the frames with the use of butt blocks provides more freedom when chosing lumber and makes replacing planks easier. Without seams at the frames, there is less chance of water intrusion and correspondingly less concern over rot in these larger timbers, which would be even more costly and difficult to replace than the planks. Viktoria uses butt blocks, a detail my host was keen to point out (see image below).

Butt blocks were also used in the replica Tumleren^{1, p54}. In his letter to the builder, Nielsen^3,p147, Ole J. Strandby requests that his vessel be built with seams at the frames. While the builders of Tumleren were worried about rot at the frames, Mr. Standby was more concerned with rot at the butt blocks. That he bothered to make this specific request could imply that Nielsen was known to use both techniques. I find them to be an interesting detail, and chose to use them for De 13 Søskende.

After installing 3 strakes on each side using the butt block technique, I recieved additional photographs of the model built by Nielsen’s grandson. No butt blocks are visible. It is entirely possible that this detail was simply omitted from the model, and I am not taking it as evidence that butt blocks were not used on De 13 Søskende. Even if originally constructed with seams at the frames, Viktoria and/or De 13 Søskende could have been re-planked during their lifetimes with the use of butt-blocks, so I am not overly worried about the accuracy of including them.

Edge Seams

The planks overlapped by approximately 1-3/8″ and fastened with clinch nails spaced approximately 5-1/2″ apart.^3,p110

Wet Well

A series of holes was drilled in the hull planking where it forms a boundary of the wet well^{1, p18}. The note says the holes were small enough to keep eels from escaping, but it does not provide a diameter nor a number or pattern. Nielsen’s model did not shed any light on this aspect, as they seem to be omitted entirely.

Waterline

The waterline was scribed so that the correct trim could be determined even if the paint were worn away^1,p22. I will scribe a very light line, which may or may not be visible after painting.

Pancakes

In the time period in which these boats were built, they would celebrate completion of the hull planking with traditional yeast pancakes (‘raised pancakes’)^{1, p16; 3,p110}. I will do the same!

Deck

Planks

Gøthche states that 7″ planks were normal, but to combat leaks due to shrinkage, 5″ planks were sometimes used^{1, p20}. The deck planks drawn for De 13 Søskende show some variation, but are all close to 5″. I’m using a consistent width of 5″ unless it leaves too narrow a strip at the outside. In that case I will widen the center plank(s).

Toe Rail

Gøthche states the toe rails are 3.5″ high before adding the rail cap^1,p20. I measure this at only 3″ on the drawings for De 13 Søskende. As he did not seem to imply any variation, I’m inclined to go with the larger dimension, which will allow more room for cutting out the scuppers.

He also points out the existence of a fairlead for the anchor chain cut out of the toe rail on the port bow. This does not show up in the drawings for De 13 Søskende, but I do have a photograph from Viktoria, which shows it reinforced as well.

Rub Rail

Gøthche states this as 2″ thick, though he may have meant deep^1,p20. However, this is called out on the drawings for De 13 Søskende as only 3/4″x7/8″. I’ll see which looks more appropriate when I get to that stage in the model. Note the staggering of the fasteners each side of the midline in the image above.

Coamings

Gøthche states this as 2″x2″, rounded over on the top^1,p20 for both the large opening and the cockpit at the tiller. I took a photo to show how the tight radius at the corners was handled.

Bitts

There were two common forms of mooring bitts. Dan and Viktoria have two vertical posts attached to a deck beam and extending well below deck. De 13 Søskende has two large wooden cleats. In both cases there was a cross piece between the bitts and cleats, to which the heel of the jib boom was secured.^{1, p18}.

Author’s photo (left). Drawing (right), Maritime Museum of Denmark.

This and the cabin roof were offset several inches to port of the centerline in order to provide more room to work the starboard side, where the net was always set^1,p20.

Cabin Roof

Gøtche points out that the cabin roof was laid directly upon the deck and fastened from underneath^1,p20. Evidence of this can be seen in a photograph taken from within Viktoria’s cabin, where the nails on the underside of the deck planks are clearly visible. This detail would only be visible when looking up from the cabin floor. I include it nonetheless, as it is mentioned in Gøtche’s book and I choose to document the vessel as thoroughly as possible.

