Trying to assess or judge which of these 

 Ughts is better than the other is difficult, 

 because, like submersibles themselves, each 

 one has assets that make it desirable for 

 certain tasks and undesirable for others. 

 Consequently, the most legitimate and real- 

 istic approach is to present characteristics of 

 each and let the user decide which serves his 

 purpose best. 



C. L. Strickland and R. L. Hittleman of 

 Dillingham Corp. (5) presented the results of 

 their tests which compared the above three 

 lights in respect to light output, sensitivity 

 to input power variations, attenuation ver- 

 sus distance, compatibility to television and 

 color rendition. The test results clearly de- 

 fine the advantages and disadvantages of 

 each type and simplify the selection of light 

 types for particular tasks. 



For these tests each light or lamp was in 

 an identical housing, envelope and reflector 

 configuration and operated at 250 watts (ex- 

 cept for the color rendition tests when both 

 250 and 1,000 watts were used). The results 

 are highlighted below. 



Light output — (lumen efficiency) 

 quartz iodide 18 lumens per watt 



mercury vapor 40 lumens per watt 



thallium iodide 75 lumens per watt 



At 120 volts the measured output showed 

 the thallium iodide to be twice that of mer- 

 cury vapor and six times that of the quartz 

 iodide. Varying the input voltage showed the 

 mercury vapor and thallium lights to drop to 

 85 percent of the 120-volt output; quartz 

 iodide has an output of less than 45 percent 

 of its output at 120 volts. 



Attenuation — The centerbeam candlepower 

 (cp) of each light was measured in clear 

 seawater at distances of 1, 2, 3 and 3.5 me- 

 ters from the source. The data at 2 meters 

 are as follows: 



Attenuation of 

 Output (cp) cp at 2ni initial output 



(2 m) 



quartz iodide 1,100 110 90% 



mercury vapor (not given) (not given) 80% 



thallium io- 5,500 1,500 72% 



dide 



Contrast Level Versus Distance — An experiment 



was conducted using gray scale targets at 2, 

 4, 6 and 8 meters from a television camera to 

 examine where the various lights fell in rela- 



tionship to the peak of the video response 

 curve. The thallium iodide provided much 

 better contrast than the other two, espe- 

 cially at 4, 6 and 8 meters. Only a slight 

 difference was measured between the mer- 

 cury vapor and quartz iodide lamps. This 

 was explained on the basis of a camera fea- 

 ture which automatically compensates for 

 the lower light output of the quartz iodide. It 

 was theorized that if the targets were far- 

 ther apart or the water more turbid, the gas 

 discharge lamps would have more clearly 

 demonstrated their superior penetration. In 

 this respect Figure 10.5 compares both the 

 spectral sensitivity of both the human eye 

 and a typical black and white TV camera. 

 The curves for both eye and camera peak at 

 about 5,500 Angstroms which is almost iden- 

 tical to that of thallium iodide's principal line 

 spectrum (5,350 A). 



Color Rendition — At 250 watts for each lamp 

 the following results were obtained by com- 

 paring photographs of a spectral color chart 

 at the varying distances: 



1 meter: quartz iodide showed strong 



green and blue attenuation and 

 the violet appeared almost red. 

 mercury vapor showed poor 

 color rendition except in the 

 blue-green region. 

 thallium iodide showed some 

 red output, blue and violet; 

 green is predominant. 



2 meters: quartz iodide (1,000 watts) pro- 



vided good color rendition at this 

 distance, 2.5 meters seemed to 

 be the limit for the 1,000-watt 

 light. 



mercury vapor showed greens 

 and blues, otherwise the light 

 had very little color rendition. 

 thallium iodide showed yellow 

 and green, other colors were 

 non- 

 existent. 

 Strickland and Hittleman conclude that 

 each light has its distinct advantages for 

 certain applications, but predict that thal- 

 lium iodide should become the primary un- 

 derwater light source in the near future. 



Subsequent to the above study, A. L. Waltz 

 (6) of the Naval Undersea Research and De- 

 velopment Center conducted an investiga- 



475 



