Sunday 19 November 2017


Much of a wooden ship’s strength comes from its planking, which provides longitudinal stiffness while unifying the frame both outboard and inboard. It is no surprise, then, that special attention was paid to HMS Terror’s planking; originally, to protect her from the recoil of her massive mortars within, and later to protect her from the immense force of polar ice without. These combined pressures resulted in the construction of perhaps the toughest wooden sailing vessel the world has ever seen. She was, in her way, the pinnacle of the wooden shipwright’s art - embodied in a squat, slab-sided bomb, with the lines and dull sailing qualities of a merchantman. How can you not love HMS Terror?

My reconstruction of Terror's 1845 planking plan. 

As a bomb vessel which needed to endure the punishing impact of her mortars, a concern for increased strength was apparent in Terror’s original planking plan. The dockyard contract for her sister ship, HMS Beelzebub (1), reveals that she was originally clad in four-inch thick oak planks, exceptional for a ship of her tonnage. English oak was used for planking her weatherworks (above the waterline) while English or Dantzic (Polish) oak was used below the waterline. Terror’s original wales were five and one-half inches thick, wrought “hook and butt” fashion, which is an interlocking design common only on warships (and bombs in particular) which greatly increased the strength of her hull. I initially believed that the second layer of planking would be wrought hook and butt fashion as well, but pictures released by Parks Canada of Erebus, and detailed images of Terror painted by Owen Stanley in 1836-1837, show that top and butt planking was employed on the second layer. How many strakes were wrought in this fashion hasn’t been reported by the archaeologists, but the very detailed paintings of Terror by Owen Stanley show two normal thick strakes below the chocks and the strakes below that wrought top and butt. This is the configuration I opted to replicate on my model.

The refit of HMS Terror for polar duty in 1835 followed a reinforcement plan that had been well-established from the time of William Edward Parry. Like HMS Fury and Hecla, Terror’s forecastle and quarter deck were joined to form a continuous upper deck, thereby greatly increasing the enclosed space below. Some minor improvements to this planking system (described in Part 2) were applied to Terror (and then Erebus) for James Clark Ross’ Antarctic expedition in 1839. These modifications were carefully described by Rice, the master shipwright who outfitted both ships for the Antarctic (2):

” the side is doubled with six-inch oak plank under the channel, increasing to eight-inch at the wale, which is three feet broad; from thence, through a space of five feet, the doubling diminishes to three inches in thickness, of English elm, and the remainder of the bottom to the keel is doubled with three-inch Canada elm….”

Midship Section Plan for Erebus and Terror. 
National Maritime Museum, ZAZ5678
Copyright: CC BY-NC-SA

Beelzebub’s contract (1) indicates that Terror was planked with a three-plank shift (meaning that on each frame, three planks occurred between each plank butt, or plank end), which appears to have also been used on the subsequent second layer of planking (at least on Erebus). We don’t know how many strakes were used on Terror’s hull, but the 1839 Erebus cross section plan (3) indicates that 34 strakes (plus the garboard strake) were used on both the outer and inner layers below the chock channels.

Lang’s 1845 stern modification plan (4) provides crucial details on the planking configuration used for the modified stern. In addition to the general planking run, it reveals that the hull planks averaged between 9.5 and 10 inches in width, and that the garboard strake was unusually broad, about twice the width of the bottom planks. Because of the extra-wide garboard strake, and the smaller overall dimensions of Terror, I estimated that Terror was planked with 30 strakes plus the wide garboard (assuming each strake was 9.5 to 10 inches in width).

1845 Stern Profile form Erebus and Terror. Note the planking detail.
National Maritime Museum, ZAZ5683 
Copyright: CC BY-NC-SA

I made two fundamental mistakes when planning to plank Terror’s hull in 2013. The first of these was deciding to double-plank her hull, just like the real ship. I began planking my model in the fall of 2016. In a previous blog post, I showed how I planked Terror’s topside weatherworks above the chocks. I followed a similar methodology for the hull, but rather than edge-bending planks, I had to carefully spile them (a technique used by actual shipwrights to calculate, then cut, the precise curve and run of a plank), as this was necessary to plank the bluffer parts of Terror’s bow.

The first strake added to the hull. For fellow ship modelers: the white line was my first inadequate
 attempt to line off the bottom of the wales. I fixed the run shortly after this picture was taken. 

The first strake at the bow. This image reveals how bluff Terror was just above the waterline.
Again, this was taken before I adjusted the reference line on the hull. 

