Connect with us

Articles

Published

on

1604-1.jpg

Many variables affect the way a printed color is perceived, and when any one of them changes, so does the appearance of the printed color. Just when you think you have produced exactly what the customer specified and used your best color-matching abilities, your color can be rejected because of the subjective eye of a viewer who didn’t sleep well the night before.

Color matching with a color-measurement device is a science. Without one, it is an art. The very thin ink deposits used in pad printing present special challenges for ink formulators, and it is necessary to employ both the science and the art in order to satisfy your customers.

Three things must be present in order for a color to exist: an object, light to illuminate that object, and an observer (such as a person or a color-measurement device). These three components contain all the variables that affect your color-matching capabilities. The majority of the variables are contained in the object to be printed, so let’s start there.

The influence of substrates

When a customer requests a particular color to be printed on an object, that color is usually selected from a color-matching system, such as Pantone, or a proprietary color master, such as a molded or painted color chip. While the color appears acceptable on the color master, it will not necessarily appear the same way on the pad-printed object — unless you take several substrate-related factors into account.

When looking at the object or material on which you are attempting to print an image, you need to consider the following four factors:

1. the substrate’s base color and surface characteristics

2. the number of print passes required to achieve the desired opacity (single or multiple hits)

3. additives, thinners, or catalysts used in the ink

4. the application process (e.g. offset, screen printing, pad printing) that was used to produce the color master (color swatch) being matched.

Base color and surface characteristics A substrate’s surface color is the single greatest influence on the final color of the pad-printed image. When you perceive a color that has been pad printed on any given material, what you are “seeing” is the light that is reflected back to your eyes from the surface of the material and the overlaying ink layer (or layers).

Since pad printing deposits a very thin layer of ink, the light reflected back to your eyes usually travels through the ink layer, then reflects off the substrate, back through the ink layer, and into your eyes. The amount of light reflected off the substrate depends on substrate color. As you might expect, the effects of substrate color on printed-image color are most apparent when you look at lighter ink colors over darker substrates.

Surface characteristics, such as the gloss level or porosity of the substrate, also play an important role. When you view a particular color printed on two substrates of the same color but with different gloss levels, you perceive the two prints differently. Color printed on a high-gloss substrate usually appears brighter or sharper. The same color printed on a textured or porous surface appears duller.

Basically, the two textures reflect differing amounts of light back to your eyes. When light strikes the smoother, glossier substrate, most of the light is reflected at the same angle and, consequently, more reaches your eyes. A textured, porous surface, on the other hand, causes the light to reflect at a variety of angles, essentially scattering it and allowing less to reach your eyes.

The lesson to learn from this is that the more closely you can match the original color master to the substrate you’ll be printing, the better off you’ll be. For example, when a specified color is taken from the pages of a PMS book, it is a color that has been printed on either a coated or uncoated white paper substrate. This presents a problem if your print involves a matte black plastic substrate.

Your ink formulator may match the color master when pad printing on a glossy white paper or plastic, but the ink formulation doesn’t match the color master after it is printed on the black substrate (Figure 2). If you have a variety of substrate colors onto which a certain color must be matched, you may need more than one ink formulation for that color in order to achieve an acceptable match on all the materials.

In certain applications, you may not obtain an acceptable match without first printing an underbase layer or two of white ink. For example, fluorescent colors cannot be adjusted to print over anything but white without losing their fluorescent characteristics. The same holds true for process colors. In both cases, the inks are transparent and need to be printed over a white substrate or a white base.

Number of print passes As previously mentioned, when you “see” a pad-printed color, you are usually seeing the color of the substrate as well. So to reduce the influence of the substrate color on perceived image color, the printed ink layer should be as opaque as possible.

In pad printing, the thickness of the printed ink-film after drying is approximately 20% of the cliche’s etch depth. If your etch depth is a typical 0.001 in., your dried ink film for a single pass has a thickness between 0.0002-0.00025 in. This is very thin when compared to screen printing or painting. And because it is so thin, the ink layer is often not opaque enough to “hide” or minimize the effects of substrate color.

One apparent solution is to increase the cliche’s etch depth. But when you increase the etch depth beyond 0.0015 in., you reduce the transfer efficiency of the ink, and the printed result is a blurry, smudged image. So that leaves only one alternative: increasing the number of times the image is printed, a process often called double printing or multiple printing. The comparative effects of single and double printing the same image color are depicted in Figure 3.

For the automotive industry, multiple printing is usually specified to build both the opacity and mechanical resistance of the ink film. But if your customer doesn’t specifically request multiple printing, you must make the call. In making this decision, you need to determine whether you can get an acceptable color match with a single print. If you decide you need to multiple print, you then have to determine whether you can still achieve the required production rates.

In situations where you can’t get an acceptable match with a single print and can’t sacrifice production time for multiple printing, you need to communicate this dilemma to your ink formulator and/or your customer. In some cases, an ink’s opacity can be improved to support single printing by reducing the amount of clear base in the formulation or adding more pigment.

Finally, on the subject of multiple printing, be sure that your nesting fixture is capable of holding the part in a precise location through successive prints. Also make sure that the part is properly supported and won’t stretch or flex during printing.

