(380) As this subject is one of some interest a few words may be profitably expended on it. There is some confusion existing as to the way in which the loss in twist should be arrived at. On the one hand it is contended that the number of coils made in one lift of the ring rail should be counted, and the loss of twist calculated from that. On the other hand, it is urged that the coils laid in a double lift of the rail should be taken as a basis. This is the view held by Mr. Charles Lancaster, of Manchester, who has given a good deal of attention to the subject, and with whom the author is inclined to agree. To arrive at a conclusion it is necessary to ascertain the smallest and largest diameter of the surface of the bobbin, calculating therefrom their respective circumferences. The mean of the latter will give the average length of yarn wound per revolution, and the difference of the two the relative loss in twist. Mr. Lancaster put this matter very clearly, and the demonstration may be given in his own words: “To prove this calculation measure the length of a ‘draw’—_i.e._, the yarn deposited in one up or one down motion of the ring rail—and multiply by the number of turns per inch, count the number of coils in this layer of yarn (which represents the actual loss), and divide into total number of turns. Thus, if the up motion of the ring rail deposits 72 inches of 20’s yarn with 16·75 calculated turns per inch, then 72 x 16·75 = 1,206 ÷ 20 coils = 1·8 per cent of loss; and if the down motion deposits 178 inches of yarn with 16·75 calculated turns, then 178 x 16·75 = 2,981·5 ÷ 46 coils = 1·6 per cent, or an average of one up and one down motion of the ring rail of 1·7 per cent.”[A] Of course the finer the yarn spun the less the percentage of loss of twist.
(381) Whatever be the mode of calculation adopted, whether only the single or double lift be taken into account, the fact remains that there is a loss of twist, and that this is of most account when weft yarns are being spun. It being most desirable to spin these upon the bare spindle, so that they may be used in the shuttles, it will be seen that the subject is one of some importance. Various methods have been tried to overcome the difficulty, one mode being to take the yarn away from the nose of the cop as quickly as possible, but it has only been partially successful. Another, and more successful plan, is to form a special traveller, so arranged that the yarn in passing to the bobbin does not give a radial but a tangential pull to the traveller. The final form of traveller adopted by Mr. William Lancaster, who has tried a large number of shapes for this purpose, is arranged in this way, and to a certain extent frames made by him have been successfully used. The mode of construction adopted by Mr. Lancaster is shown in Figs. 198 and 199. In these the traveller is made at one end with an open fork, and at the other is =C= shaped, fitting the ring as shown in Fig. 199. Referring to that figure it will be seen that the yarn is carried through the =C= shaped eye, and then round the foot of the open fork =B= to the spindle =A=. The result is that the point where the traveller rests on the spindle acts as a fulcrum, and the yarn exercises a pull on the end of the arm carrying the =C=, this portion of the traveller practically becoming a lever. In this way the direct radial pull upon the yarn is avoided, and the traveller is readily drawn round the ring. Its actual position on the cop is shown in Fig. 198, where =B= is the fork, =A= the yarn, and =C= the spindle. In this form a number of frames are working with considerable success.