1. Side of cabin roof
2. Deck plank
3. Deck half-beam
4. Wall of sleeping cabin or equipment locker

Keel, Frames, and Beams

Keel

A detail pointed out by Gøthche that is not shown in the plans for De 13 Søskende is the hourglass-shaped reinforcement between the keel and sternpost^1,p22. Being an inexperienced modeler, I did not know this was a common practice within and outside of Denmark, and would have missed including this feature. But after having seen it mentioned, I recalled the horseshoe-shaped reinforcement at the stem of Alert in the Anatomy of the Ship series. I looked at the back side of the dust jacket to see if there was a reinforcement at the stern. Sure enough there was, hourglass-shaped as well.

Deck Beams

Gøthche states that the deck beams forming the fore and aft ends of the deck opening were 3″x3″ with about 4″ of camber. He goes on to mention a single knee and series of half-beams supporting the side deck, but without mentioning dimensions^1,p20. The drawings for De 13 Søskende do not show any variation between the deck beams, knee, or half-beams, which appear rectangular, approximately 2.5″ wide and 2″ deep. I will opt to go with the slightly larger dimensions given by Gøthche for the deck beams.

Forecastle

If I choose to open up the cabin for viewing, the book by Gøthche offers some details that could be included^1,p18:

• A square opening on the forward bulkhead of the main space, leading to a smaller space where the boys bunk would be fixed athwartship.
• A removeable hatch on the next bulkhead forward, behind which was stowage space.
• Decorative edging, as photographed on Viktoria
• Bunk on the starboard side, two lockers for clothes and provision on the port
• Space to hang clothes aft of the lockers on the port side
• Small bench on aftermost bulkhead for cooking equipment and other utensils, at the same height as the wet well deck.
• Small benches in front of the lockers and bunk
• Folding table on the foremost bulkhead, with fiddles to prevent crockery from sliding off the edge.

It also points out that on Dan the sailor’s bunk is enclosed, whereas it is open on De 13 Søskende.

Rigging

Shrouds

Mainmast supported by two sets of shrouds of twisted wire, attached with shroud seizings. The mizzen was unstayed^1,p28.

Forestay

The forestay was passed through a hole in the stem and seized to the bitts. Alternatively it could be shackled to an eye fitting on the stem^1,p28.

Dan and Viktoria both have the forestay led through the stem, as does the model of De 13 Søskende made by Nielsen.

Mainsail Gaff

The mainsail gaff was equipped with a patent parrel, a rotating, bow-shaped fitting that could slide up and down the mast. This fitting made it possible to lower the peak of the gaff all the way to vertical, so that the gaff hung down along the mast, allowing the sail to be furled to the mast^1,p28.

Sails

Bands from an original sail from Dan were 53cm, and another sail from another vessel an equivalent width^1,p30.

Drifters usually had four reef bands on the mainsail and two on the foresail [staysail]. Only the three lower reef bands were equipped with reef points. The uppermost band was left with reef eyelets only, for loose reef points to be inserted when necessary. The mainsheet was attached to the clew becket with a hook so that it could be easily moved to a grommet on the leech of the sail, inline with the reef band^1,p30.

The sails were treated with a mixture of horse fat, ochre, wood tar and water, or with catechu (an extract of acacia bark imported from the far east). The former would take on a mottled, golden color, as in the right-hand image below. Catechu would become rusty red or brown. The majority of the images I’ve seen (like those of Elisabeths Minde and Haabet shown ealier) are rather dark, implying catechu was more common. The sails shown in the one known photo of De 13 Søskende are rather light, but this photo is later in life after a change in owners, so it is not definitive in selecting a color for the model’s sails. The model by Christian Nielsen is more representative of the time at which it will be modeled, and he chose tan colored sails. This is also not definitive, as he may have simply generalized based on common practice at the time. I would probably be justified in choosing either style, or even a combination. As sails wore out or became damage, a fisherman might purchase whatever was readily available, or perhaps a used item to save money. Refer again to the photo of Haabet, which has dark sails, apart from a light-colored staysail. Choosing such a combination might add more interest to the model and a greater sense of realism – in life, few things are perfect.

Finishes

Below the Waterline

Under the waterline the hull was treated with coal tar^{1, p22}.

Freeboard

The freeboard of most drifters was grey, but some (including De 13 Søskende) were painted Schweinfurt Green^1,p22, a.k.a Paris Green, Emerald Green, Vienna Green. This appears to be confirmed by Christian Nielsen’s model.

