Back to Basics

Fluorescent Penetrant Testing and Color Vision Deficiencies

by William H. Bailey^*

 

Introduction
Inspectors of various materials and in different industries have one thing in common. They must have the ability to see adequately in order to conduct the proper testing of products, items or systems being fabricated or installed. Based on many inherent factors, the inspector may be requested or required to pass a normal or, in some cases, a special vision test. Individuals working in the field of nondestructive testing (NDT) are generally required to be certified to the quality of their vision capabilities by specifications used in their specific area of employment.

 

Vision Testing
Vision requirements for NDT inspectors is a topic given little attention over the years. It is usually assumed that if the inspector is able to pass the normal vision test given by the average optometrist, then the inspector would also be qualified to do fluorescent penetrant testing. Most optometrists know very little about NDT requirements and do not attempt to assist companies in providing vision tests sufficient to meet specifications. For example, most are not familiar with the use of ultraviolet light for testing parts or the limitations of color deficient personnel who may be using the equipment. The test usually covers vision acuity, where the inspector must read a specific print size at an acceptable distance, usually 305 to 381 mm (12 to 15 in.). The color vision portion of the examination normally consists of reading at least 14 of 36 pseudo-isochromatic plates. The number of missed plates, 22 of 36, would be 61.1% of the total. This percentage, when considered as 22 missed indications, cannot be considered acceptable when testing materials that may be used in the manufacture of critical parts for aerospace or aircraft, pressure vessels or vehicles.

---

Stringent color vision testing is necessary when safety is a major concern.

---

In today's field of electronics, color coding is essential. In the fabrication of highly stressed pressure vessels, the piping and various assemblies are connected to each other and are usually assigned particular color codes for what they convey and where they are going. Inspectors with only a 38.9% acceptable color vision rating (having read 14 of 36 correct plates) should not be considered acceptable. Stringent color vision testing is necessary when safety is a major concern. In many cases, tests are carried out at different stages of processing or fabrication in order to ensure that the final product is safe for use.

 

Background
In 1966, a report was issued that described how the Navy was attempting to design a method of testing personnel and placing them for specific jobs (Paulson, 1966). Those who took the color vision tests and would normally have been rejected by the pseudo-isochromatic plates could still be deemed acceptable for some jobs. This routine testing was designed to pass those with normal vision and fail all those who were color deficient. Ten percent of the applicants failed. With the new tests described in the report, 30% of the color deficient applicants were found to be acceptable for specific job placements. These tests had been used since 1955.

The tests and equipment were developed for Navy use by Commander Dean Farnsworth while he was stationed at the Naval Laboratory. Subsequently, based on this work and the development of the "Farnsworth-Munsell l00-Hue and Dichotomous Tests for Color Vision" (Farnsworth, 1943) and Deane B. Judd's work (1943 and 1945), the Farnsworth Dichotomous Test for Color Blindness, Panel D-15 (Farnsworth, 1947) was created. This test was designed to distinguish the differences in visual ability between individuals who are functionally colorblind or moderately color deficient and those having normal color vision. Those individuals with color deficiencies can view similar hues under variable conditions and see different colors. Red and green, blue and green, blue and pink, yellow and blue, green and brown, and tan and amber are some of the combinations that get mixed. The test was not designed to separate the partially colorblind from those with normal vision nor to identify the color aptitude of the persons with normal vision.

Likewise, selecting people for color matching, grading or other work requiring special color differentiation would have to use individualized special testing. The test can quickly separate individuals into three categories: normal, colorblind and partially color deficient. The color deficient group may further be subdivided into the color mix that they misinterpret. Color discrimination begins to descend in desirability when using northern daylight, daylight blue incandescent lamps, daylight fluorescents and ordinary incandescent lamps. The testing of individuals with a moderate color deficiency and the color range of indications all fall within the parameters of their deficiencies. This alludes to a poor combination of color deficient inspectors in an area where the colors are the most difficult for them to identify. Whether the contrast ratio offsets the color deficiency of the inhibited inspector has not yet been explored, but should not be forgotten.

