TEST OF IRON BY MAGNETISM. 



717 



of tho bar, us indicated l,y (ho two small ar- 

 rows there. On traversing tin- needle t'i\v:inl 

 .<! and :i quirk l-iaiid-fro motion is of 

 groat use near a sii-p.-ct, .,1 .sput it got gradu- 

 nlly debated at. A, until it entirely reversed at 

 (', indicating tin- presence of the flaw, shown 

 in tin- t'ul' ^eetion, nftcr cutting the 



originally larger Lar at that spot. I 



'viiicly interest I:ILT, as they all'ord 

 proofs of the groat accuracy of the method. 



i _ 



Figs. 6 and 7. 

 4 f> 6 7 8 9 10 11 12 18 14 





r* r 



q 2 K 



No. 61 



mtm 



No. 8 



.NO. is 



Bar shifted end for end. 



No. 6 



No. 5 [ JH-. | , ^ j 



No. 2 



No. 6 



E 



SO 50 43 36 |2 19 48 45 44- 5559 63 6a [S 



N - 5 n * : MW , r r\ s 



^ 1 1 J , 1 1 ! 3 ! 1 



^:^- * 



No. 2 " 



^HU__!_L- 



Bar shifted end for end. 

 No. 6 I -72-el^78 -7^6f+60t6747T+pTt74f754-7ci7^|9 



No. 8 



No. 8 



No. 2 



The actions of different small needles nnder 

 the influence of an imperfect bar, accurately 

 measured in degrees of a circle, are compared 

 together. No. 6 needle was 1.5 in. long; No. 

 5, 1.2 in. ; No. 4, 0.63 in. ; No. 3 was 0.54 in. ; 

 and No. 2, 0.34 in. long. The amounts of the 

 temporary disturbances are of course measured 

 by the number of degrees the needle was de- 

 viated. The bar of- iron was first tested in the 

 magnetic equatorial plane, and certain disturb- 

 ances of the needle showed themselves, indi- 



cating internal faults. On turning the ba: 

 for end. but still in the mai/m-ii.- equal 

 plain-, the same deviations occurred, pointing 

 to the same faulty spot-;. With the l,ar 

 in the magnetic meridian tin- n<-".|!- WB 

 larly deviated, though the indications thus 

 ohtained would evidently not be, so practically 

 clear and easily obtainabl-- a> when the bur 

 was in the magnetic equatorial plane. 



Mr. Saxby has been allowed to test his 

 method in various ways in the royal dockyards 

 of Sheerness and Chatham, in tli-- pre-ciice of 

 the master-smiths, the foremen of the testing- 

 houses, and several of the chief engineers of 

 the royal navy. Mr. Saxby, for instance, was 

 requested to find out the weakest spots in a 

 number of bars, and to tie a string or make a 

 chalk-mark on each spot. Immediately after- 

 ward all these bars were put into the testing- 

 machine and broken. Their history is given 

 below in the annexed cuts (Fig. 8), the predic- 

 tion having in every case been verified. The 

 bars are shown by lines to scale, and a scroll is 

 placed where the weakest part was found out 

 by the needle. The vertical dotted lines indi- 

 cate the spots where the several bars broke : 

 FIG. a 



Broke at 2$ tons. Off same bar ns D. Not annealed. 



The smiths of the royal dockyard seem to 

 have properly tried Mr. Saxby's powers in al- 

 most every possible way, and most ingenious 

 devices were sometimes resorted to for the 

 purpose. A. examples out of many, in the 

 centre of a bar (Fig. 9), of 1 in. square forged 



Fia. 9. 



iron, was welded a piece of unmagnetized steel 

 about 5 in. long. The needle detected a fault 

 at about the centre of the piece of steel. 

 In the \ in. round bar, 12 in. long (Fig. 10), 



Fio. 10. 



it was discovered by the oscillations of the 

 needle that the bar was flawed, and more es- 

 pecially at A. It was then gently cut, and the 



