Commercially matching colors is done by well paid experts with fancy equipment, but it can be done reasonably well at home. Since some people are color blind in certain or all areas, color perception is a personal thing. Many other environmental factors effect the appearance of colors. Scientifically colors are measured by electro-magnetic wave, light frequencies with a spectrometer, which is a fancy prism. As seen in a rainbow, a prism bends light at different angles by frequency, with low frequency red at one end of the spectrum and high, violet at the other. Most colors viewed are very complicated mixtures of these frequencies and their magnitudes, somehow interpreted by the eyes and brain.
There are two different basic methods of controlling and mathematically handling colors. For transmitted and filtered light, such as a monitor or TV, frequencies are subtractive. For reflected light, such as a paint finish, they are additive. For convenience the spectra are often looped around into a colorwheel, as often seen in color preferences on a computer. The primary and secondary colors are quite different from what was taught in art classes in school. Most people are now familiar with the primary color RGB (red, green, blue) adjustments and designations. Two of the secondary colors have strange names. Around the wheel are red, yellow, green, cyan, blue and magenta. On the art wheel primaries are red, yellow and blue, while with the secondaries the sequence is red, orange, yellow, green, blue, and violet. Examining the two side by side colors are shifted due to the mathematics of the two methods. The conversion problem is quite evident when programming printer drivers. Fortunately painted colors use the old familiar art color wheel.
In 1973 Munsel published a large color comparison book with thousands of color swatches in different arrangements for easier indexing. In depth instructions explained usage. A companion book listed common names and general locations on charts plus formulae to compute differences and mixing from available pigments.
The whole system of designating paint colors is based on this wheel with expansion into a cylinder. Since this is a three dimensional object, three variables are required to locate and designate any color in it. Around any wheel slice, HUES are divided into twelve name groups based on primary colors, which are divided into ten parts between them. A reddish orange or reddish brown often found on freight equipment was designated by red-yellow-red or RYR with a digit following to indicate distance from pure red.
From the center of the wheel out to the rim, CHROMA or SATURATION is divided into ten subdivided parts from grey = 0 toward higher saturation. Most hues reached about six, but some yellow reached about eight. The outside, curved surface resembles a bumpy cylinder.
Designated VALUE, wheel slices run from 0 = black at cylinder bottom to 10 = white at the top. The center axis with 0 chroma represents the pure grey scale, commonly found on computers. Lighter tints like pink, baby blue, and light tans are near cylinder top. While dark shades like navy blue and brunswick green are near the bottom.
Having very accurate , glued-on swatches the Munsel book presents three groups of arrays to aid search. Fixed value, color wheels present hues around circles of equal chroma with grey as a center. Since maximum saturations of hues at different values vary, outer circles may vary in number and may not be complete. Another group using a flattened cylindrical cut with fixed chroma presents hues horizontally and values vertically . Due to the maximum saturation limits, higher chroma sheets may have very incomplete fill-ins. The other group presents equal hue cuts, with chroma horizontal and value vertical. Every color is represented in its relative location in each of the three groups. To limit the number of swatches, adjacent ones vary to a minor degree, which is closer than the average human will normally discern.
Source light can have a drastic effect on reflected colors, due to its own spectrum and pigment qualities. Fluorescent lamps tend to shift color toward the cool violet, while incandescent lamps tend to shift toward the warm red. Developed by artists, based on human emotional reaction, these adjectives are reversed from physical properties. Due to a higher frequency, violet requires higher radiation temperature than red. Often photography film is rated in temperature degrees. Direct sunlight varies according to the elevation of the sun, effected by different distances and angles through the atmosphere. As seen in the blue sky, atmospheric reflections on colored surfaces often shift color toward violet, even in the shade. To reduce effects and establish a standard, a NORTH LIGHT, through a clean, clear window in the north side of a building, is used. Commercially very expensive substitutes are used. For any initial matching, a north light should be used, but final appearance will be effected by lighting on the layout.
Several manufacturers have offered less expensive fluorescent type tubes and spiral bulbs that claim to be true light. These are suitable for model paint matching purposes. Compare specifications and claims very carefully, before buying. Lamps tend to be expensive and over priced, but the tubes are standard sizes that fit less expensive fixtures. Although useful for matching, one folding lamp with magnifier was tried for modelling with mediocre success. The light was too close to the work, limiting tool use and the use of customary binocular magnifiers was next to impossible, even without the built-in lens. The base was always in the way of hands.
