PART THE FIFTH.

ON UNIVERSAL JOINTS.

CHAPTER I.

HISTORY AND APPLICATIONS.

507. DR. JOHNSON gives seven different definitions of the word joint,' of which the second is that which is applicable to mechanism; namely, Hinge; junctures which admit motion of the parts; or rather, of parts that are connected.

But in scientific language I prefer to employ the term 'lines of flexure' for hinge-joints. Such joints were termed in Old English, gimmals or gimbals. The derivation of these words is doubtless from the French gémeaux (gemella, Lat.), twins; which is applied properly not only to a hinge composed of two portions of exactly similar form and size jointed together, but to anything else which is formed of twin pieces of like dimensions united in any manner.'

The contrivances which bear the name of universal joints' have been employed for two different purposes.

First, to connect any object, such as a lamp, mariner's compass, chronometer, or wheel-carriage, with its base or support, in such a manner that when the support is moved into different angular positions the object shall remain parallel to its normal position. The connection must have the property of compelling the two parts, object and base, to preserve one constant point in common, about which their relative motions are performed.

Secondly. As a point of flexure to connect in such a manner, two rods or shafts whose directions meet in a constant point, as to enable one to bend or incline in all aspects with respect to the

* Vide Promptorium, p. 194, by A. Way; published by the Camden Society. London, 1843.

other about this intersection point, but so that a rotation of the one rod about its length, will compel the other also to rotate about its own length.

For the first purpose this contrivance has been used from remote antiquity. In the manuscript sketch-book of Wilars de Honecort, an architect of the thirteenth century, there is a drawing, and coeval explanatory inscription. Fig. 356 is a reduced fac-simile of the drawing, accompanied by a literal translation of the inscription.†

If you desire to make a chauferette (calefactorium) or handwarmer, you must construct a kind of apple of brass in two halves which fit together, inside the apple place six brazen circles, let each circle have two pivots, and in the middle place a little brazier with two pivots. The pivots must be placed in contrary directions, so

Fig. 356.

that in all positions the brasier may remain upright, for every circle supports the pivots of the next. If you make this contrivance exactly as the description and drawing shews it, you may turn it about in any way, and the cinders will never fall out. It is excellent for a bishop, for he may boldly assist at high mass, and as long as he holds it in his hands they will be kept warm so long as the fire remains alight. This machine requires no farther explanation.'

* Published at Paris in fac-simile in 1858, with notes, by Lassus, and afterwards by myself in 1859 in London, with many additional comments.

+ Plate xvi. p. 54.

It must be remarked that one intermediate ring between the object which is to be kept in a horizontal position, and the outer case which may be inclined in any direction, is sufficient, as will be seen in the next chapter.

The last purpose of this class to which the gimbals were employed was in the fifteenth century, for the construction of wheelcarriages that when overturned would nevertheless preserve the body of the vehicle and its occupants in their level position without injury.

It is remarkable that when Hooke was appointed Curator of the Royal Society in 1662, and engaged to supply at every meeting three or four of his own experiments, one of the first things was a Chinese cart with one wheel; and in the next year he showed a scheme of an engine or carriage which goes on one wheel and with one horse, and will not fall but be kept perpendicular, even on the declivity of a hill. In one of these carriages a man was once overturned, and as he afterwards related, I knew it not till I lookt up and saw the wheel flat over my head.'

Hooke continued his projects for carriages of this kind in succeeding years, and took a patent for several new-fashioned chariots August 31, 1664.

We may now consider the history of the universal joint in its application to the connection of rods whose directions meet in a point.*

The earliest representation of this connection is to be found in the Technica Curiosa' of the Jesuit Schottus, published in 1664, where we find in plate vii. the drawing which I have given in fac-simile (fig. 357). It occurs in the ninth book, entitled Mirabilia Chronometrica,' which, as Schottus informs us in the preface, is composed of extracts from an unpublished manuscript entitled Chronometria Mechanica Nova,' the work of a writer whom he terms Amicus,† and in a manner which intimates that he was no longer alive.

The universal joint, therefore, as represented in the engraved figure, might have been invented many years before the publi

*The universal joint is attributed by French writers to Cardan, who lived in the sixteenth century (1501-1575). But the only trace of such a machine that I have been able to discover in his voluminous works (10 vols. folio), is a diagram of three hoops joined to each other in succession by diametral axes, as above described. Cardan tell us, that he saw it in the house of a friend, and is unable to assign a use for it. It appears to me to be a portion of a rolling lamp or brasier, like that of W. de Honecort. (Op. Cardani, t. x. p. 488. De Armillarum Instrumento.)

† Pp. 618, 727.

cation of the Technica Curiosa,' and not necessarily by this 'Amicus,' whose account of it I subjoin.

In clock towers it frequently happens that from want of room it is impossible to place the dial in the same part of the building as the wheel-work of the clock. Therefore the motion must be communicated from the mechanism in various directions obliquely, upwards, downwards, or sideways, to the hands of the dial plate. 6 This can be effected by means of axes provided with conical (bevel) wheels, but much more simply by the following wonderful device "paradoxum." It must be premised that there is a wellknown combination of concentric rings, each one connected to the next by pivots, which is employed for the construction of lamps which can be rolled upon a plane surface without spilling the oil. Our paradox is not unlike this, but is rather its twin brother, as the subsequent description and figure will show.'

Fig. 357.

OH

ABCD is a cross consisting of four arms united by a small ball or any convenient connection. The opposite arms AB, are inserted in holes at the ends of the prongs of a fork (fuscinula) ABF. The other pair of arms CD are similarly received by the fork CDH, and the forks are supported by fixed rings, GE.

E

B

If the extremity H of one fork receive rotation from the clock, the other fork, by virtue of the connection described will necessarily revolve with exactly the same velocity. Therefore if one fork receives a uniform circular motion, the other fork will also be compelled to rotate uniformly.* I must here observe that this unfortunate remark, which the next chapter will show to be quite contrary to the truth, proves that the writer had not accurately examined the laws of the combination.

But he states truly that the prongs of one fork will not strike against those of the other, if the angle made by the axes is greater than a right angle.'

Our author also adds that a series of axes may be arranged, each connected to the next by a joint of this kind. These axes may be disposed in a zigzag course, or along the sides of a polygon, or gradually rising in a spiral form against the sides of a

*

necesse est sequatur et altera fuscinula parique cum priore illa feratur velocitate; unde si fuerit unius fuscinulæ motus regularis circularis, erit similis et alterius ac omnium quotquot artificio simili connexarum.

polygonal prism so as to convey motion from the bottom to the top of a tower. They may also be disposed in an endless chain."

Three years after this publication of the joint of Amicus, namely, March 14, 1667, at the meeting of the Royal Society, Mr. Hooke produced a contrivance to make a motion of a clock Fig. 358.

to go along with the shadow on a wall, for which he offered a demonstration; affirming that the same instrument would be applicable to all planes, to make all sorts of dials, &c. He had previously announced (in November 1663) that he was occupied with the contrivance of a machine to describe all kind of dials, and he