298 ELECTRIC PROi*ULSiON OE MERCHANT SHIPS. 



the problems of the gyro-compass I have often sent my staff back on the job with the injunc- 

 tion that the real cause of the difficulty is as big as a barn, and with a little more patience 

 there should be little trouble in locating it. 



Further, the case is cited where the loading of the teeth is four times and still working 

 perfectly successfully. Mr. Emmet seems to think that there is something about ship con- 

 ditions that applies. I am inclined to think he is right, but I do not believe that a gradual 

 application and withdrawal of 75 per cent overload, as is shown in Plate 120, will go far in 

 explanation. The seven seconds marked is simply the natural period of pitch of the ship; 

 but I do think that Plate 120 may give a slight clue to one possible source of the real trou- 

 ble in the minute periods, which can be very clearly seen on close scrutiny; there being eas- 

 ily traceable a high period sub-vibration of something on the order of five per second, which 

 harmonizes with the number of revolutions per minute multiplied by four, which is in all 

 probability the number of blades upon the screw at the end of this shaft. 



Critical periods are likely tO' exist where little suspected in machinery, especially where 

 this machinery is subjected to rhythmic or periodic oscillation of any kind. The latest Ger- 

 man submarine Diesels sent over here have two such criticals. These are very serious. The 

 engines in all submarines also have periods, one case notably occurring exactly at the normal 

 running speed. In studying these periods and their causes and finding means for their en- 

 tire suppression and elimination, we have learned much concerning this whole phenomenon. 

 Such familiarity as I have with this problem leads me to believe that here is another occa- 

 sion where criticals are on our track and are the prime cause of all the trouble. However, one 

 might say we have no reciprocating engine connected up in this instance, but a turbine in- 

 stead. Where, therefore, is the source of oscillation? It is true that in the case of the sub- 

 marines, where the reciprocating engine speeds are only 350, it is easier to study these crit- 

 icals than at the higher speeds of the steel -turbine, but in these geared sets in ships there 

 are always two sources of periodical oscillation, either one of which, or a combination of the 

 two, may, as we know, constitute all the cause that is necessary to set up a true critical. The- 

 oretically a critical of large magnitude can be set up and maintained by an almost infinitesi- 

 mally small source of disturbance when this disturbance occurs in proper period. 



Now in every ship's gear installation we have two sources of such periods : — First, the 

 unequal action of each successive blade, as it comes down into stiffer water, or as it passes 

 the stern post eddies when in stiffer water, leading or lagging pitch of a blade, or through 

 any number of possible combinations of disturbances, can easily set up sufficiently periodic 

 oscillation to become the source of criticals; second, the gear teeth impact themselves, but we 

 say these are helical and operate with perfect smoothness. This statement, however, for the 

 present purposes, is nothing less than nonsense, because, as stated, an infinitely small source is 

 all that is necessary ; let these gears be cut with the greatest accuracy possible, they still have 

 sufficient inaccuracies to perform the full function of setting up a critical. 



When criticals exist, it means that certain mass moments are not only present but are 

 so located and distributed in the whole moment system that a proper inter-relation exists. 

 My conception is that such a distribution doubtless exists and can be fotmd and analyzed 

 completely upon close study of the situation. The enormous over-stresses and general havoc 

 provoked and maintained by criticals correspond almost exactly to the phenomenon that has 

 been observed and that we are here considering, which causes disturbances that have no 

 limit in creating unsuspected stresses and pressures that are repeated with a frequency and 



