ON THE FRICTION OF METAL COILS. 541 



with the exception of flexible metal bands and wire rope, metal could be 

 employed in coil friction. In exactly similar circumstances this certainly 

 could not be done ; but suppose the conditions to be altered, and the 

 metal coil to be only used under circumstances which do not require it 

 to be unwound, as is done with rope, then the properties of coil friction 

 may also be taken "edvantage of if the coil is sufficiently flexible for the 

 purpose. It is true that metal is inferior in frictional resistance to rope, 

 but this advantage may easily be obviated by using a greater number of 

 coils, whereas the much higher tensional strength and durability of metal 

 point to many valuable applications. The chief applications, which have 

 been made with more or less success, have been in connection with 

 clutches and brakes, and the principle of operation is very simple. The 

 metal coil is either wound round a shaft or the sleeve of a pulley, or else 

 it is contained within a cylinder attached to either. One end, which will 

 always be referred to in the present paper as the ' tail,' is by some means 

 or other brought into frictional contact with the shaft, sleeve, or cylinder^ 

 and is thus carried round if the surface be in motion, or retarded if the 

 coil itself be in motion and the surface at rest. The attachment of the 

 ' head ' of the coil prevents its following the tail until a considerable 

 tension is put upon the whole coil. Thus, if the coil encloses the shaft or 

 sleeve, it is made to wind up upon the shaft, becoming of less internal 

 diameter and taking a frictional grip throughout its whole length ; but 

 if, on the contrary, it is enclosed in a cylinder, it is made to unwind, and 

 so expand. In either case the result is the same, and a force of any 

 required magnitude may be transmitted by this means. In order to 

 illustrate this action, a piece of apparatus has been devised by the 

 authors which may interest those who have never seen any example of 

 the power of metal coil fi'iction. A weight of 56 lbs. is raised by a drum 

 and handle. The shaft — 1 in. in diameter — to which the drum is attached, 

 passes through its bearing in the frame, and is attached to the head of a 

 metal coil of iron wire ^ in. in diameter. The coil encloses a sleeve 

 carried by the frame, and the tail of the coil is free. As long as this is 

 the case the whole weight of the 56 lbs. has to be sustained by means of 

 the handle ; but if a weight of 10 oz. is now suspended to the tail, a 

 grip is obtained throughout the coil, by means of which the weight is 

 sustained, which from the difference of leverage is equivalent to a force 

 at the head of the coil of 200 lbs. The load has in this way been in- 

 creased at a previous trial until the head of the coil was torn off without 

 adding to the weight at the tail. If the small weight is raised, the load 

 falls ; but its fall is instantly checked by releasing the small weight 

 again. To show the efiect of the number of convolutions various coils 

 are substituted, and it is seen that when there are only four coils instead 

 of eight, an increased weight, many times as great, is required to sustain 

 the load. The effect of simply twisting the tail of the coil is very 

 remarkable, as may be ascertained by means of the small wire coils now 

 passed around the room. In this case it is not necessary to hang any 

 weight at all upon the tail itself; and the self-sustaining action of the 

 coil, which permits perfectly unconstrained motion in one direction but 

 instantly checks any motion in the other, is very striking. This effect is 

 still more strikingly exhibited by means of the experimental apparatus. 

 The tail of the small coil is twisted so as to just touch the shaft with a 

 slight pressure. The load is then raised, and no resistance whatever is 

 felt from the coil as long as the handle is being turned in the correspond- 



