(1) Incorrect velocity of propagation 



(2) Non-linearity of range circuits 



(3) Initial zero correction. 



A velocity of propagation of 299, 690 km per second at a sea level 

 pressure of 760 mm is assumed . The velocity is subject to change due to 

 changes in barometric pressure, humidity, etc. , but it is probable that the 

 error caused by these assumptions is not greater than 5 parts in 200,000. 



The non-linearity of the range circuits can be determined in the lab- 

 oratory. Corrections can be applied if necessary . Generally they do not 

 exceed 0.1 microsecond and can be neglected in hydrographic work if the scale 

 is smaller than 1 : 100,000. 



A small initial zero correction must be applied to all distance 

 measurements . This error is caused by the difficulty of making the vernier read 

 exactly zero to correspond to zero time . The error can be determined at the 

 will of the operator and applied or corrected at any time. Daily calibration is 

 necessary to correct for normal instrument drift. 



b. Random Errors 



The random errors are more serious, with no way of predicting their 

 magnitude. They occur both in the shipboard equipment and shore station equip- 

 ment and are caused mainly by human error. 



At the ship they are: 



(1) Misalignment of local and distance pulses 



(2) Careless reading of the distance verniers and dials 



(3) Poor alignment of pulses . 



At the ground station they are: 



(1) Poor synchronization of the time fix 



(2) Incorrect setting of the balance gain . 



The effect of all of these possible errors in the EPI system is to cause 

 an uncertainty in position which is directly related to the magnitude of the error 

 (algebraic summation) and the angle at which the distance arcs intersect. Ex- 

 perience has demonstrated that negligible error is introduced when the inter- 

 secting arcs are kept within the limits of 30° - 150° . 



114 



artbur ZD.Itittlc.Knr. 