I did not know about this pigment or how widespread was its use. Unfortunately, it is highly toxic. More on that topic can be found here.

As for the actual hue of Schweifurt Green, the Wikipedia article states “The color of Paris green is said to range from a pale blue green when very finely ground, to a deeper green when coarsely ground.” However, I presume the coarser grind would be more economical and thus more likely to be used for a fishing vessel. I have two shades of Ronan Japan colors, one of which is named ‘Emerald Green’. It is a close match to the old paint can and jar. I also have ‘CP Green Light’, which also looks to be the proper hue, but a shade too dark. Perhaps a mix of the two will yield a good result.

Photo credits. clockwise from upper-left: 1) Wikipedia,  2) Straus Center for Conservation and Technical Studies, Harvard University 3) Utrecht Art Supply 4) Ronan Paints

Rub Rail, Cabin Roof

Normally green^1,p22, but vessels from Askø^3,p017 were painted brown, as on the extant vessels Dan and Viktoria^1,p22. De 13 Søskende was also from Askø, but on Christian Nielsen’s model the cabin roof is green, like the freeboard. I have to decide if I think it more likely Nielsen’s model is accurate or De 13 Søskende actually stuck with local tradition. Since the green freeboard is also a break from the norm, I presume the cabin roof is also, and will paint it green on the model.

Rail Cap

Varnished^1,p22.

Deck

Varnished, but the varnish was often mixed with a little ‘English Red‘,  a pigment derived from iron oxide^1,p22. The RGB color as specified by Wikipedia appears to be a good match for the deck of Dan. Viktoria‘s deck lacks the brownish hue from the iron oxide. Though worn, Dan‘s deck appears rather opaque for a mixture of varnish and paint. I wonder if it is just paint. I may use a wash so that it looks more like a high-solids stain, as seen on many home decks.

Author’s photos.

Launching

In addition to the pancakes served at the completion of hull fastening, another tradition I aim to honor is the post-launching feast, at which coffee, tea, æbleskiver and Fejø-punch would be served^1,p24. My wife and I snacked on æbleskiver as we walked around Den Gamle By (the old town) at Christmas time. I searched in vain for the recipe for Fejø-punch until I noticed it in the margin of the book.

Fejø-punch

• 2 liters^a of rum
• 1-1/2 bottles of red wine
• 2 pounds sugar
• Cinnamon sticks
• Equal quantities of water and rum^b

a) The recipe actually calls for 2 pots of rum, a pot being 0.966 liters. I figured, since my model will be 3% too small, why not over compensate by adding 3.4% too much rum 😉
b) This one confuses me; equal parts water and the mixture above?

Fishing Rig

Here are two images depicting the seine net and its use.

Gøthche gives details about how the boat is configured when drifting. The far right of the image shows the boat drifting directly downwind. He states that when retrieving the net, the boat is pointed into the wind. I assume a few points off, as in the upper-left of this image, so that the boat is not pointed directly into the mouth of the seine. The position shown at the bottom-left depicts the boat drifting slightly astern, which could be achieved by pulling the aft seine line in slightly^1,p33.

My intention is to provide a visual indication of the vessel’s historic use and how it functioned by modeling the boat in the process of fishing. The seine is quite large, and when fully deployed it is quite a distance from the vessel, as shown in the diagram. I’ve seen models like this, but only in museums where space is not an issue. To keep the display size reasonable for my home, I can depict it as the seine is being set or taken in. I considered having the net still on the deck, but seriously doubt my ability to model a natural-looking pile of netting. I also doubt that I could well-depict the seized bag full of eels being hauled onto the deck. Doing so would also prevent me from showing the sails set close-hauled, as described below. I think having the sails set in an unusual fashion would make the model more interesting. It seems the best bet, for both size and interest, is to show only the arms of the net, in the water a short distance from the boat, as if it the net has just been let out. The next image is a concept of the display. The rectangle is the display case. The arms of the seine net are at the bottom and the viewer must imagine the rest of the net. I would show the sea floor and a number of eels among grass to convey the quary of the fisherman. A few plaice and a sunken log could be added for interest. I’ve never modeled water before. If trials are successful, the hull would rest within a hole cut in a plexiglass sheet with the water’s surface modeled with acrylics. The wake of the boat moving sideways would further convey the method of drift seining. We shall see if this comes to fruition.

