170 I>KFU:< To.METERS. 



which is beiug tested upon two steel bars, fitting at the 

 two ends into the brackets which carry the revolving 

 supports. During a test the deflections may be either 

 measured directly from this plate to the under-suri'ace of 

 the beam by means of a pair of ordinary calipers, an inside 

 micrometer caliper, or a separate multiplying detiectometer, 

 like the one shown in Fig. 85.* This instrument is simply 

 a lever of large multiplying ratio clamped to the plate, and 

 having the short end of its lever resting against a point on 

 the underside of the beam. The end of the long arm 

 forms a pointer, which is arranged so as to move over a 

 fixed vertical scale, and the deflections of the beam are the 

 readings on the scale multiplied by the velocity ratio of 

 the lever. This arrangement is found to work very satis- 

 factorily, in spite of the fact that it is in principle faulty 

 in one or two respects as, for instance, the support of the 

 measuring lever is not carried by the beam itself. The 

 error introduced from this cause is very small, because 

 the heavy design of the supporting brackets and the large 

 bearing surface upon which the beam rests render it 

 difficult for any appreciable yield to take place. ~ 



For given loads, the deflections are relatively great as 

 compared with the extensions of a tension piece for the 

 same loads, and therefore the measuring gear need not be 

 of quite so delicate a character, especially for large beams. 

 In the multiplying gear that has just been described, a 

 lever ratio of 10 to 1 is found to work well for beams of oft. 

 span or upwards, but a larger ratio may with advantage be 

 employed. 



FIG. 86. t 



For small beams a gear similar to that shown diagram- 

 matically on Fig. 86 may be used. In this arrangement, 

 A is a light frame carried by the specimen S itself, which 

 it clips through two pairs of screws C C. At the centre of 



*See Secundo on Beams, Proc. Inst. C.E., Vol. XCVIII., p. 32; 

 also Carpenter's " Experimental Engineering," p. 115. 



t This is similar in principle to the instrument used by Professor 

 Goodman at the Yorkshire College. 



