It was suggested to measure the total luminous flux of linear light source as a fluorescent lamp, using an approximate ellipsoidal closed surface, with both ends of the cylinder being rounded off.We found that this closed surface had been used by some manufactories; we examined, therefore, the fundamental character of the flux integrator, using “Coefficient of Relative Sensitivity” and “Coefficient of the Final Illumination of Interreflection” given by N. Muroi, one of the authors. A lso we discussed some errors due to the substitution method that compares an unknown linear light source with a standard incandescent lamp or with a known light source whose distribution is similar to the linear light distribution.As the major axis was equal to the minor axis x (1.5, 2.0, 2.5, 3.0, & 3.5), we estimated the flux integrator's errors.For instance, if the major axis was equal to 2.5 times the minor axis, the substitution of a standard incandescent lamp for an unknown linear light source resulted in less than about 4% errors.Also, the substitution of an unknown linear light source for a known light source whose distribution is similar to the linear light distribution, their light lengths being equal, resulted in far less errors.
Read full abstract