48 



NATURE 



[March ii, 1920 



the movements of the ship would react through the 

 friction and cause errors in the reading. 



Anschutz, in his early form of compass, by use of 

 an air blast, gets rid of this connection with the ship. 

 The air blast was arranged to oppose the movement 

 in azimuth when the wheel tilted, and thus he ob- 

 tained an effective method of damping. The strength 

 of the air blast, which varies proportionally to the 

 tilt, should be nothing when the compass is at rest 

 on the north — that is, when the tilt is nothing — and 

 this would be true with the compass on the equator. 



In other latitudes, however, the compass rests at 

 the north with a tilt still remaining. It does not 

 come back to the horizontal position because the axis 

 of the wheel is trying to set itself parallel to that of 

 the earth. This leaves a residual air blast continuously 

 acting, producing a permanent twist in azimuth and 

 a constant error. It is, therefore, preferable to damp 

 the swings of the compass by acting upon the tilt 

 rather than upon its movement in azimuth, because in 

 this case there will be no latitude error. The tilt is a 

 maximum at the middle of each swing — that is, when 

 it is moving throuf^h the north position — and it is 

 the return of the weight to its trulv vertical position 

 that is resnonsible for the continuation of the oscilla- 

 tion ; we therefore require some method of neutralis- 

 ing? the action of the weight, not before, but after, 

 the compass has reached the north. This I accom- 

 plish in the Brown gyro-compass by automatically 

 moving a liquid from one bottle to another, and in 

 such a direction as to counterbalance the weight, pro- 

 cessing the gvro-wheel, and I delav its action by 

 means of a valve or constriction in the tube joining 

 the two bottles. 



The force with which the compass seeks the north 

 is proportional to the product of the rotation (one 

 revolution in twenty-four hours) and the spin of the 

 wheel. The faster we can spin the wheel, the more do 

 we obtain directive force. It is for this reason that 

 the wheel is rotated at its maximum speed and 

 strength consistent with the rise of temperature. 



Taking the Brown gyro-compass as an example, 

 the wheel, which is 4 in. in diameter and 4J lb. in 

 weight, runs at 15,000 revolutions per minute. The 

 maximum directive force of the earth on this wheel — 

 that is, when the spindle is pointing east to west— is 

 onlv the wei.<^ht of 30 grains, with a leverage of i in. 

 This small force is continually diminishing in value 

 as the axis approaches the north direction, and 

 vanishes absolutely in that position. If the comoass 

 was deflected, say, 1° from the north, then the force 

 of restoration is only \ grain at a leverage of t in. 

 It will therefore be seen how important it is to 

 eliminate as completely as possible any friction on 

 the vertical axis that would tend to oppose the direc- 

 tive action of the earth. 



There are three forms of gyro-compass now in use : 

 the Anschutz CGerman), the Sperry (.'Kmerican), and 

 the Brown (British). In the Anschutz the vertical 

 axis is supported by a bath of mercury, and in the 

 Sperry by a suspended wire, the twist, if any, being 

 taken out by a follow-up motor through an electric 

 contact, which switches on the current to the motor ; 

 while the Brown is operated by a hydraulic system 

 of support. The lower end of the vertical spindle acts 

 as a ram and stands upon a column of oil. The oil 

 is under great pressure, some .i;oo lb. per square inch, 

 and is kept pumning up and down, and thus raising 

 and lowering the vertical axis continually some 

 i8n times everv minute. 



The continual movement of the spindle results in 

 a practicallv frictionless vertical support, so that the 

 total movin?^ nart. some '^\ lb. in weight, can be 

 carried round in azimuth by the smallest force, due to 

 the earth's rotat'on : in fact, -^o small is the friction 

 NO. 2628, VOL. 105] 



that the compass, if deflected, will always come 

 back again to its true north position, certainly within 

 one-tenth of a degree. I think I am safe in saying 

 that it is the most perfect frictionless support yet 

 given to the vertical spindle of any gyro-compass, 

 or, indeed, of any machine. 



In an earlier part of this lecture it was stated 

 that the period of oscillation given to a gyro-compass 

 is of the order of 85'. I will now try to explain 

 why this is so. The earth has no angular movement 

 from south to north, but has one from west to east, 

 due to the daily revolution on its axis. A ship, how- 

 ever, sailing to the north at, say, twenty knots an hour 

 introduces an angular movement in that direction 

 because it is moving over the curved surface of the 

 ocean, and would complete a revolution of the globe 

 in forty-five days. 



If there were a gyro-compass on the ship the 

 instrument would be sensible of these angular 

 movements, set itself so as to make a com- 

 promise between them, and, as a consequence, point, 

 not to the true north, but one or more degrees west 

 of the actual pole. This division is termed the " north 

 steaming error." Knowing the latitude, the speed 

 of the ship, and its direction towards the north and 

 south, the extent of the error can be accurately cal- 

 culated, and speed-correction tables have been pre- 

 pared so that this error can be determined for any 

 latitude, speed, and heading of the ship, and can be 

 allowed for. 



Automatic means have also been devised to make 

 these necessary corrections in the reading of the 

 compass. For instance, my special form of repeater 

 has been designed so that the card can be set 

 eccentric, and. when once set, the correction will 

 be automatically applied without any further reference 

 to the tables. 



When a ship is in harbour a gyro-compass On board 

 points due north, but when the ship starts steaming 

 to the north the compass begins an oscillation so as 

 to bring the axis of the wheel into the new resting 

 position to include the north steaming error 

 in the reading. Getting up speed will, however, 

 have another effect on the compass. We know that 

 the gyro-wheel is acted upon by a pendulous weight, 

 .^s the ship changes its speed the acceleration will 

 act upon the pendulous weight and cause an oscilla- 

 tion to be started. This oscillation is termed the 

 "ballistic deflection." 



The permanent north steaming error and the transi- 

 torv error due to the ballistic deflection are in the 

 same direction, and mathematicians have calculated 

 that with an undamped gvro-compass, if the time of 

 its oscillation is set to 85' in any particular latitude, 

 the ballistic deflection can be made exactly the same 

 as the deflection due to the north steaming^ error; 

 this being- so, the compass should move into its new 

 resting-place without further oscillation. This would 

 be true if, as before indicated, the compass were un- 

 damped in its swings, but the mathematicians have 

 overlooked, the fact that all gyro-compasses are 

 damped, and the ballistic deflection must, therefore, 

 include a term due to the damping. 



This dampinff term up to the present has been 

 neglected, but in practice it is found that when a 

 ship is steaming and turning to alter its course the 

 compass does not come dead-beat to its new position, 

 but has an oscillation started which is common to 

 all existing gyro-compasses. The extent of this 

 oscillation mav be termed the "damping error. " On 

 a merchant ship the damping error is of little moment, 

 but on a war vessel which is manoeuvring it may 

 be serious, as it mav swing the compass off its cor- 

 rect reading by several degrees. 



{To "he continued.) 



