The Hunley Reconstruction

The Hunley Reconstruction

began reconstruction with the center hull, the most straightforward part of the submarine. A number of dimensions were published although I don’t know if they include the concretion that covers the exterior and interior.

In the interest of getting a fairly accurate reconstruction in a relatively short time, I decided not to worry too much about an inch or two. If you use this material please keep in mind that it is my interpretation of the limited available data and is not authoritative. Still, in the many years this page has existed I've not found any better source of this information.
Click the graphic at right for a 3D low-resolution cutaway reconstruction (opens in new window, model takes some seconds to load - browser must support WebGL).

Move the mouse pointer over the line drawing below to highlight features of the reconstruction. Click a highlighted area or the page icons below to go to a description of that feature or to open a related popup window.

A set of ballast weights are attached to the hull by bolts, some of which could be released from inside the cabin.

The captain's station included controls for the rudder and dive planes, a compass, and depth gauge.

The bow and stern are iron castings.

The crew section has stations for eight crewmen although only the seven cranking stations were manned for the last mission.

The Hunley's portholes are in noticeably different positions on the two hatch towers. These differences are very likely by design.

The hull begins to taper at the hatch sections.

Click for some thoughts on the hull shape.

Historical accounts report the center section of the hull was made from a boiler.

The dive planes are positioned forward of the center of floatation.

The propeller was driven by a crank chained geared to a flywheel on the propeller shaft.

The rudder mechanism looks odd. How did the Hunley steer?

A bellows air pump was found beneath the snorkel box.

The upper and lower parts of the hull are joined by strakes along the sides.

The ballast tanks are separated from the cabin by bulkheads, open at the top.

The tapered tank sections are constructed from quarter panels.

The internal steering linkage differs from historical depictions.

The Hunley's torpedo was partly carried on a spar bolted to the bottom of the forward casting.

The deadlights on the top of the hull were more complex than thought.


Ports	Taper	Gearing	Steer	Rudder	Air	Com	Spar

Keel	Tanks	DLs	Hull

The Reconstruction

     In early 2017 the Naval History Command published an extensive report detailing the Hunley recovery operations and much of the archaeology conducted on the sea floor. This material is augmented with photos and drawings that are much more recent. The latest revision of my Hunley plan takes advantage of this report as well as photos after the hull exterior was cleaned of concretion. I have added the likely torpedo configuration and speculative placement of two brackets found near the Hunley that likely fastened the upper spar boom to the hull as shown by Chapman.
     I started my original plan way back with dimension and detail information from the National Park Service Site Assessment Report, available via a link on my Hunley page. Reports and photos of removal of the hull plates provided significant data on the size of the plates. The central hull took form from rivet counts and examination of a number of photographs, and from reported dimensions. I’m confident of the accuracy of the four removed plates, but the two remaining plates are more speculative. Finally, I adjusted the hull shape based on close examination of laser scan images published in the Friends of the Hunley Blue Light newsletter. A more recent change, dividing the formally long most forward and aft tank quarter plates in two, is much less accurate, based on a two or three cropped Post and Courier photos. The photos show a narrow plate on the top of the hull joining the forward ballast tank quarter panels together. I've assumed this plate runs the length of the tank section and that a similar plate is used on the bottom. Likewise I've added these structural details to the aft section. This narrow plate is consistent with Chapman's painting, where it appears as dark line along the top of the aft hull. Chapman also depicts the division of the aft tank quarter-panels mentioned above.
     Click the now obsolete low-resolution drawing below for a detailed five-view plan of the Hunley. (Note that Internet Explorer reduces the image size to fit its window. Click the expansion icon in the lower right-hand corner to view it full size.)

In-work plan - Click for a new 5-view hi-res version (110K)

Click the plan above for a higher resolution version with an educated guess at rivet patterns or click here for a PDF version that prints at 1:24 scale on four letter paper sheets (~300 KB). The latest revision is Y.