 

Physiological Factors
There are two standard eye responses. At luminance levels above 3 cd/m² (0.3 cd/ft²), the eye's spectral response is primarily determined by the cone shaped receptors of the retina and is called the photopic response. At luminance levels much lower, such as 30 µcd/m² (3 µcd/ft²), the rod shaped receptors respond and they are more sensitive in the blue color spectrum. This is called scotopic response. The rod receptors are more sensitive to light than the cone receptors (Ness, 1987). Luminance levels to which the eye responds with a variable spectral response are called mesopic. They range from approximately 680 to 1746 lm/W. Observations made utilizing the cone receptors require photopic conditions or the foveal area of the retina, with the rod receptors at the near threshold illumination level surrounding the center. Scotopic vision is not used for color recognition while photopic vision is used. The fovea centralis is an area of tightly packed cone receptors located in the optic center of the retina. The area is surrounded by both rod and cone receptors with the ratio of cones to rods decreasing with the increasing distance from the fovea. Faint indications are seen better when indirectly viewed. This is how a person in a darkened room uses peripheral vision and can see to a better degree. The fovea becomes a blind spot when low luminance and maximum sensitivity occur at approximately ten angular degrees from the fovea center. The fovea itself has about 1 to 2 degrees in the visual angular field (Figure 1).

The age of the inspector should also be considered. With age, pupil size is reduced, as is the luminous flux at the retina. This does not include contrast differences of the background reflectance and the indication response. In Figures 2 and 3, the electromagnetic spectrum is shown, with the visual spectrum of colors identified in Figure 2 and the ultraviolet region shown in Figure 3. The near ultraviolet A region, used as the light source for penetrant testing, is on the side of the ultraviolet region bordering the visible spectrum. To a normal observer, the color spectrum appears as a series of chromatic colors, starting with the shortest wavelength and proceeding through the longest color wavelength in the visible range of the electromagnetic radiation spectrum: violet, indigo, blue, green, yellow, orange and red.

Figure 1 - Components of the human eye in cross section. 

 

Figure 2 - The visual spectrum: human eye response at 1070 lx (100 ftc). 

 

Figure 3 - The electromagnetic spectrum with an enlargement of the ultraviolet region. 

 

 

Ultraviolet and Visible Spectra
Ultraviolet A is considered long wave and is not considered harmful to the human eye. Ultraviolet B is considered middle wave and can cause sunburn as well as be harmful to the eyes. Ultraviolet C is considered short wave and is used for germicidal purposes and is harmful to the eyes and body. Daylight viewing is considered to be a range of 540 through 570 nm (2.1 x 10^-5 through 2.24 x 10^-5 in.), with the average being 555 nm (2.19 x 10^-5 in.). Toward either end of the visual range, brightness diminishes. As the colors appear to diminish to either side of the average wavelength, the luminosity function located at 555 nm (2.19 x 10^-5 in.) diminishes with the color.

Colors in the visual range of any wavelength can be made by mixing lights of the three basic colors (red, green and blue) in specific proportions. These are the basic colors that account for normal vision being classified as trichromatic. The normal observer may be classified to discriminate light and dark, yellow/blue and red/green. An observer having abnormal trichromatic vision requires a different mixture of the three colors to see the same color as the normal observer. If the observer had mild trichromatic abnormalities, his or her description of the created colors would be the same. If the observer had extreme trichromatic abnormalities, the colors described would be similar to those of a dichromatic, who would make the same discriminations, but with one of the three lights at a lesser percentage, usually red/green.

 

Testing for Color Vision Problems
There are two main types of trichromatic abnormalities, called protanomaly and deuteranomaly. With either of these conditions, inspectors could make color discrimination decisions with difficulty, while a dichromatic could not. The inspector with protanomaly of all degrees is deficient at the long wave end of the spectrum due to a loss of luminosity function. Trichromatic and deutermatic inspectors have a luminosity function within normal limits. Dichromatic observers require just two color wavelengths to be mixed to form a wavelength color that may be discriminated. They can make only two types of visual decisions. One is achromatic (light/dark) and the other is chromatic (either yellow/blue or red/green, usually the former). Dichromatic observers are limited in their visual range and they would not pass the color vision requirements given by an optometrist.

Today, the method of testing individuals for color vision capabilities relies on the Farnsworth-Munsell 100 Hue Test for the Examination of Color Discrimination (Farnsworth, 1957). It encompasses all phases of color vision. The Farnsworth Dichotomous Test for Color Blindness, Panel D-15 (Farnsworth, 1947) covers only a portion. Fifteen different color paper disks of 12.7 mm (0.5 in.) diameter are sunk 2.5 mm (0.1 in.) below the 2.5 mm (0.1 in.) rim of plastic caps. The backs are numbered sequentially. A box with a terminal color is fixed at one end of the box. The candidate is given the 15 colored caps in a random mix and is required to match them in consecutive order to the terminal color without looking at the numbers on the caps. About 80% of adult males make fewer than 12 transpositions, most of which are simple two cap exchanges. These errors are random, seldom reoccur on retests and are not restricted to one region of the color scheme. The instruction booklet that accompanies the box of caps has all the information required for a qualified test of vision status. It can define the area of the colors that may be missed, therefore limiting the type of test that could be performed. All the noted testing is performed using white light of about 6500 K (11 200 ºF), average daylight.

