fast as ten minutes per down-up cycle should be possible without 

 overloading any components. 



Work at the Hydrographic Office indicates that electronic bathy- 

 thermographs which employ a single thermistor and contain the 

 electronic circuitry which is required to convert the output to a 

 frequency signal in the "fish" have a tendency to drift. This drift is 

 caused by changes in the ambient temperature which in turn affects 

 the characteristics of the circuit. 



As a result, much work has been put into an instrument called the 

 Hydro Linear BT. In this device only the sensors are in the "fish". 

 The temperature element consists of two thermistors in a series/ 

 parallel circuit, one of the thermistors being in series with a precision 

 wire-wound resistor. The purpose of this arrangement is to overcome 

 the nonlinearity of the thermistors. Also in the n fish n is a pressure 

 transducer of a helical tube-type. The ranges of the instrument are 

 0° to 30°C and to 1,500 feet. Initial tests indicate that an accuracy 

 of ±0.01°C can be achieved. The data are plotted on an x-y recorder. 

 Full scale ranges of 0° to 1°, 0° to 5°, 0° to 10°, 0° to 20°, and 0° to 

 30°C can be selected. Similarly, depth scales of to 150, to 350, 

 to 750, and to 1,500 feet can be selected. A depth accuracy of 

 ±7.5 feet for the 1,500-foot range is claimed. At present, the device 

 cannot be used from a ship that is underway. 



The use of thermistors is proving to be a problem. The facts that 

 they are nonlinear, require aging to give repeatable results, and are 

 difficult to match raise many problems in their use in temperature 

 probes. 



Other problems also have become apparent. If the electronics 

 are taken out of the n fish n , then a multiconductor cable is necessary. 

 The increased cable diameter requires larger winches and makes 

 it difficult to get the "fish" to depth if an underway device is being 

 sought. 



b. Thermocline recorder (multiple sensors, temperature 

 continuous with time) 



Thermistor arrays such as the Richardson chain are increas- 

 ing the understanding of temperature distributions in the ocean. A 

 requirement exists for one or more instruments of this type. The 

 Hubbard isotherm plotter, which is used at present with the chain, is 

 not considered to be very well suited for the subsequent processing 



III -9 