The hull taper begins with the hatch panels. The tapered sections consist of 16 quarter panels and the bow and stern castings. The ballast tank sections are forward and aft of the hatch panels (the tank bulkheads are actually located several inches into the sections defined by the plate seams).

Both the bow and stern are iron castings. Chapman’s painting shows the stern casting clearly. The historical reports and the Chapman material have been found to be sources of accurate information. I used them extensively.
For my reconstruction, I've positioned the dive plane above the hull centerline based on interpretation of Chapman's painting and web cam views. Similarly, the painting appears to attach the propeller guard as I've placed it.

The Expansion Strake

The expansion strake, historically described as a metal plate inserted between the upper and lower halves of the boiler, appears to extend the length of the hull between the end castings. Photos released during the excavation indicate that the port-side strake includes a narrower interior plate between and butted to the upper and lower hull plates. Later photos show no such butt-plate on the starboard side. Such asymmetrical construction might be weight compensation for the off-center crank.

Return to line drawing above.

The Cabin

Crew cabin older view, looking forward; starboard is right

The rendering at left shows the interior view looking forward. Many details are speculative, especially far forward. An historical account describes a Hunley predecessor as painted white inside. A white interior would have enhanced cabin lighting. Traces of white paint were found on the bench but Paul Mardikian reports no other paint has been found.

Click the icons below for pop-up panoramas of the cabin.

	(Java required)

The slightly out-dated view above shows my sixth version of the crank. I've mounted it one inch below and ten inches to starboard of the centerline, supported by new brackets. Late interior photos clearly show crank mounts bolted to the strake by an upper plate bolt. The lower part of the mount is hidden in the photos but is most likely bolted at the bottom of the strake. There is a small possibility it is attached to the adjoining frame ring as I thought originally, but I have no photographic evidence of this. Notice that much of the crank mechanism would be below the halfway mark of the excavation, consistent with the delay in uncovering it (remember, the hull lists about 45 to starboard).
Visual evidence places the frames in the sections exposed for excavation about 25 inches apart. I placed the others using interpolation and extrapolation from 22 to 25 inches apart. The crank radius pictured is eight inches. The published photos show varying angles from crank position to position, perhaps more consistent with Alexander's description ("cranks at different angles") and archaeological findings ("evidently bent at odd angles"). To me this suggests that the crank is constructed of sections bolted together. There may have been some loosening and relative shifting of the sections during operation that was not corrected when the joints were tightened. There is a clear benefit of offset cranks: each crewman would rock back and forward out of phase with the next, preventing the action to induce a rolling motion to the boat.

The slightly more recent rendering at right shows details gleaned from the 2002 Smithsonian Seminar presentation and from photos from National Geographic magazine and the updated Friends site. The forward sea-cock is located on the port side of the shelf just below the inlet for the depth gauge. In 2005 the actual gauge was revealed to be three feet high, so this reconstruction is much too small and probably incorrectly positioned (see Depth Gauge below).

Looking forward at the commander's station. (earlier rendering)

     It's still not clear exactly how the tiller connects to the rudder linkage. The photos appear to show a bend at the bottom of the tiller connecting to another feature a few inches farther to port. A long sturdy cylinder seen running the length of the cabin on the floor almost under the bench is actually a pipe between the two ballast tanks but a narrower rod further under the bench is probably the steering linkage. It is possible the sharp angle at the bottom of tiller is a separate feature not associated with the steering mechanism but I think it is connected to the second rod. The depiction above incorrectly shows the tiller attached to the inter-tank plumbing.
     The hatch latching mechanism is in its approximate position and I've made an attempt to show the hinge I believe attaches it to the cabin overhead.
     Lastly, the rendering shows a bench for Dixon, floating in space for now until I can work out mounting details.

I've place the crewmen in the rendering at right. The camera is about at the location of the bellows. It's definitely a tight fit. The rendering below, looking forward, shows one crewman cranking at his station.

