Philadelphia Quaker Joshua Humphreys, who designed the U. S. Navy’s first frigates in the 1790s, believed that a young country with limited resources should concentrate on building ships big enough to outfight anything of their type—and fast enough to get away from anything stronger. He also believed in building ships to the highest standards, using the finest materials available to ensure long service lives— a realistic goal, because the nation’s forests could furnish a cornucopia of all woods a shipwright of that era could desire.
Beginning with a hull form influenced by French practices, and using British norms for dimensioning structural members, he added his own unique contribution—prestressed diagonals—and created the design for the unparalleled champion USS Constitution—’’Old Ironsides”—and others.
Sailing frigates required wood from more than 1,500 trees, cut and shaped into many thousands of parts that were joined into what was hoped would be a strong, unified structure. But when that structure was a ship more than 200-feet long that carried large weights even near its extremes and had to contend with the dynamic motion of the sea, it was almost inevitable that the structure would suffer some distortion. The major manifestation was a condition called “hogging,” where the ends of the ship bowed downward, leaving the midpoint of the keel higher than either extreme. This affected the ship’s speed and maneuverability, not to mention her longevity.
In the late 18th century, many of the world’s navies commonly used short, heavy pieces of wood attached diagonally over sections of the inner hull to help stiffen it and counteract hogging. Joshua Humphreys, with warship design experience dating back to the American Revolution, was aware of this, of course, but he took the idea much further in his design for a 44-gun frigate.
Some understanding of wooden-ship construction is necessary to appreciate what Humphreys did. The backbone of a ship is its keel, a single beam composed of several interlocked pieces that is nearly as long as the ship’s final length. At right angles to this backbone, and at regularly spaced intervals, are the frames—the ship’s curved ribs that give the final product its shape. In Humphreys’s design, these frames fitted over and around the keel, clamped to it by the keelson—a near-twin to the keel fastened over it by stout copper pins more than one-inch thick and peened over at both ends to hold them securely.
The frames themselves consisted of several pieces called “futtocks” laid parallel to one another and pinned together with locust treenails in such a way that no point where the ends of two consecutive futtocks butted together was directly opposite a similar point in the adjacent futtocks. The frames were sistered so that each pair presented a unit two feet wide and nearly one foot thick. Humphreys’s design allowed just two inches of space between each pair of sis- tered frames.
The horizontal wales and planking of the ship’s external skin were fastened parallel to the waterline on the outside of the vertically rising frames. The thickest timbers were placed above the waterline and just below the gun ports, and all the timbers were joined end-to-end. Similar horizontal ceiling planking was laid inside the frames, and the three layers were united by metal pins driven through them and clinched at both ends for security.
Once the shell of the ship was erected, its volume could be filled in with the necessary decks. Where deck beams spanned the interval from side to side, two strakes of timber—thicker and heavier than the ceiling planking and together called a clamp— were installed to support them. At this point, with the berth-deck beams in place, Humphreys’s introduced his innovation.
He called for long, curving pieces of wood called “diagonal riders” to be installed between the keel and the berth-deck beams—three pairs of diagonals in the forward part of the ship and a similar number aft.
In the forward set, the rearmost butted to the keel/keelson—and to the forwardmost of the after set—at the ship’s midpoint and ran outboard, forward, and upward to attach to either end of the eighth berth-deck beam (numbered aft from the bow). The next began two frames forward of the midpoint and ran up to the sixth berth-deck beam. The forwardmost pair began another two frames forward and were connected to the fourth deck beam from the bow. The after diagonals were mirror images of those and connected to the eighth, sixth, and fourth berth-deck beams (numbered forward from the stern).
The diagonals were through- bolted to the hull at two-foot intervals. All, except for the midship diagonal riders that butted against each other, were tenoned into the keelson.
Humphreys also called for the installation of two “transom riders” under the berth deck aft of the diagonal riders, specifying only that they be 18-feet long and one-foot square. These probably were placed parallel to the diagonal riders—perhaps with one-half the spacing—but precisely where their respective end points were fastened remains unknown; they were bolted to “every other timber.” At the forward extremity of the ship, Humphreys specified five massive breast hooks straddling the stem and keelson from just below the berth deck to immediately aft of the foremast.