If available at you local library, examination of the Munsel books can be very enlightening and educational. Browsing the red-yellow sheets will reveal many of the prototype freight car colors, derived from earth pigments. Deriving its name from the Greek word for black, the pigment responsible for almost all hair, skin and organ coloring in mammals and humans, melanin falls in this area. Also most wood colors are included. Due to a lack of melanin, blue eyes and light hair tints reflect substrate colors. Toward the right, brilliant oranges have high chromas and medium values. Nearer the upper left, tans like ecru and beige have higher values and lower greyish chromas. Toward the bottom, browns have lower values and somewhat reduced chromas. Many of the freight car colors have hues toward red and are commonly referred to as red.
Visualizing the cylinder as a solid, presents a clearer understanding of the absolute relationships of colors. Often very large, each named color is represented by an irregular, ill defined, solid; whose dimensions vary in hue, chroma and value. They may approach a spherical shape, but more often stretch out in any direction. Pink forms a relatively thin, high value layer that extends from an orange hue to a purplish red.
Names are often confusing, non indicative or overlapping. Often they refer to the location of the origin pigment source like Tuscan or the chemical used like chromate. Others derive from similarly colored objects, which vary like peach or cherry. Many are coined. Often referred to as green, the largest part of olive is a low value, reduced chroma yellow. The companion book lists those known at the time with their customary limits. Locating those colors on hand and the desired resulting color on the solid, will lead to more successful approaches in mixing.
Before attempting to apply all this confusing knowledge, it would be wise to examine the goal. A trip to a freight yard, with strings of cars from the same railroad with supposedly the same color, will reveal that, except possibly for cars freshly painted at the same shop, no two cars will have exactly the same color. Often the ends and opposite sides do not match the first side viewed. Sometimes even the same side varies in some areas. Often acquired from the lowest bidder, successive batches of paints are not exact matches. On older equipment, smaller railroads and regions on larger railroads often bought paint more locally. Before standardization, prior to 1900, paint matching was usually at the whim of the shop foreman. Passenger cars were more often pampered, resulting in better matches.
Since cars are subjected to different environments between paint jobs, many factors can change colors. Direct sunlight often effects certain pigments in the mix, changing hue or fading colors. Frequent exposure to industrial fumes can alter colors. Various density deposits of dusts and water mixes from almost white through soil colors to black soot, quickly change colors in some areas. There is no constancy in color.
In spite of prototypical names, too often model paints are not even close matches for prototype colors. Two colors used by the "standard railroad of the world", Pennsylvania, created heated discussions and claims in the modelrailroad world. Since Tuscan red, derived from clay in Tuscany, Italy, is a common art pigment, passenger cars are usually a close match. But erroneously, sometimes it is called maroon. Some argued that locos were painted black and not very low value and chroma brunswick green. In photos the color often appeared bluish, due to sky reflections. A paint chip and Munsel proved it to be green with the description, one more drop of black and it would be black. Hundreds of thousands of cars were painted freight car red as were cab roofs and tender decks, but what color is a freight car? Only the Pennsy seemed to know. In the model world, arguments persisted and several poor matched recipes were offered. Varney came close, but MDC and Athearn used way less red, boxcar red or dark Tuscan red. Even though it was probably the most common, single freight car color until the late 1960's, after around 40 years, Floquil and some manufacturers like Bowser, finally discovered Varney was close. Even blacks and whites vary.
Although, if possible, initial matching should be done under north light, models normally will be viewed under artificial lights with varying luminance and spectra, which shift colors around. In some cases, different pigments used to achieve match will reflect differently under various lights. Due to vision quirks, painted area size and background alter the brain's interpretation of colors. In storage pigments settle in strange color layers dissimilar from the mixed result. Many customers were observed shaking bottles to determine selection, but store light changes colors and its reflection from glass disguises colors. Far worse, most colors change drastically between the liquid and dry states, making liquid matches almost impossible. Some paints continue to change, after apparently dry, until cured in days or weeks. All matching should be done with thoroughly dry and set paint swatches. Since it appears that obtaining exact matches is highly improbable and unnecessary, the aim should be to make it believable.
