Catapult Dimensions and Assembly

III. The after cross-piece. This is 15 in. high and 1 ft. thick.

IV. The forward cross-piece. This is 21 in. high and 1 ft. thick.

The cross-pieces (III, IV) are cut into tenons at their extremities and mortised into the side pieces I, II.

V. The small cross-piece (6 in. square). This gives additional support to the sides of the catapult to enable it to resist the immense force of the skein of twisted cord.

The inside width between the sides of the catapult (I, II), when the cross-pieces (III, IV, V) are fixed, is 4 ft.

A, A. The skein of twisted cord. The ends of the skein turn over the crossbars of the large wheels (B, B), which twist the skein. See figures. 197 and 199.

C, C. The pinion wheels which turn the large wheels, B, B.

By turning with long spanners the spindle heads (D, D), of the pinion wheels (C, C), the large wheels (B, B), revolve and twist the skein of cord (A, A), between the halves of which the catapult arm (E, E), is placed.

The skein of cord (A, A), is 8 in. in diameter.

F, F. The roller (7 in. in diameter), which winds down the arm, E, E.

The two small cogged wheels, with their checks, which are fitted on the ends of the spindle (G, G), prevent the roller from reversing whilst the catapult arm is being wound down, figure 194.

H, H. The mortises cut in the sides of the catapult to receive the tenons of the two uprights. Between the tops of these uprights is fixed the cross bar against which the arm of the catapult rests, or when released from its catch strikes. The uprights and the cross bar are shown in figures. 194, 195, 196.

It will be noticed that the mortises for the tenons of the uprights, are placed well away from the circular openings in the sides of the catapult through which the skein of cord passes. If these mortises were cut too near the openings for the skein, the side pieces of the catapult would be weakened.

K, K, K. The mortises for the lower tenons of the three sloping supports which prevent the two uprights, and their crossbar, from giving way under the blow of the released arm of the catapult, figures 195 and 202.

The upper ends of the two side supports are mortised into the tops of the uprights, to which they are also bolted, figure 194 and 202.

The top of the middle support is mortised into the center of the cross-bar that connects the uprights, figure 195 and figure 202.

 Catapult Experiments and Testing

The upper part or bend of the rope is hitched by a slip hook to a ring bolt which passes through the arm of the catapult. Figure. 200 describes the ringbolt and the slip hook.

B. The position of the catapult arm when it is fully wound down by the roller. The stone may be seen in the cup of the arm.

By pulling the cord (E), the arm is released from the slip hook and , taking an upward sweep of 90 degrees (see curved line of arrows), returns to its original position, as at A.

C. The position of the arm of the catapult at the moment when the stone leaves it. The stone is projected upwards at an angle of about 45 degrees, as represented by the straight line of small arrows that indicates its flight after it leaves the arm at C.

When the arm reaches the point in its upward sweep at which its speed is greatest, the stone instantly flies away in front of it. That is to say, when the arm decreases in speed, however slightly, it cannot keep pace with the stone it projected the moment it reached its maximum velocity.

This principle should apply equally to the bow and its arrow. In this case I believe the arrow leaves the bowstring before the latter has returned to its position of rest, or as it was before it was pulled back by the archer to discharge the arrow.

When I originally directed my attention to the construction of a catapult I concluded that the medieval catapult plans and drawings, which depicted the arm of the catapult in a perpendicular position, as in A, figure 194, were incorrect.

My surmise was that a catapult design with a perpendicular arm would merely bowl its stone along the ground, on the principle that the stone was retained in the cup of the arm till the latter was checked by the cross-bar.

Carrying out this idea, I placed the winches of the first catapult I made in front of the uprights and not behind them as in the weapon here described. By this catapult design  the arm when released had of course an upward inclination when checked by its cross-bar. Such a position for example as half-way between C and A, figure 194.

The result of this intended improvement on the ancient catapult design was:

1. The cross-bar which checked the arm of the catapult was soon knocked loose through being struck in an upward direction.

2. The range of the projectile was unsatisfactory through the arm being wound down only a short distance from its state of rest.

3. The projectile - as in the case of a perpendicular arm - left its cup a considerable time before the arm encountered the cross-bar.

On the other hand I found that:

1. The cross-bar was struck a level blow, or one that was taken by the three supports which lean against its center and ends.