Ink additives Anything added to the ink can affect its color. Thinners, catalysts, and flow agents are almost always clear, causing them to reduce the opacity of the ink.

For example, when you add a catalyst such as a hardening agent to an ink at a ratio of two parts ink to one part hardener, you are essentially reducing the opacity by 50%. Depending on the ratio, the difference between the same ink with and without catalyst may be as evident as the difference between the single and double print shown in Figure 3.

For this reason it is important that the person doing the color match is advised of any required additives andthe specific ratio at which they are to be added. The color-matching technician can then make the required additions prior to verifying the formulation for color accuracy.

Color-master production Lastly, consider the process and material used to generate the color master. Once again, your concerns relate to the ink-film-thickness limitations inherent with pad printing. Every printing process deposits a different ink thickness, so whether your color master was produced with screen or offset printing, you must adjust your expectations accordingly and make formulation adjustments when possible.

When a color master has been spray painted or molded in the desired color, it is frequently difficult to get the same level of saturation (or purity) with a pad-printing ink. So when you get a painted or molded color master, don’t assume the pad-printing ink is going to have the same appearance.

The viewing environment

Light is electromagnetic energy and, as such, is part of the greater electromagnetic spectrum. The small portion of the spectrum that humans can actually see, commonly called “visible light,” ranges from red (which has a long electromagnetic wavelength) to violet (which has a much shorter wavelength). Since every color you “see” is just a specific wavelength of visible light, the light that you use to illuminate the object for viewing has an effect on the perceived color of the object. If the color you’re matching is to be viewed under daylight conditions, but you’ve obtained an acceptable match under fluorescent lighting conditions, you may not really have an acceptable match.

Even the color of the area surrounding the object has an effect on the appearance of the image’s color. For example, if your customer lays the object on a dark brown desk for viewing, its color appears different than when it is viewed on a neutral gray workbench. Figure 4 illustrates what happens when the same color printed on identical substrates is viewed under different lighting conditions.

When you send a customer a printed swatch or sample for approval, try to make sure that the person who judges the color views it under the same lighting conditions you used when formulating the ink. Alternatively, adjust your own lighting conditions to match those the customer will use to view the color.

Now that you understand how the substrate and viewing conditions influence the color of a printed object, it’s time to consider the third and final item necessary for color to exist: the observer. The observer may be either a human or a color-measurement device (such as a spectrophotometer). In either case, no two observers will view the same image in the same way.

Human observers can differentiate between roughly 10 million colors. In doing so, each individual’s physiology affects the way he or she perceives a given color. A well-rested person may perceive color differently than he would when tired, ill, or stressed out. An elderly person will see color differently than a teenager. Also, certain people are less sensitive to color changes within a range of wavelengths–they may even be colorblind. So, any time a color is judged for accuracy, it is a good idea to get the input of more than one individual.

Color-measurement devices, such as spectrophotometers, measure the light reflected off (or through) an object and give the color a set of coordinates in an imaginary, three-dimensional color model called “color space.” In taking measurements, each individual device is calibrated to measure the sample under a given set of conditions. These conditions involve the illuminant (light source) type, the observer’s viewing angle, spectral inclusion (taking gloss into consideration), or spectral exclusion (not taking gloss into consideration).

When different devices are calibrated using the same set of conditions, ideally the measured sample yields the same coordinates in color space. In other words, they see the same color. But subtle differences in two separate instruments yield different results. In fact, the same instrument may yield different results when two consecutive measurements are taken without changing anything. But in most cases, these subtle differences don’t result in color rejection.

When two separate devices are not calibrated with the same set of conditions, however, or when the color tolerance (calculated difference between two sets of color coordinates) is too narrow, problems arise. Both you and your customer might use color-measurement devices, but unless they are calibrated to see the color in exactly the same way, they will yield results that are as different as the subjective judgments of two human beings.

For example, if the person specifying the color measures the color master using spectral inclusion and you measure your color match versus the color master using spectral exclusion, you may end up with a non-acceptable result, even though your naked eye says the colors match. Therefore, if you are matching a color to fall within a specific computer-generated tolerance range, it is important that you ensure your color-measurement device is set up to use the same set of viewing conditions as the device used when the color was specified.

I prefer to use a color-measurement device as a tool in color matching. It generates a useful reference that indicates the positions of the matched color and color master in the color space. This tells me which way to adjust my formulation to come up with an acceptable match. But I don’t rely solely on a color-measurement device for the purpose of accepting or rejecting a color, because on numerous occasions, I have seen them reject visually acceptable colors or accept visually unacceptable colors.

This is where the art of visual color matching comes into play. In the end, it is really people, not color-measurement devices, that look at the color. For this reason, I prefer to get the visual perceptions of several people on color matches, rather than relying on tools alone.

Color management is made simpler when you consider the variables represented by the object being printed, the lighting conditions under which a printed part is viewed, and the observer who views the color. Understanding and controlling these variables can turn a daunting challenge into a profitable–and even enjoyable–area of production.

Advertisement

Most Popular