[ 199 ] 



XXV. On a Method of Testing Submarine Telegraph Cables during 

 Paying-out. By Thomas T. P. Bruce Warren, Electrician 

 to Hooper's Telegraph Works } Limited; Member of the Society 

 of Telegraph Engineers *. 



IT is a singular circumstance that although within the last 

 few years we have become, as it were, inundated with new 

 appliances for testing submarine telegraph-cables during their 

 manufacture, so little has been effected towards the improve- 

 ment of electrical testing during the paying-out. 



Apart from the uncertainty which has lately been shown to 

 exist in galvanometric measurements themselves, there are many 

 little difficulties to encounter when testing on board ship, which 

 at times are so embarrassing as to make one forego a very large 

 share of confidence in the results. The electrical condition of 

 the cable must consequently be a matter of great anxiety, until 

 a steady observation enables the electrician to decide that every 

 thing is all right. 



The practice has hitherto been to make the duties at the 

 shore station too subsidiary a matter in the system of testings 

 instead of relying upon the results obtained on shore as of 

 primary importance. Considering the facilities offered when 

 the instruments are perfectly at rest and consequently admit of 

 much more accurate adjustment and indication than is possible 

 on the ship, it is evident that we should look to the shore 

 observations with an equal, if not greater interest. 



In the present systems of testing, the shore operations have 

 been usually so formulated that any separate or independent 

 observation is quite inadmissible, and, unless distinctly and 

 definitely preconcerted by the ship, would lead to serious results. 



Any system of testing on the ship should be capable of being 

 synchronously carried on at the shore station without impeding 

 any modified operation required on the ship ; and either the ship 

 or shore should be capable of being made the controlling 

 station, and in such a way as not to interfere with each other. 



The electrician has to provide for two kinds of faults, one of 

 which is brought about by the rupture of the conductor, and 

 the other by a flaw or defect in the dielectric. 



The rupture of the conductor involves the cessation of signals 

 from one end to the other, and may be either accompanied by 

 or without defective insulation on one or both sides of the broken 

 conductor. 



As the shore must transmit its results to the ship telegraphi- 

 cally, it is evident that in the case of "no continuity "j- the ship 



* Communicated by the Author. 



t The technical expression for a broken conductor. 