Although the companion book includes involved mathematical methods, most will find common sense methods preferable. Understanding the relative positions on the solid leads to the solution. Mixing two colors, the resultant will be along a straight line between them and the distance determined by the quantities of each. Adding a third color will pull the resultant toward it. Adding black by drops or less will quickly lower value, but also reduces chroma. Since both a nearly pure red, adding tuscan red to lower value in caboose red, might be more desirable. Reducing chroma can be done by adding sufficient equal value grey, but far quicker is the addition of a color diametrically opposite on the color wheel. Blue will reduce orange chroma. Deviating from the line with the additive will swing the hue around the wheel in the half on that side,
Increasing value or chroma presents a different problem. Adding white to 1 oz of darker color to raise value, may result in a gallon of mix. To minimize quantities, colors should be added to a white base slowly. The same applies to chroma, where lower level color should be added to higher slowly. In general adding darker colors produces very rapid changes, while brighter color additions yield much slower changes
Altering hue slightly is tricky, since locating the nearest correct color additive may involve all three variables. Adding very small drops of a higher chroma, almost equal value color to pull in the right direction can work effectively. If resultant chroma is excessive, drops of a similar value grey will reduce it. Matching factory applied paint for touch-up is quite difficult, even when the line and color is known, since often very small amounts of black are added to improve opaqueness or finishes are baked to speed setting.
The quantity of source paint is very critical to the results and initially should be measured accurately. Often mixing formulae use a "standard" drop, but droppers are not all created equal. Some medicine or eye droppers have produce around 1200 drops to the ounce while kitchen blasters may yield less than 100. The Perfect Science Center has droppers, which yield repeatedly, close to 480 drops = 1 oz. Counting drops is tedious at best. For larger quantities droppers may be calibrated and marked by counting drops. In some cases, whole 1 oz bottles are used for both quantity and convenience. Beware, Floquil's older 1 oz bottles were actually 1 1/8 oz. For intermediate measurements, Perfect's graduated cylinders are excellent, but are marked in cc, requiring conversion. Calibrated pipettes might be used. For the nit-pickers, chemical titration tubes with stop are more accurate; again marked in cc. Often when tinting or adding black, very small quantities are obtained by dipping the tip of a glass medicine applicator or fine brush.
Source paints from different lines are not usually mixable. Including military and railroad colors, Polly S provides the widest selection, while Floquil including Flo-Paque and military is narrower. Military bottles are usually 1/2 oz. With the transfer of Floquil by parent corporation, RPM, to the Testors division; Flo-Paque, the metallics, stains and military colors have been dropped along with many railroad colors. Due to EPA requirements, formulation has changed for the worse. The new version is not compatible with the old. Larger quantities, up to a gallon or more, are available in some lines. Most dry pigments are too coarse and may not be compatible with vehicles when mixed.
Select and obtain a comparison standard; possibly a piece of acceptably colored rolling stock. Final comparison should be under layout lighting, where color will be viewed, Since paints should be thoroughly cured before comparison, sufficient patience is required. Often taking weeks, prescheduling is wise. Here haste can very easily make waste.
It is very important to record all recipes developed or obtained, for future reference. A very simple and convenient method is to use small file cards, with name at top left, formula in preferred units below and a paint swatch to the right. Good sources for additive color swatches are paint color cards offered by some lines or swatch strips from old rack displays. Sorting these by apparent hue, chroma then value aids search. To simplify mixing odd batches percentages may be computed and added. Floquil once offered a Paint Factory, which contained a few source colors and formulae with swatches of all extant colors at that time. After cutting to size and interpolating these into the file, the additional information helped expand data on cards and improve assortment.
Note: Adjust brightness and contrast for optimum viewing.
FLOQUIL SWATCH CARDS
In an attempt to preserve hand written lettering, colors are not true. The cards are an "antique" white. Cut from the kit sheets, the left card shows the mixture for a military color, while the right is homemade for a PRR red with the mixed sample brushed on. The printed text show the parts of standard colors to be mixed. Barely visible at the bottom are the percentages of the primary mixing colors provided with the kit. These aid in sorting cards and in approximating future mixes.
BACK TO PAINTING AND COATING
BACK TO METHODS INDEX