The Depth Gauge

I initially positioned a small depth gauge as shown in the older rendering above, but the actual gauge, composed of metal and glass tubing, was nearly three feet tall. Contrary to press reports, comparison of an x-ray published in The Blue Light with older interior photos verifies it was mounted on the forward bulkhead approximately as pictured at left. The shelf mounted on the tank bulkhead appears to have a notch on its bulkhead edge to accommodate the gauge apparatus.
Close examination of earlier photos, especially one published on the Friends of the Hunley conservation images page, reveals features that match details in the x-ray. It’s not clear what is glass and what is metal and there is some informational discontinuity, but I've assumed the tubing visible in the x-ray is metal and that a fairly short sight tube was glass.

It might be helpful to provide just a little information about mercury manometers. A column of mercury is exposed at one end to the medium whose pressure is being measured. The height of the column changes as the pressure changes. Normal atmospheric pressure is about 15 psi or 30 inches of mercury ("Hg) so the large size of gauge should not be a surprise. Depending on the gauge configuration, the pressure is indicated by the difference between the mercury levels in the two legs of gauge or by the level in one leg. The inlet port for the Hunley's gauge is approximately at the waterline that Alexander indicated was the "surface" position. Mercury is 13 or 14 times as heavy as water so a change in depth of one foot would be indicated by a change of less than an inch in the mercury level in the tube. The reports do not say how much of the gauge is metal and how much glass. Only the sight portions of the gauge tubing need be of glass.
The Fall 2016 Blue Light (Vol. 48) reports that the glass tube length was sufficient to indicate 30 feet of depth. Photos show that the upper part of the board is missing, so it would have extended higher than my illustration shows.

I've used a small card as a depth scale for the reconstructed gauge, but it occurred to me that the metal plate suggested recently as part of a possible battery might be part of the gauge. Its 4-inch width is similar to the gauge mounting board and it could fit behind the glass tube. Although it is thicker than needed, its 16-inch height could have provided a depth scale of 15 or 16 feet. However, this is inconsistent with the the longer glass tubing length reported in 2016.

The Crank

The web cam views and published photos show that the handles do not extend at right angles from the axis, but are angled in. The earlier reconstruction above left, with the single crewman, is simpler than the actual crank appears in the photos. It uses a uniform diameter where the photos imply a narrowing from the handle to the journal, as shown above in the other views. The cross sections may not be circular and, as mentioned above, it is possible the crank is constructed from discrete parts rather than a single shaft. Apparent offsets at the handles visible in one early photo appear on closer examination to have been an illusion caused by unremoved fill. The concretion and preservative wrappings made this kind of remote interpretation chancy.
I’ve positioned the crank largely through the analysis of one photo with a wide enough field of view to show several sections of one mounting frame. Unfortunately, the conservator’s hands obscured the crank axis in the photo so its precise positioning is problematic. A later photo provided a clear view of the crank in the foreground but not enough of the frame for easy positioning. Several reference points are needed to account for distortions of perspective and the camera lens.

What are characteristics of the crank?

It needs to be strong and rigid enough to take the torque produced by the strenuous cranking of seven men.
The handle should fit comfortably in the clenched hand of a cranker.
The handle should be wide enough to fit two hands, but narrow enough to fit between the crewman’s legs as he cranks. There needs to be sufficient clearance for his arms as the crank rotates.
The handle probably needs to rotate as the crank turns. Considering the careful engineering revealed as the sub is excavated this is more likely than other less desirable solutions to a basic rotating machinery problem.

The forward section, under the snorkel box.

Alexander’s drawing shows eight cranking stations between the two hatch areas. In fact there are only seven crank positions. There is room on the bench for an additional crewman, but he would only have had the bellows and pump to operate. The diving plane axis crosses the hull just aft of the forward hatch in the snorkel box section at the forward end of the crew bench.
Removal of the air box plate confirmed the location of a frame ring in the expected location. Another forward of the diving plane appears in some later photographs and there are more in the ballast tanks. See the discussion below for more on the rings.

