i io8

MODERN STEAM PRACTICE.

ing to ceilings cannot be produced so as to dass with what is termed

pound work,” and engineers who have attempted to reduce theweight of line shafts and fittings, have generally had to contendwith rival estimates, in cases where the difference could not be fairlyexplained to those who purchase.

Two other things in connection with the arrangement of “over-head” and “underfloor” shafts, may be noticed—danger from fire,and floor space. In regard to the first, especially in factories wherefibrous material or wooden wares are manufactured, there is no mana-ger of experience who will not admit that a concealed pit shaft causesmore anxiety and calls for more circumspection than a whole factorybesides. Belts passing down through a floor carry with them, bythe air currents induced, much of the light dust arising frommachines, and generally, we may say unavoidably, shavings or otherd*$bris , so that shaft pits become in effect tinder boxes. A piece ofiron, a stone, sometimes even a nail, coming in contact with an ironpulley running at great speed, will cause a shower of sparks to bedischarged, and in an instant all is in Harnes. In respect to floorspace it is perhaps useless to say that belts coming down from over-head shafts to machines consume the least possible amount of use-ful room; besides, the arrangement of most factory machines, espe-cially machine tools for iron, renders overhead counter-shafts indispensable. Holes cut through floors to accommodate belts, with thesafety cases which must surround them, consume valuable room.

There are, no doubt, in many cases, especially where heavy shaftsare employed, sufificient reasons for mounting line shafts on earthfoundations, but such reasons are too well understood to requirediscussion. What has been explained thus far is preliminary tosome notice of couplings for line shafts.

Experience has shown that conditions other than a capacity foitorsional resistance should govern the size of line shafts and coupledjoints. The engineer who after careful calculation had reached atheoretical factor of safety for a line shaft, and then instructed hisdraughtsman. to “quadruple it,” was but following what ordinarypractice demands. Solid sleeves, to embrace both ends and Hangecouplings with clutch faces, have long been abandoned by Americanengineers, and they were succeeded by plain flange couplings, asshown in Fig. 768; the two halves being joined in some cases bytaper bolts, as drawn, in other cases by fitted parallel bolts, and'not unfrequently by loose bolts, depending for adjustment on a