The diagonals in effect transferred some of the bending moments at the frigate’s extremities to a point near the ship’s center of bending at the keel, and largely counteracted the detrimental effects. Rough modern model tests found that, whereas a load of about 800 pounds would cause the keel to deflect almost three-quarters of an inch on a model without diagonals, almost three times the load— 2,250 pounds—caused a deflection of less than one-half inch on a model with diagonals installed. Subsequent computer modelling by the David W. Taylor Ship Research and Development Center in Maryland confirmed the importance of the diagonals in offsetting hogging.
At the berth-deck level, Humphreys’s design provided for four pairs of white oak “thick planks” to be installed along the length of the ship. One pair was located on each side of the hatches piercing this deck; the others, farther outboard, were midway between the hatches and the ship’s hull. Each was five-and-one-half inches thick and at least ten inches wide, and the sections were lock-scarfed together. All were worked to fit over and into the beams and ledges they crossed, and all were firmly attached to the ends of the ship with stout knees of white oak.' These became an upper keel, a series of unified members running the ship’s entire length at that level.
Also at the berth-deck level, Humphreys called for the pieces of the two strakes of spirketing—the thick pieces occupying the space above the waterways filling the line of jointure of deck beams with the ship’s frames— to be “hooked and joggled” into each other. This made them fit like pieces in a jigsaw puzzle, and meant that the spirketing running all around the perimeter of the berth deck became a fence that tied thick planks, deck beams, and frames into a single, large framework. Structurally, everything in the hull was tied, in turn, to the keel/keel- son by the diagonals.
Humphreys further reinforced the berth deck against the strains imposed on it by the diagonal riders thrusting up under six of its beams by providing a series of standard knees fastened to the ship’s sides and, through the deck plank, to the ends of the deck beams below. A total of 12 were provided for each side, so that not only were the beams abutting diagonals reinforced, but so, too, was every other beam in the midships’ area between the upper ends of the midships’ diagonals.
As innovative and effective as this system was, Humphreys did not stop there. Between the berth-deck beams and those of the gun deck above it, he provided for three rows of vertical stanchions, one along the centerline of the ship, and another parallel to it on either side. The outer-line stanchions probably were placed on the outer row of thick planks under gun-deck beams located immediately inboard of gun ports. This put each stanchion approximately under each 24-pounder long gun and helped to distribute the weight of the guns—about 6,500 pounds each—throughout the ship’s structure.
Finally, six-inch-thick planks were installed in the gun deck in the same pattern followed one deck below, and again attached to the ends of the ship with knees. The spirketing here also was hooked and joggled to form a solid perimeter about the deck. The effect of the thick strakes and locked spirketing on the gun- and berth-decks was to limit the spreading possible at butt joints. This, in turn, reduced the structure’s ability to droop or hog at its extremities. A centerline row of movable stanchions helped support the spar deck.
Humphreys’s system resulted in a unified gun deck that relied on stanchions to help distribute its enormous weight to a unified berth deck that, in turn, formed one side of a longitudinal girder, triangular in cross-section, whose other two sides were diagonals with their apex at the ship’s keel. This system gave the ship enough flexibility to work easily in a seaway while retaining the inherent stiffness that counteracted the natural tendency of a long wooden warship to hog.
Joshua Humphreys left no clues as to the source of his inspiration, although it may have come from the Dutch- originated bow roof, incorporating curved beams, found in barns and houses in Pennsylvania, or it may have come from the spate of bridge construction that occurred in the United States during the 1790s, when architects rediscovered the truss systems employed in European bridges two-centuries-and-more earlier and began incorporating them into their designs. Diagonals similar to those Humphreys used were features of both the “queen truss” and “Warren truss” bridge designs.