2. The range of the projectile was much increased owing to the additional distance the catapult arm was wound down, and which caused the skein of cord to be far more tightly twisted than it was when the arm rested against the cross-bar in a sloping position before it was pulled back.

3. The projectile left the cup of the arm as shown at C, figure 194, and as it did with a sloped arm.

Detailed Catapult Plans Design

Figure 95 shows the large front cross-piece (IV, figure 193), between the sides of the catapult, as well as the three supports that hold the uprights and the cross-bar from movement when the latter is violently struck by the released arm.

We also see in figure 196 the padded cushion against which the arm strikes with terrific force when its upper end is checked by the cross-bar. The cushion is of the same depth as the cross bar. It is 16 in. long and about 6 in. thick. It is made of soft hide, doubled and packed with horsehair, and should be nailed to the cross-bar. 
Without this protection the catapult arm and cross bar would soon be shattered.

The tendency of the arm of a catapult is always to draw out of the skein of cord, in which its butt end is placed. This is the result of the strain applied to the arm when it is being wound down by the roller. To prevent this slipping of the arm its butt end should be slightly increased in bulk, as shown in figure 193.

The cup or circular hollow at the end of the arm - in which the stone is laid  is 5 in. wide and 2 in. deep at its center.

The arm should be tightly bound at short intervals with lashings of quarter inch cord, figure 196. Sometimes an arm will endure the great strain applied to it from the first and show no sign of fracture, though it may bend not a little when, it is wound down to its full extent.

It is, however, probable that the first arm or two tried in the catapult will give way, especially if too much initial pressure is put upon them. The arm should be tested by degrees and only pulled down its full distance after several trials at shorter ones.

The ancients had the same difficulty in obtaining arms for their large catapults that I have experienced with smaller ones. For this reason their engineers constructed the arm of a catapult of three longitudinal pieces. They first fastened three smooth and closely fitting planks together with glue and with small rivets; then they shaped the planks, thus held together, into an arm of correct size and outline.

The catapult arm, except its enlarged head end, was next wrapped tightly round its entire length with several layers, one above the other, of strong linen soaked in glue, the linen being cut in strips about 3 in. wide.

Finally strong cord, also soaked in glue, was closely lashed over the linen from the butt end of the arm to the cup for the stone.

The arm was made on the same principle as a carriage spring, or a longbow of several pieces, and was infinitely stronger and more elastic than one formed of solid wood.

These are the most important parts of the catapult, and generate its projectile force. However carefully a catapult may be built, its effectiveness chiefly depends upon the two winches that twist the skein of cord in which its arm works.

The plans in figure 197 show a winch and its cross-bar in various positions.

In the catapult plans I am describing, the dimensions of each winch are :

Large Wheel.  - 14 in. diameter across its top surface. Its bore (i.e. the aperture for the skein of cord), 8 in. diameter. Total length of the wheel, 8 in. Length of its flange that fits through the iron plate, 3 in. Thickness of the flange, 3/4 in.

Pinion Wheel. - 6 in. diameter. Its length, 4 in.

The projecting ends of the spindles of the pinion wheels are each 2 in. square and 5 in. long. On these ends heavy spanners are fitted for twisting up the skein of cord (see below).

Roman Catapult Model
Full Size Model of a Roman Catapult
Built by the Author
Weight 1 1/2 tons Range, with 6 lbs.
stone ball, 300 yards

1. Twisting up the skein of cord by means of the winches.
2. Winding down the catapult arm.
3. Releasing the arm when fully wound down.

Though this was the method of fixing the cross-bars adopted by the ancients, I have had my winches cast with their cross-bars solid with their wheels and not as separate pieces.

The wrought iron plates through which the flanges of the large wheels of the winches pass and on which the projecting rims of these wheels revolve, are each 1 in. thick. These plates are bolted to the sides of the catapult, figure 202.

The round shanks of the spindles of the pinion wheels (secured at their ends by washers and nuts), also pass through these plates as well as through the sides of the framework of the catapult, II, figure 197.

An almost inconceivable strain can be applied to the skein of cord by four or five men turning the winches of the catapult, a strain so immense that no arm of serviceable dimensions could be made to withstand the force that would have to be applied to wind it down.

Some medieval writers describe the devices formerly employed for reducing the friction created between the rims of the large wheels of the winches and the iron plates on which they revolve.