Spiling greatly slowed my progress, and I was only able to complete one or two strakes in an evening. With 120 strakes necessary for both layers of hull planking, not including the ice chocks, stern, and upper deck, I rapidly realized that I was facing a crisis. The deadline to deliver my model for the Death in the Ice exhibition was in June of 2017, and it quickly became apparent that I would not make that deadline if I did not increase my output. Thus, I began the great planking of 2017.

It began with a compromise. To speed up the first layer of planking, I decided to double the width of each of the lower hull strakes and to not follow the stated plank shift pattern (which requires more cutting). While this layer will never be visible, I regret not having completed it to scale; in addition, not having a photograph of Terror’s original planking configuration remains a sincere source of dissatisfaction for me.  

My second mistake was to plank the second layer of my model using accurate scale plank thicknesses. Some of the planks on Terror’s wales are 8 inches thick, representing a daunting task at 1:48th scale. While the three and four-inch scale planks could easily be bent with a crimping tool and some heat from a blow dryer, this technique simply would not work on planks thicker than five scale inches. Every thick plank had to be soaked in near-boiling water for 20 minutes, carefully crimped with a plank bender, and then pressed into shape using a bending iron and a curved jig. On top of that, each plank had to be carefully spiled before bending, and the distortion caused by swelling wood and heat treatment caused no end of difficulty. An added complication was that the thickest strakes, at the wales, had to be laid top and butt fashion, which further complicated the spiling process.

The second layer of planking in progress. You can see here where I made the decision to
widen each strake on the first layer (about January 2017).

Detail showing the transition to 8" strakes at wales. The upper two 8" stakes were sanded 
to provide a smooth run to the 6" planks above them.    

The third strake of  8" top and butt planking at the bow.Notice the
 drop strake below it in the first layer of planking. 

Top and butt planking in progress. The tape protects
the wood at the stern and bow during planking.

Detail of the completed top and butt planking. 

Adding the absurdly wide garboard strake on the second layer. 

Close up of the garboard strake at midships. According to contemporary plans,
the garboard strake of the second layer was not rabbeted into the keel.
Interestingly, it was on later polar vessels, like HMS Investigator. 

The final garboard planks at the stern, after bending them into shape. 

Bottom planking in progress. This photo shows how I lined off the second layer of planking. 

The most difficult part of the hull planking occurred with the stern. The first layer of planking was relatively simple as it abutted the rabbet on the original stern post (hence this layer was planked like every other ship). However, Lang’s conversion of Erebus and Terror to steam locomotion required that the second layer of planking form the walls of the propeller well (4). This meant that the second layer extended over the original stern post and propeller well and was rabbeted into the new rudder post. I’ve known for some time what shape this configuration would take, but implementing it required a lot of trial and error, despite Lang’s detailed plans and a block model (5) for guidance. The most difficult chore was bending and spilling the planks into the proper shape, especially the strake forming the lower margin of the well. It also required the use of two “stealers” to accommodate the increased area of the stern. However, once installed, I’m convinced the model respects Lang’s design, the 1845 block model, and the practical reality faced by the shipwrights who had to plank this unusual ship.

I use masking tape to make spiling templates. This shows the extreme shape of the
first stern plank above the propeller opening at the stern. Lang didn't make the
shipwright's task easy!

The resulting pearwood plank. 

To achieve the complex bend in the plank, I soaked it in hot
water, then clamped it in place until it dried.

One of the stealer planks in the stern, after it had been bent to shape. 

Terror's unusual stern, prior to sanding and finish. Oliver Lang designed only one stealer
in this area , but I found it impossible to plank without a second.  I'll discuss the
planking of the transom and chock channel in Part 2.

The completed second layer at the bow. Note the drop planks below the wales.
On the finished model this is completely covered by a third layer of wood
and "iron" plating. I didn't need to spend such care at the bow, but a
modeler can only accept so much compromise. I'll discuss the
planking of the chock channel "ice bumper" in Part 2.

The completed planking prior to sanding, bow plating, and finish. I'll discuss the
planking of the chock channel "ice bumper" in Part 2

Part 2 of my post details the planking of Terror’s chock channels (or “ice bumper”), her transom, and her upper deck. Stay tuned!


(1) National Maritime Museum, ADT0010
(2)Ross, Sir James Clark, 1847. A Voyage of Discovery and Research in the Southern and Antarctic Regions, During the Years 1839-1843: Volume I. John Murray, London. 
(3)National Maritime Museum, ZAZ5678
(4)National Maritime Museum, ZAZ5683
(5)National Maritime Museum, SLR2253

Tuesday 8 August 2017


In my last post, I revealed that my Terror model would be displayed in the “Death in the Ice”
exhibition, now showing at the National Maritime Museum in Greenwich. On July 13th, I was fortunate to see the exhibition at a private event for experts, dignitaries, and those who had contributed knowledge or items to the exhibition. I was very impressed by the show, which provides a detailed review of the history of the expedition and tracks the fascinating trail of clues recovered by myriad Franklin searchers, culminating in the identification of HMS Erebus and Terror by Parks Canada. 