Individuals have variations in their spectral responses, therefore setting up a single standard to which they would have to comply would be quite difficult. Any standards being used must be able to encompass all of the shortcomings of an individual inspector but be strict to the point that any test will be capable of detecting vision problems that cause harmful conditions in the use or operation of the article being tested.

 

Color Discrimination and Penetrant Testing
Fluorescent penetrant testing is carried out in a secluded area where the ambient room lighting is controlled to less than 21.5 lx (2 ftc) of white light. Total darkness is never achieved. The sensitivity of discontinuity detection is controlled by the inspector's limitations, which may include protective eyewear, reflective clothes, reflected near ultraviolet light from the part surface, the angular reflections of the discontinuities and their size and quantity of luminescence based on the type of penetrant used. The nature of fluorescent penetrant testing raises the question as to whether an individual with color deficiencies may make more errors when performing a test of fluorescent indications as viewed under near ultraviolet illumination than would an inspector with normal vision.

The viewing of fluorescent indications is done using a mercury vapor lamp and a blue/purple filter. That allows only light at a wavelength of 365 nm (1.4 ´ 10^-5 in.) to be used. The fluorescent dye that makes up indications is yellow. There are other dyes that may be used for special applications, such as red, orange, blue or green. These color groups are the ones that confuse the moderate color deficient individuals. Fluorescence, also know as "cold light," may occur in gases, liquids or solids. When the material is radiated by the ultraviolet lamp, the energy excitement causes a secondary emission of longer wavelengths to be emitted from the material. In this case, the material is a liquid that has been applied to an object and allowed sufficient time to seek out discontinuities that are open to the surface. Capillary action carries the fluid into the discontinuity. When the surface residue has been removed, the only fluorescent fluid left is that which entered the discontinuity. With time, the fluid will again, by capillary action, creep out to be exposed on the surface. In order to make the fluid easier to detect, a blotting powder may be applied to the surface (it acts like a sponge placed on a spill.) This coverage also acts as a contrasting background. The ultraviolet testing lamp is held at approximately 305 to 381 mm (12 to 15 in.) from the test surface and has a radiation value of 1200 µw/cm². The contrast ratio is much higher than it would be under normal daylight test conditions.

 

Conclusion
Based on the research performed, nothing was located to show that any recent work has been done involving the testing of color vision utilizing fluorescent materials and ultraviolet illumination. Quite by chance, contact was made with the newly developed Eye Institute of the Cleveland Clinic Foundation, Cleveland, Ohio. They were not aware of any new developments in this direction, but they are doing some work in the color vision area. Perhaps they may get new data regarding the use of color deficient persons in NDT.

 

REFERENCES
Farnsworth, Dean, "The Farnsworth-Munsell 100-Hue and Dichotomous Tests for Color Vision," Journal of the Optical Society of America, Vol. 33, No. 10, October 1943, pp. 568-578.

Farnsworth, Dean, The Farnsworth Dichotomous Test for Color Blindness, Panel D-I5, New York, Psychological Corporation, 1947.

Farnsworth, Dean, The Farnsworth-Munsell 100 Hue Test for the Examination of Color Discrimination, Baltimore, Maryland, Munsell Color Company, 1957.

Judd, Deane B., "Facts of Color Blindness," Journal of the Optical Society of America, Vol. 33, No. 6, June 1943, pp. 294-306.

Judd, Deane B., "Standard Response Functions for Protanopic and Deuteranopic Vision," Journal of the Optical Society of America, Vol. 35, No. 3, 1945, pp. 199-219.

Ness, Stanley, "White Light/Visible in the I/II Dye Penetrant and Fluorescent/Non-fluorescent Magnetic Particle NDT Inspection Processes," Visual Committee Report, American Society for Nondestructive Testing Fall Conference 1987, Columbus, Ohio, ASNT, pp. 1-8.

Paulson, Helen M., The Performance of the Farnsworth Lantern at the Submarine Medical Research Laboratory and in the Field from 1955 to 1965, Report Number 466, US Naval Submarine Medical Center, Submarine Base, Groton, Connecticut, 1966.

 

* 1162 Dover Center Road, Westlake, OH 44145-1315; (440) 835-4017; e-mail <wbailey@quack.nacs.net>.

 

Copyright © 2003 by the American Society for Nondestructive Testing, Inc. All rights reserved.