In the catapults I have made, I have not however found anything of the kind, such as ball bearings, necessary, other than plenty of grease inserted between frictional surfaces.

In the first catapult I made I fitted a skein of thick rope for the arm to work between, but I found it was impossible to put an even strain upon the rope when twisting it up with the winches.

The result of this uneven strain was, that the lengths of rope which formed the skein,  each 1 1/2 in. thick, broke one by one like rotten thread, owing to the force applied by the winches affecting them in detail instead of collectively.

After a series of experiments with various kinds of cordage, I discovered that the finer the cord used within reason, the more elastic and compact was the skein and hence the less its liability to break. The fracture of a few strands of a large skein of fine cord is of no consequence, but the breaking of one stout rope amid a skein of a dozen lengths of such rope, means a noticeable loss of power. The ancients were well aware of this and made the skeins of their catapults of thin cords of twisted hair.

In cases of emergency, woman's hair was made into skeins for catapults and balistas, and of all material nothing was so elastic or enduring for this purpose. When the inhabitants of Carthage commenced the heroic defense of their city (149-146 B.C.) they were forced to hurriedly manufacture weapons of all kinds to replace those which they had recently surrendered to the Roman general Censorinus. In various modern works we read of how the noble matrons of Carthage cut off their long tresses and twisted them into ropes for catapults.

At the siege of Salona by Marcus Octavius, one of Pompey's generals, the Roman women cut off their hair that it might be made into ropes for the catapults of the besieged.

I can find no authority for any such picturesque writing, as ancient authors simply record the fact ' that women's hair was used at Carthage. 'For instance, Florus, in his Roman History, a chronicler who flourished early in the second century, writes ' and the women parted with their hair to make cordage for the catapults. 'Zonaras, Byzantine historian, Chronica, ix. 26, says ' for the ropes of the catapults they used the hair of the women.'

If horsehair were not available in sufficient quantity, sinews from the necks of horses or oxen were used. I do not find that ordinary rope was ever employed to build a catapult skein. The elasticity of hair is so great, that however tight a large skein of it is twisted its extreme stretching or breaking limit cannot well be reached. for this reason, there is always sufficient life or spring in the most tightly twisted skein of horse hair to give the requisite velocity to the arm of the catapult.

It is evident that if the skein of a catapult were twisted up to its extreme limit, it would break under the further strain entailed on it by winding down the catapult arm.

After testing every kind of material for the skein of a catapult I find that horse-hair rope - 1/2 in. thick - is far the best. Failing horse-hair, pure flax in the form of sail maker's sewing twine is a fairly good substitute.If this twine is used for the skein of a catapult it should be spun into a cord 1/4 in. thick.

Insert a thin stick into the ground halfway between the winches. Place it upright inside the framework of the catapult. This stick will serve to keep the halves of the skein separate as it is being made, so that when it is completed the arm of the catapult may be placed in position without difficulty. Turn the winches till both their crossbars are perpendicular to the ground and in line with the stick.

Next secure one end of the cord you are using for the skein to the corner of the cross-bar of one of the winches.

Pass the other end of the cord through the holes in the sides of the catapult and round the cross-bar of the opposite winch, and then back again over the bar of the first winch. Do this in regular rotation to and fro, first on one side of the stick then on the other. Be careful not to cross the lengths of cord as you pass them between the winches, but keep them individually straight, tight and regular and alternately on either side of the stick, A, figure 199.

Do not wrap the cord at haphazard round the cross-bars of the winches, but lay the turns regularly from one end of each cross-bar to its other end and then back again.

When a complete layer of cord is wrapped over a cross-bar, place on it a strip of paper 1 in. wide. By concealing the last layer the paper will show you how to proceed with the next. The last few turns of the cord will have to be passed through the winches by the aid of a piece of stout wire with a loop at its end.

The stick may now be removed and the butt end of the arm placed between the halves of the skein. The skein should appear as in B, figure 199.

If the skein is formed of hemp or flax and not of horse hair, the material should be previously soaked in neat's foot oil. The oil will preserve the skein and save it from wear and tear; it will also make the skein into one solid mass, so that when it is twisted up by the winches its strands receive an equal strain.