Accompanying the exhibition is a beautiful display of 129 flags, one for each crew member of the Franklin Expedition.
The flags for Franklin and Crozier can be seen in the front row.

My model is placed relatively early in the show, in a section detailing the construction and outfitting of the expedition’s ships. Remarkably, it is surrounded by some of the greatest art and technical plans associated with Erebus and Terror.  

The case containing my model. On the left is the original 1839 Admiralty builder's
model of HMS Erebus; my Terror is on the right.  A large scale reproduction of the
1839 cross section plan for Terror and Erebus is placed at the back of the case. 

In my previous post, I outlined that my model is not yet complete because I intentionally paused construction for the show. Currently, my Terror is a slightly more detailed version of a “builder’s model,” and is missing many fittings. The reason for this can now be revealed; my Terror is displayed in the same case as the original 1839 Admiralty builder’s model of Erebus. The exhibition team wished to show the differences in size between the two ships and, in particular, the new fittings installed for the 1845 polar expedition when compared to the 1839 configuration. For this reason, they requested that Terror approach the same level of detail as the 1839 builder’s model.

A view from their bows . 

I admit a degree of anxiety came with this great opportunity. My version of Terror is modeled at the standard 1:48 Admiralty scale, the same as the builder’s model of Erebus. However, my plans are custom hybrids created by concatenating details from multiple plans spanning a 33-year period, with additional alterations based on contemporary historical sources. I had measured, scaled, remeasured, and measured again to ensure the scale and dimensions were correct, but I still worried that something was amiss.  I had nightmares that museum staff would open the packing crate only to find that my Terror was larger than Erebus, or had the wrong bow shape, or some other fatal flaw.

I’m happy to say that my Terror appears to have just the right proportions, with all the necessary similarities and differences to Erebus. It is slightly smaller than Erebus in length, height, and breadth, but with nearly identical curves and lines, and with the characteristic difference in bow shape. In short, all appears as it should be.

A view from the stern. 

It is sincerely humbling to have my Terror displayed next to the 178-year-old model of Erebus. The list of famous Royal Navy explorers and shipwrights who studied and touched this object is astonishing; a very truncated list includes Sir James Clark Ross, Francis Crozier, Sir John Franklin, Sir John Ross, Sir Edward Belcher, Sir William Parry, Oliver Lang, John Rice, and Sir John Barrow. Sir Ernest Shackleton and Robert Falcon Scott are also thought to have consulted the plans and models of Erebus and Terror before outfitting their ships for the Antarctic. I’m frankly overwhelmed that my model is now breathing the same air as this incredible object of history.  

Tuesday 13 June 2017


My blog has been silent for nearly five months, but I have an excellent excuse.  Since December, I’ve been working night and day on HMS Terror. Why the rush?

This week, my model is carefully being packaged for transport to Greenwich. In one month, it will be displayed in a new exhibition on the Franklin tragedy at the National Maritime Museum.

Mini-Crozier stands at his taffrail. 

Created by the Canadian Museum of History, Parks Canada, and the National Maritime Museum, the exhibition will open in Greenwich on July 14th.  In January, the model will travel back across the Atlantic with the exhibition, where it will be displayed at Canada’s national museum beginning on March 1st, 2018.  

A view from the bow. 

Port side planking. It took nearly six months of work to double
(and in some cases tipple) plank this hull with scale timber. 

What does this mean for my project?

  1. First and foremost, it is a sincere honour to have been asked to display my model alongside iconic artifacts related to the Franklin Expedition. I jumped at the chance to loan it when it was presented to me, despite the short time frame involved.
  2. Due to time constraints, progress on my model has now far outpaced my blog. I’ll be playing catch up for the next several months; keep visiting to see all new historical research and build photos.
  3. My model is not finished. Currently, it most closely resembles a shipwright’s “builder’s model” which typically only show the design and major fittings of a ship. The reason for this will become clear when you see the model in its position in the exhibition. I admit that I couldn’t help but add a few extra details, but it’s essentially just a builder’s model right now.  When the model is returned to me, I’ll complete all the finer details; I estimate it’s about ¾ complete.

Missing details can be seen in this view; the tiller, deck houses
and conning (ice) plank are all absent.

