720 



TEST OF IRON BY MAGNETISM. 



the flaw was found and cut out in the progress 

 of the work. 



ITio. 22. 



In another case, in which Mr. Saxby's ex- 

 periments were carried out in the presence of 

 a large number of naval chief-engineers, he 

 put down in writing the results of his magnetic 

 examinations, in order that they might be sub- 

 sequently compared with what was known as 

 to the actual quality of each bar. A bar, 1 

 in. round and 3 ft. 11 in. long, was pronounced 

 by the compass-needle as being not of the 

 same iron throughout, and with a south end bet- 

 ter than the other. It was then stated by the 

 master-smith to have been made up of pieces 

 good and bad. A rather shorter bar was found 

 to be good iron, but doubtful in condition ; it 

 was afterward explained to be "uncertain," 

 and on testing it in the machine it was stated 

 to be "crystallized." A third piece was found 

 to be of very good iron, but with slight irregu- 

 larities ; the smiths stated it to be scrap iron, 

 and the best to be got in the shop. Two pieces 

 of f in. manufactured iron were discovered to 

 be not good. Another piece of .1^ in. bar was 

 found to be good iron, though made of differ- 

 ent qualities it had been afterward annealed. 

 "With another bar, to Mr. Saxby's written ques- 

 tion whether it was not steel, it was answered 

 that the bar in question was a near approach 

 to steel, being a piece of galvanized wire-rope 

 welded up. To the remark that another bar 

 was unfit for use, he was told that it had been 

 twisted round when at a low heat, and then 

 hammered cold. Some singular proofs of the 

 power of magnetic testing over the ordinary 

 methods of determining quality and condition 

 of iron have been shown. Pieces of iron 

 brought for testing by most able and experi- 

 enced master-smiths, of such quality as would 

 be selected for the most important work, have, 

 on being tested, been marked at spots as de- 

 fective, and on cutting have accordingly been 

 found at those spots to be partially fibrous, 

 partially crystallized. 



The following experiment was made in order 

 to throw light on an important practical ques- 

 tion in smith's work: A certain If- in. round 

 bar 17-J in. long was specially worked, and had 

 been brought to be tested without anything of 

 its history being, known to Mr. Saxby. He 

 found that in the middle of its length it was 

 seriously faulty, and even unfit for use. He 

 was then told that the bar, though solid, had 

 been "upset" in the middle of its length, and 

 then hammered down to its original diameter 

 at a temperature below welding heat. This 

 will be held to confirm the opinion of good 



workmen that "upsetting" should be done at 

 a temperature as near as possible below that 

 of welding. 



Mr. Saxby has not yet been successful in 

 testing rolled plates for lamination. In these, 

 again, the neutral or zero lines should run at 

 right angles to the dip in a homogeneous plate; 

 but the more complex structure of the plates 

 has made the investigation more difficult. An- 

 other difficulty doubtless consists in the fact 

 that the usual shape of a plate does not allow 

 the magnetism to separate itself in such a 

 marked way as in a bar, usually longer by many 

 diameters. The investigation, with a resulting 

 perfect method, can scarcely be said to be 

 completed in this direction. The chief diffi- 

 culty at present seems to be that the internal 

 structure is too irregular. 



Up to the present but few experiments have 

 been made with steel, and very few with cast 

 iron; those already made have, however, been 

 satisfactory. Any difficulty that might be sup- 

 posed to 'attend the presence in wrought-iron 

 of what is termed by the Astronomer Royal 

 subpermanent magnetism is easily overcome. 

 A few taps on the end of a bar of wrought 

 iron, when lying east and west, sufficient to 

 cause vibration, w'ould demagnetize it, and 

 leave it in a fit state to be examined by the 

 needle; any polarity subsequently found would 

 indicate either a steely nature of the bar or 

 inferior iron. As showing the necessity for 

 caution as to whether or not bars being tested 

 are permanently magnetized, may be cited the 

 following experiment: Three bars were pre- 

 sented to Mr. Saxby to be tried. He found 

 that all three were permanently magnetic, 

 using after the trial with the pocket-needle 

 percussion to determine this point. He at first 

 declared that all the three showed the same 

 behavior as to the magnetic test. He was then 

 informed that one bar had been made up of 

 " scrap galvanized iron," or of old pieces of 

 zinc, galvanized plates, and wire-rope. The 

 zinc is not, as might at first sight have been 

 expected, sublimated by the heat, and accord- 

 ingly the iron produced is of about the worst 

 possible character. The second bar consisted 

 of the best Bowling iron, and the third was of 

 the best Chatham make. But Mr. Saxby was 

 soon enabled to trace out the fact that the 

 three bars had not merely been lying side by 

 side for some five hours previous to the test, 

 but that they had all been kept on racks lying 

 nearly in the magnetic meridian for above a 

 year. A similar but new Bowling bar from 

 the same rack as the bar tested was found to 

 be strongly magnetic; and another similar 

 zincified bar was partially magnetic. The bar 

 made at Chatham was found to behave toward 

 the needle as a piece of good iron should. Mr. 

 Saxby hence concluded that the Bowling bar, 

 which, indeed, had much more the character 

 of steel than of iron, had magnetized the two 

 others. Tested as a steel magnet, it was found 

 free from fault. 