When a skein is made of fine cord, it will be necessary to wrap this (in forty yard lengths) on a number of large netting needles, such as herring net makers use. It would be out of the question to pass and repass fine cord in one length through the winches. My largest catapult, for instance, required 1,400 yards of cord to make its skein.

When short lengths of fine cord are used, they will have to be knotted together as occasion requires during the process of making the skein.

After the skein is finished and the arm of the catapult has been placed in position therein, the former may be twisted (C, figure 199). For this purpose a heavy spanner, 6 ft. long, is necessary.

The eye of the spanner is fitted over the squared spindle (D, figure 193) of one of the winches. By means of the spanner, three or four men turn one winch slightly. They then remove the spanner and go round to the opposite side of the catapult and give the other winch a turn.

Numerals may be painted on the large wheels of the winches, so that it may be readily seen if the same number of revolutions are given to each wheel. This is important, as if one winch is turned more than the other the skein will be more tightly twisted on one side of the arm than it is on the other, and a Joss of power will ensue.

The winches should be employed to twist up the skein gradually, till it is impossible for three strong men (without the aid of the windlass) to pull the catapult arm back, even a quarter of an inch, from the top cross-bar against which it presses.

Three complete revolutions of the large wheel of each winch should be sufficient to create this amount of pressure. The winches are, of course, always turned in the same direction.

A stout iron slip hook was then attached to the rope that wound down the arm. The bend of the rope passed through the ring of the slip hook. The point of the slip hook was hitched inside the eye of the bolt and projected about 1 in. through it, figure 200.

By pulling the cord attached to the lever of the slip hook, the point of the latter instantly slipped out of the eye of the bolt and in this way released the catapult arm.

The point of the hook should be short and slightly tapered to its extremity, or it will not easily slip out when required to do so. For the same reason the point of the hook and the inside of the eye of the bolt should be smooth and round.

However great the strain on the slip hook it will, if properly made, easily effect the release of the catapult arm.

This simple method of releasing the arm of a catapult was far the best as the hook that pulled down the arm was also the means of setting it free. The slip hook was able to release the arm at any angle, whether it was fully (as in figure 200) or only partially wound down. The trajectory of the weapon was, therefore, controlled by this form of release, as the longer the distance the arm was pulled down the higher the angle at which the projectile was thrown.

On the other hand, the shorter the distance the arm was drawn back the lower the trajectory of its missile. If, for instance, a town was being bombarded by a catapult, the arm was wound down to its full extent of 90 degrees so that the stone it cast might strike the defenders on the ramparts, or else travel high over the defenses and fall upon the houses and people inside the walls.

If, however, the besiegers were threatened by a sortie from the gateway of a fortress, the arm of the catapult was set free at a point which was about a quarter less than its full sweep.Though the force of the missile projected by the catapult was then, less than when its arm was fully extended before it was released, the stone traveled low, and bounding along the surface of the ground was more likely to encounter an enemy advancing on horseback or on foot.

In this case the arm of the catapult was wound down to its full extent and could only be set free from this position; hence when the catapult was on level ground the trajectory of its stone did not vary.

To alter the trajectory of the stone thrown by a catapult of this description, the framework of the engine was elevated or depressed, figure 192. If it was desired to throw a stone at a low trajectory, the after end of the catapult was raised and wedges were inserted under the ends of its sides.

If a high trajectory was required, as when as it was wished to drop a stone into a town on an eminence, the front part of the catapult was propped up. Even in the case of the release described in figure 200, this was also necessary when a fortress was built on ground considerably above the level of the catapults attacking it.

(I) F. The end of the arm of the catapult as held from escaping by the projection B, of the hinged catch D, B.

By knocking down with a mallet the end of the lever A, the leg of the catch (D, B), is freed from the notch in A, at E.

(II) The catch (D, B), being then free to swing, the end (F) of the arm of the catapult is instantly released from the projection B, as seen in II, figure 201. This figure may also be taken to represent the catapult arm being wound down by the rope and roller.

When the arm is a little lower than shown in II (taking it as coming slowly down and not as flying up), then by lifting the handle (D) of the catch its projection (B) drops over the end (F) of the arm. The leg of the catch at the same time snaps into the notch of the lever A, at E. In this way the catch is reset and the arm again secured, as in I figure 201.

(III) Plan view of the  catch holding down the catapult arm.

(IV) Perspective view of the catch.