A cathead with its iron knee. 

Mini-Crozier keeps watch on the voyage. 

I won’t show all the details of the model in this blog post; I intend to catch up over time while it is away from my workbench. To see the entire thing, you will have to wait for my blog to catch up to the model, or go see the exhibition!

When I began this hobby project four years ago, I couldn’t have imagined the interest it would generate, or the great connections and friends I would make. I’m very happy that my model will help to tell the fascinating story of the Franklin Expedition in this new exhibition. In fact, what they have planned for my HMS Terror is beyond any of my expectations. You’ll have to see the show to find out, but I promise it will have great company. 

The bow plating is made from 100 chemically blackend brass plates. 

Sunday 12 February 2017


The 1835/1836 modifications to HMS Terror introduced a new fitting to Royal Navy polar exploration vessels– the hatched companionway. Covered companionways began to replace open ladderways on smaller Royal Navy ships in the late 18th century. However, the special type of hatched companionway associated with Erebus and Terror is rare on Royal Navy plans and models until the middle of the 19th century.

The aft companionway of HMS Terror was a simple box-shaped structure, with a large sliding hatch on its roof. To access the ladderway, the crew would slide the hatch back, and then pull open two small starboard-facing doors. They would then descend the ladderway backwards, pausing midway to close the doors, and then the hatch, behind them.

The Companionways on HMS Terror

The 1836/1837 Terror profile plan (1) indicates that this companionway could be removed and replaced with a tall winter deckhouse with a standard-sized door (it was apparently shaped like Terror’s water closet). However, pencil marks on these draughts indicate that this special winter coaming was abandoned sometime after Back’s 1836/1837 Arctic voyage.
Prior to 1839, there was no raised companionway in the forward part of Terror. Instead, an unusual hinged trapdoor system was used to access the forward ladderway. Curiously, this ladderway was located on the starboard side of the vessel, and not on the midline as was typical. This feature was changed in 1839, and a raised companionway identical in design, but somewhat larger, to Terror’s aft companionway was installed (2).
Unlike its aft counterpart, its doors faced port, and it was located just behind the funnel for the ship’s stove.  The ladderway below it descended just aft of the ship’s stove. This must have been somewhat inconvenient, as using it would have introduced terrible drafts to the lower deck mess and sleeping area during the winter months.

Besides the clear labeling on the 1839 Terror and Erebus profile plans (2), evidence for this change in the forward companionway’s position can be found on the 1836 profile plan of Terror (1) which shows that the 1836 trapdoor system was scratched out in pencil. The forward ladderway is scratched out on this plan as well, and a new position, consistent with the 1839 profile plan, has been penciled in.

The companionways on the 1839 builder’s model of HMS Erebus (3) are consistent with the 1839 plans (2). However, the model reveals an additional detail, namely that the companionways had two horizontal tracks which facilitated the sliding of the upper hatch. These rails would have been sheathed in a thin layer of brass or bronze on their upper surfaces, and the hatch itself would have had two grooves cut into its bottom surface (and through its forward edge), permitting it to slide on the tracks.

On both the 1836/1837 and 1839 plans, and crucially, the model, the roof of the companionways and the hatches themselves are flat. This contrasts with most companionways of the era, which had a slight camber (often matching the camber of the upper deck). The reason for this unusual trait is unknown, but plans for HMS Investigator show the same flat-topped companionways, suggesting it may have been specific to polar exploration. Whatever the advantage of this specific trait, the overall design must have been very durable; indeed its efficacy is easily reflected in the fact that the basic design is still used on modern sailing craft.  

(1) National Maritime Museum ZAZ5672
(2) National Maritime Museum ZAZ5673
(3) National Maritime Museum SLR0715

The basic components of the companionway were cut from pear 
wood stock using my local Library’s laser cutter. 

Tracks were added to the roof of the companionway. 

Terror’s companionways were very simple box-like structures.

I had trouble simulating the bronze tracks with brass sheeting, 
so I opted to use a brass foil product here. 

Preparing to cut the grooves in the aft hatch. 
The companionways with hatches and tracks 

The completed forward companionway. 

The completed aft companionway. 

Mini-Crozier inspects the workmanship. A coat of Minwax 
Wipe-On Poly provides a protective finish that enhances 
the wood. 

These structures were neither large nor comfortable. Robustness 
seems to have been the primary design feature. 

The 1839 Erebus model indicates that the doors had small knobs, 
which I recreated using brass pins filed to the correct size.  I 
elected to show the doors with hidden door hinges (the doors 
swung outwards), which is common on modern 
companionways of the same design.