ON THE SMALL SCREW GAUGE* 439 



produced over the surfaces aa, a a, . . . , and the compression produced 

 there by the act of screwing up relieves any pressure over the surfaces 

 hb, bb, . . . , Contact and pi-essure at the points c c c . . . depend on 

 the relative diameters of the screw and the tapped hole. The spaces 

 shown in the figure along the surfaces b b, b b, . . . ai-e of course greater 

 than would occur in a well-titted screw. Now if the thread may be looked 

 upon merely as a means of supporting the tensional strain on the bolt, 

 without offering much frictional resistance to screwing up, it is clear that 

 this will be most efficiently done if the pressure is evenly supported over 

 the whole of the working surface of the thread a a, a a, . . ., and within 

 the assigned dimensions of the thread this surface should be as large as 

 possible. Contact and pressure at the points c, c, . . . depending on the 

 respective diameters of the screw and the tapped hole may interfere with 

 the fair contact of the working surfaces, involve extra resistance to screw- 

 ing up, and, so far as the support of the tensional strain is concerned, serve 

 no useful purpose. The best design for the thread, in view of its function 

 of supporting the tension, is that which secures most perfectly a continuous 

 working contact over the surface a a, a a, . . ., and freedom frooi pressure 

 at other points. These conditions are met best by a thread having straight 

 sides, a flat top, and a clearance space at the top and bottom of the thread 

 such as is shown in fig. 2 (p. 441). The provision of straight sides gives 

 a form to the originating tool which can be j^roduced with more ease and 

 accuracy than one of a curved form, and assists to secure correspondence 

 between the surfaces of the sci'ew and nut : the provision of a flat top gives 

 the largest possible area to the working surface within the given limits of 

 the thread : the provision of a clearance space at top and bottom removes 

 the possibility of any interference with the fit of the working surface by 

 irregularities of form at those points, and avoids unnecessary friction. 

 Screws with straight sides and flat tops are perfectly satisfactory in in- 

 strument practice, are employed in France and Germany for the most 

 important engineering work, and are uni'v'ersal in America for work of all 

 kinds, for instrument work as well as heavy engineering work. We 

 understand that the provision of clearance is well recognised in the practice 

 of American and French engineers, who use the Sellers thread, and Mr. 

 Hewitt, at Prescot, gives a very liberal clearance in the screws manu- 

 factured by him. The ease with which such threads are originated is a 

 point in their favour, though it would be of small importance if it were 

 shown that the thread is practically defective in other ways. 



As regards the reduction of the sectional area of the core by the pro- 

 posed deepening of the thread, the figures tibtained by Messrs. Gorham and 

 Price, corroborated by common experience, show that screws give way 

 under tension by breaking across the core rather than by stripping their 

 threads or those of the nuts ; and it has been urged against the proposal to 

 deepen the thread that it weakens the screw in its already weakest part. 

 The reply to this is that the strength of the screw is really determined by 

 the strength of the core, and that the British Association series is so 

 closely spaced that a screw can always be found whose core is of the 

 required size. Moreover, in well-designed work, screws have so large a 

 factor of safety that a reduction of the section of the core by an amount 

 varying from 8 per cent, in large screws to 12 per cent, in small screws 

 will not generally be a matter of great importance, though it will be 

 remembered tliat the resistance to torsional fracture varies inversely as 

 the square of the sectional area. 