The normal shape of a ship’s hull allowed Humphreys to strengthen the system further by using prestressed timbers for the diagonals. These great baulks of wood, softened in long “trunks” of boiling salt water and then bent to shape, took on all the characteristics of an archer’s bow, resisting any forces tending to distort them—particularly on the outside of the curve. Humphreys transmitted their advantages to his entire structure by fastening the diagonals through the total thickness of the ship’s external planking, frames, and ceiling timbers fabric at every frame they crossed. Furthermore, as a result of the way he placed the diagonals, more than half the hull was reinforced by a least two of them. It was an original concept not used before by any nation’s shipbuilders—a true stroke of genius.2
Humphreys’s diagonal system is known to have been built into the Constitution’s sister, the United States, and her near-sister, the President. Not all the ships got them, though; David Stodder, constructor of the 38-gun Constellation, was ordered by Secretary of the Navy Benjamin Stoddert, to omit them—at the request of Captain Thomas Truxtun, her prospective commander—and Josiah Fox, one of Humphreys’s subordinates who later controlled the building of the 36-gun Chesapeake, is known to have disagreed with Humphreys on this and many other design features, and undoubtedly left them out of his redesigned ship—the last of the original six frigates to be built. Diagonals were included in the 36-gun Congress, as built, and in Humphreys’s design for ships of the line authorized in 1799, but subsequently canceled by the Jefferson administration. After that, Joshua Humphreys designed no more ships for the Navy.
At some point in the Constitution’s long history, possibly as early as the mid-1850s, when she was being readied as a stationary school ship for the U.S. Naval Academy, the diagonals were removed and not replaced. That they ever existed was lost to historians until more than a century later, when the ship’s captain of that era rediscovered Joshua Humphreys’s paper calling for them in his design. This discovery came too late to be confirmed as anything more than a design concept and therefore no diagonals were used in the ship’s restoration then in progress—the information was merely filed. Now, as she is being readied for her bicentennial, the original system of diagonal riders is being installed in conformance with Humphreys’s original design.
This integrated structural system with its six pairs of diagonals spanning much of a sailing ship’s lower hull, then, is Joshua Humphreys’s unique contribution to the art of warship design. The British and the French, at least, already had built large 44-gun frigates, but Humphreys was the first to employ full-length (keel to lower deck beams), one-piece, pre-stressed diagonals in frigate design and to unify strength members at lower- and main-deck levels.
Humphreys did far more than successfully blend French proportions for speed with British practices for sturdiness— his diagonal riders greatly increased his ships’ strength, ensuring that their size and power would endure.3
1. The term “knees” refers to a naturally angled piece of wood, such as might be cut from the area of a tree including a portion of the trunk and the near portion of a large limb. Those involved here subtended obtuse angles, roughly L-shaped, the longer arm extending horizontally on deck and the shorter one rising against the hull.
2. Humphreys’s innovation points not only to his genius; it is a testimonial to the American shipwrights’ skills in those days. To work, all of the interlocking pieces of this huge “Chinese puzzle” had to fit together with precision. That the constructors were able to do it is clear evidence of their high technical abilities.
3. Frederick H. af Chapman, the noted late-18th century Swedish naval architect, designed a “privateer frigate” which included a ship-long system of stanchions and diagonals limited to the plane between the keelson and the gun deck. The first British use of diagonals, close to Humphreys’s concept although used in a ship of the line, was introduced by Sir Robert Seppings in the Tremendous, built in 1811. After becoming Surveyor of the Royal Navy in 1813, he altered the diagonals idea into almost a basket-weave structure, termed a trussed frame, employing shorter timbers in a criss-cross network of reinforcement from keel to berth-deck beams against the inner sides of the ship. He probably was forced to do this by the lack of sufficient timbers of adequate size in England, a factor Joshua Humphreys did not have to consider.
Note: Joshua Humphreys’s paper, “Dimensions and Sizes of Materials for Building a Frigate of Forty-Four Guns,” can be found in American State Papers, Naval Affairs, Vol. I, Paper No. 2 (enclosure), [Washington, D.C.; Gales and Seaton, 1834]. Several articles and comments on Humphreys and his designs have been published in Naval History: “Who Did Design the First U.S. Frigates,” W. P. Bass, Summer 1991, pp. 49-54; W. M. P. Dunne, pp. 3-4, and T. G. Martin, p. 4, Fall 1991; and W. P. Bass, Winter 1991, pp. 7-8.