The boat would be sailed “hove-to”, with the mainsail set as usual but the [staysail] set on the windward side, with its clew tied to the foremast shrouds [E]. The jib and mizzen would also be set as well, to balance the boat during its sideward drift.^1,p32

The set for the jib and mizzen are not specified as it is for the main and staysail. It appears it would be difficult to bring the mizzen clew to windward, and it seems the boat would be unbalanced if neither that nor the jib is set to windward, with only the staysail aback. So, I presumed the jib clew is also brought to the windward side. This was fortunately confirmed later when I received more images of Mr. Nielsen’s model. He shows the staysail sheet secured to the shroud, as it would only be when hove to. The staysail is set to the same side.

Left: Staysail secured to starboard shroud.
Middle and right: Jib sheet secured on the same side as the staysail.
Original photos: Maritime Museum of Denmark.

The magenta arrows below indicate how the forces on the boat were balanced – the staysail and jib, being brought to windward, would push the boat aft. The main and mizzen would push it forward.

Seine nets were stored on the foredeck to starboard, just forward of the shroud [A]. They would be set from here, paid out from the jib boom forward and the drift boom aft, while the sails were set. …^1,p32

Does that mean “while setting the sails” or “with the sails already set”? I presume the net was payed out gradually, to ensure nothing gets tangled. To pay it out gradually, the boat must be making noticeable leeway, if not moving directly sideways. I’m not an experienced sailor, so it’s not clear to me if the staysail would have to be to windward for this, or if the jib to windward would be enough, particularly if the mainsheet is eased to take some power out of that sail. The foot of the jib is rather high, and perhaps would not be in the way at all. It appears the staysail is cut just high enough to work the net while the sail is to windward. It will be easier to envision when the model reaches that stage.

The end of one of the seine lines was fashioned into an eye that could be slipped over the tip of the jib boom [B]. The other line was passed through a hole in the outer end of the drift boom aft [C] and then led to the forward end of the boom, where it was belayed on a cleat [D]^1,p32.

Above it was stated that the net was “paid out from the jib boom forward and the drift boom aft.” But if the forward seine line is terminated with an eye, set over the end of the jib boom, it cannot be paid out from there. I presume it is paid out from the foredeck until all the slack is taken out, at which point the boom will take the strain.

On the fore seine line, an additional light line [G] was attached at a little distance out from the boat and made fast on a cleat set into the foredeck [F]. When the haul was over, this line was used to pull the seine line in towards the boat, which now pointed into the wind. A corresponding line was used aft [J].

In the diagram above, the main lines are shown in dark blue, and the lighter lines (G and J), are shown in light cyan. The solid lines show how I imagine they would be before letting out the seine, and the dashed lines as they would be after the net is completely out. Note that the aft line must be led outside the shrouds.

References and Links

1. Eel Drifters, Morten Gøthche, published by the Viking Ship Museum, Roskilde, Denmark, ISBN: 978-87-85180-67-4
   Note: Inch dimensions referenced in this source correspond to the Danish ell of 1835. There is a difference of ~3%.
2. Wooden Boat Designs, Christian Nielsen, Charles Scribner’s Sons, New York, ISBN: 0-684-16432-9, English translation of:
   Danske bådtyper : opmålt og beskrevet, Høst & Søns Forlag, 1982 ISBN 87-14-28286-0
   Note: Inch dimensions referenced in these two sources correspond to the Danish ell of 1835. There is a difference of ~3%.
3. Bådebyggerie på Fejø og de Danske Åledrivkvaser, Christian Nielsen
   I have provided a rough translation.
4. DAN og VIKTORIA, to åledrivkvaser, Morten Gøthche
   I have provided a rough translation.
5. The Encyclopedia Britannica: A Dictionary of Arts, Sciences, Volume 27, page 221
6. Karen of Bogø (website)
7. Replica drifter Tumleren (website)
8. Fejø Drifters (website)
9. Fejø Boat Building (website)
10. Video on eel drifters, Viking Ship Museum, Roskilde
11. Drifter Kvasen (website)
12. Eels in Culture, Fisheries and Science in Denmark, Suzanne Rindom et al., National Institute for Aquatic Resources / Technical University of Denmark, 2014

More links for me to follow on this page