The iron framework of the catch was bolted to a cross-piece of wood which connected the aft ends of the sides of the catapult.

The roller that wound down the arm was fitted on the front side of this cross piece, as shown in the catapult in figure 198.

In figure 198  there is a grooved piece of wood (in the form of a shallow trough) on the top of the catapult. This trough is intended to hold the javelin, the projecting butt end of which is supposed to be struck by the released arm of the catapult. I do not believe the catapult was ever employed to project a javelin. It certainly could not do so in the manner here depicted, for the reason that the arm of the catapult could never strike a true blow. Besides this, the arm of the catapult casts a stone with a slinging motion and does not recoil with the quick snap of a spring, such as would be necessary to flip a javelin forward .

In addition, the winches for winding the skein of cordage are put in the weakest part of its framework in this catapult, i.e. between the uprights in stead of the sides where they should be.

The rope attached to the roller was hitched by a hook to a ring lashed to the catapult arm, figure 198. When the arm was safely secured by the catch, the rope that pulled it down was unhooked and the catapult was ready for action.

In some catapults, one end of the rope which pulled down the arm was spliced to a crossbar of metal fixed in the framework of the siege engine; its other end being fastened to the winding roller, figure 198. This arrangement, using mechanical advantage, halved the exertion required to pull down the arm and also halved the strain upon the roller, but it doubled the time occupied in winding back the catapult arm. By using longer levers for turning the roller, the same effect is produced as in the above method and without the loss of time it entailed.

When its skein of cord is tightly twisted, the catapult I have described will hurl a round stone weighing 10 lbs. to a distance of about 350 yards. Though this is a trivial range when compared with the result obtainable from a small mortar, it would be a more or less effective one in the days, for example, of the Crusades , in days when the besiegers camped within a quarter of a mile of the town they were attacking and even conversed with the defenders on its walls.

This catapult might easily be fitted with a pair of winches each larger by half than I have given in the plans. This would entail a stronger and slightly longer arm, and also heavier sides to the framework of the catapult. With these alterations, the catapult would cast a stone weighing 20 lbs.

The stones thrown by catapults do not increase in weight in proportion to the increase in diameter of the skeins of the catapult. For example, a catapult with a skein 1 ft. thick will throw a stone three times as heavy as will a catapult with a skein half the size, or 6 in. A skein of 1 ft. in thickness would, however, be double the length of the skein, which was only 6 in. in diameter, as in the former case the framework of the catapult would be much wider than in the latter one.

The great Roman catapult was about twice the size in length and breadth of the one I have given a detailed plan for. This immense and powerful machine had an arm of from 10 to 12 ft. long. A catapult of these dimensions ,according to the size of its skein , threw a stone of from 40 to 60 lbs. to a distance of from 350 to 400 yards, the most powerful catapult of the kind being probably able to attain a range of nearly 450 yards.

The velocity of the stone propelled by a catapult was very low as compared with that of a ball from a cannon. It was the ponderous nature of the projectile and not its velocity that did the execution. A stone of 50 lbs., falling from a short range on battlements and the tops of towers, or among crowded troops and lightly built houses, would be as destructive as a ball of half the weight fired from a cannon at a much longer distance than was possible with a catapult.

The damage to buildings and the slaughter of people must have been terrible, when we consider that 150 to 200 great catapults were often employed at the same time for pounding a city and its defenders, and further, that these engines could be used as freely on the darkest night as by daylight. Not only were heavy stones thrown among the besieged, their fortifications and their houses, but flaming projectiles were also used which set fire to everything combustible upon which they fell.

Each side of a large catapult was made of two huge logs of wood. The logs were squared and then placed one above the other and bolted together. Winches suitable for twisting a skein of cord such as a 10 to 12 ft. arm required - would necessitate timber of so great a size, that the ancients found it easier to construct the sides of their largest catapults of two longitudinal pieces.

The skein of cord for a catapult with an arm 12 ft. in length, was much larger in proportion to the size of the engine than was the case with a weapon that had a framework of half the dimensions. The catapult with an arm 10 to 12 ft. in length, also cast a stone three times as heavy as that thrown by a weapon half its size.

Read More >

Read More >

Read More >

Read More >

Read More >
 

Read More >

Read More >

Read More >

Read More >

Read More >

Read More >

Read More >