The Bench
The bench has three separate sections totaling nearly 18 feet in length. An early report mentioned bench brackets "bolted to the hull exterior" and fastened to the bench "with square-headed and countersunk fasteners". The brackets are not apparent in any photos released to date. My first guess was that they would be similar to those for the crank and depicted two possibilities in the drawings at right. The upper version would not have been uncovered until excavation was nearly complete. The part fastened to the hull would be embedded in concretion and the bench part only visible from underneath. The second possibility, which seems sturdier, would have been visible sooner, but still might have been hidden by the canteens and other artifacts and features under the bench. In January 05 the conservators removed the bench. At that time the Friends revealed it is at least an inch thick, with the forward section six and a half feet long. Except for cryptic mention of "the various metal brackets holding the bench in place" and the intriguing "C-shaped brackets mounted on the sub's portside hull" no other information or photographs of the mounting system was provided. A photo of the bench did show three fasteners in a line very close to the forward end of the center section. The Post and Courier mention of a "small block of wood attached under the bench" confirms these were fastened to a joiner plate holding two bench sections together. The article emphasizes the discomfort of sitting on these seams, but the joints occur near the stiffener rings, between the crew seats.

Photos published in the the Friends' Blue Light newsletter led me to consider bench mounts attached to the stiffener rings, but the defining information was the Hunley segment on Turner South's Southern Living broadcast in 2005. The lower part of the brackets is not visible in the video and speculative in the graphic. At least one may have the upper fastener bent down rather than up as illustrated. Such a bracket with a different lower part could be easily described as C-shaped.

An informal exchange with reporter Brian Hicks, author of the Post and Courier article, and close examination of the Southern Living video led me to my current reconstruction using the third possibility above, illustrated at right and below. Parts of several brackets are clearly visible in the video, shot during the removal of the aft bench section. All are bolted directly to the hull plates, next to stiffener rings.
Brian informed me that just two fasteners were used to attach each of four brackets to the bench itself and that they are about four inches wide.

A bracket seen from below

Examination of the video supports Brian's recollection of four brackets. There appear to be two for the first section and one each for the others. The joining block, visible in the graphic just above, provides required additional support for the second section. The third must have had another support, probably near its aft end in the jumble of plumbing and concretion visible in available images.
Move the mouse pointer over the graphic at left to see several brackets under the bench.

Many thanks to Kim Johnson for providing Southern Living video, not available in the North, for my analysis.

That the bench is made up of three sections of approximately equal length is interesting. Maria Jacobsen stated that the pieces don't fit well, indicating possible modifications. An 18-foot bench could have been installed while the hull was under construction, but after that replacement could be managed only with shorter pieces. Six-foot sections could not be brought into the cabin through the narrow hatches with their nearly foot-tall towers, but they could be jockeyed in through the snorkel opening if the box and bellows were removed. Chapman's starboard-side drawing indicates the box was removed to access the cabin after the second sinking, indicating this was not necessarily a very difficult operation.

The Stiffener Rings

     The frame rings were the first interior features revealed and presented the first puzzle to me. As the cabin fill was removed, the web cam began to reveal the ring structure. Although there was considerable concretion, I interpreted a center ring sandwiched between two slightly thinner ring plates with a smaller outer diameter. A more massive feature was visible on several rings close to the top of the cabin.
     The relatively massive frame rings separate the crew stations. They are not attached to the upper hull plates and I've seen no other evidence of specific attachment hardware. According to Paul Mardikian, the lower part of these stiffening rings is attached to the hull. My original thinking was that all of the internal mechanics were attached to the rings and that this separate structure slid into the mid-hull plate cylinder. Better photos didn't support this hypothesis. First, the rings appear to have a larger outer diameter than the hull plate bolt ring inner diameter, which would prevent the assembly from sliding in. Secondly, the crank mounts and bench brackets are bolted to the hull.
     The purpose of the rings must be to add strength and rigidity to the hull. The hull plates, strakes, and rivet rings that form the hull are all relatively thin rolled iron. This material is strong but offers very little resistance to perpendicular compression forces. The ribs formed by the frame rings distribute such forces across the hull, effectively countering them. I expected to find similar ring structures in the ballast tanks. A photograph of the initial aft tank excavation show signs of such a stiffener and a corresponding Post and Courier report mentions "internal frames that run inside the sub that gave it lateral strength". Indeed early photos of the 2003 excavation of the ballast tanks revealed rings.
     Photos of the interior taken near the end of the cabin excavation show less concretion on the lower parts of the rings and no obvious evidence of the outer sandwich layers there. The main element appears heavier than the hull plates, perhaps ½" thick. A good deal of the additional material I interpreted as structural may only be concretion, although I'm confident the more massive features are real. These are probably joining plates, used to bolt separate parts of each ring together.

Return to line drawing above.

Keeping an Even Keel

The Hunley has at least one design characteristic that would have helped keep the boat on an even keel. Offsetting the individual crank handle positions by an angle prevented the crew from synchronous back and forth motion. There are other less obvious details. The crew, with the exception of the captain, sat on the port side. Although the confines of the cabin would force them to lean toward the center, contrary to a press release most of their mass would still be on the port side.

This was partly offset by the weight of the crank, which is on the starboard side. More than 60 cast iron ingots of various shapes and sizes visible on the cabin floor likely compensated the rest.
Example weight with hole Speculative rope loop
Many of these have holes that would accommodate cord loops that could have been used to move them around for trimming. The result would be an even keel with the full crew aboard. It’s possible the trim weights would not have been moved when the crew left the boat at dockside. The graphic at right estimates the waterline and list of the empty boat.

The Hunley without crew aboard - speculative

How big is the Hunley? These renderings show the current reconstruction with a figure to scale, based on the Chapman painting. The soldier is 5 foot 8 or 9.
(Chapman actually scaled his figures a bit too large.)

Here are some crewmen seated in the crew cabin. I've placed a horizontal ten-inch wooden bench and crank based on interpretation of early published interior photos.

Poser figures rendered in RayDream Studio

Interestingly, the set used for the TV movie was 10% larger than the published dimensions, probably to make it a little easier for the actors to move around. I'm not tall and I didn't find the cabin set roomy.

What's coming?

I'm slowly refining and improving the 3D model as I speculate about interior mechanical details. Almost no new information has appeared and I'm not optimistic for the near future. I am replacing the generic Poser figures with my reconstructed crew and working on a detailed Chapman reproduction.

Check back now and again for updates

E-mail me with comments

Experience my Hunley archaeological interpretation.

*My Reconstruction Tools* I've occasionally gotten queries about the tools I use to create this reconstruction, so I've added this little section. They've served me very well on this and other projects and their utility goes far beyond this application. I also use several paint programs including Micrografx Picture Publisher, Corel PhotoPaint, and Adobe Photoshop, each of which has a particular strength.
My plans were created in CorelDraw, which I also use for 2D analysis and to create templates for individual 3D objects. Bundled as a graphics suite that includes PhotoPaint, you can buy the latest version this versatile and useful tool at amazon.com by clicking the link. (Earlier versions are also available.)
The 3D models featured on these pages were created in RayDream Studio, a non-defunct program. For later images, I transferred the modeling to Carrara, the RayDream follow-on that combines most of the capabilities of the older program with many new features.
The human figures were developed and posed in Poser, a unique entry-level program that has improved considerably over the years. It works by itself or with Carrara directly or through a plug-in, depending on the version. You can purchase the latest version of Poser at amazon.com, by clicking the link. (Earlier versions are also available.)