Expanded Color Gamut Primaries: The FQC EG Project

Recommendations

Using the theoretical, ideal color aims illustrated above, the committee developed practical aims for a 1.5 to 2 BCM ink transfer from an 800 to 1000-line anilox that represented a reasonable flexographic process print. The aims were given to a small number of converters who requested ink sets or received ink sets from inkmakers that are close to these aims. Note that there are two BLUE pigments; both are used in flexible packaging, though the redder shade blue is not available as a mono-pigmented ink. In fact, there are not many really good red-shade blue inks, capable of being printed in a process set, so this ink needed to be produced by blending two colorants.

The inks were printed on typical job stocks (coated paper, clear film, opaque film, foils, board) using an IT8/7.4 target to capture the fingerprint of the inks (CMYK in one run and then substitute OCMK, YGCK and finally CMVK), as demonstrated in the Esko report. An experienced inkmaker determined whether a specific pigment could be dispersed into a flexo solvent or water born vehicle in a way that resulted in an ink that printed a strong or bright color using low volume aniloxes and still remained transparent enough for overprinting.

To fully characterize the increase in gamut during a pressrun, the converter obtained a known, reliable CMYKOGV profile target and processed those targets into plates. Such targets are available from high end profiling packages, such as X-Rite’s ProfileMaker for Packaging. There may also be third party forms available from some inkmakers or from some prepress providers. Label printers chose a substrate that was appropriate to their business and wide web similarly to their business. Thus, the committee received prints on both paper and flexible film substrates. The printed substrate was cut and sampled as per ANSI/CGATS Recommended Industry Practice, Color characterization dataset development—pressrun guidelines.¹¹ These guidelines describe how many charts to read and how to select those charts from the full finger printrun.

A 7-color test chart contained a very large number of patches and it was impractical to try to read those patches manually using a handheld spectrodensitometer. Fortunately, X-Rite has included the EyeOne and iSIS in the new XRGA¹² standardization scale, so that readings from either of these instruments were comparable to the scales obtained from the handhelds. The printed sheets were collected and provided to two or three locations for color characterization using a modern digital spectrocolorimeter, such as the EyeOne-IO, the EyeOne-iSIS or a similar X-Y scanning system, following the guidelines of ANSI/CGATS Recommended Industry Practice, Color Characterization Dataset Development – Procedures for Color Measurement System Process Control and for Inter-lab Coordination.¹³ From the measurement data and profiles created, the gamut was plotted and the volume of the gamut solid computed, as well as the overlap or lack of overlap between the 4-color gamut and the 7-color gamut.

Field Trials

To validate the reports from the literature, a call was sent out to flexographic converters who would be interested in investigating, or were in the process of investigating, the use of extended gamut process colors in the printing plant. About 10 converters answered the call and volunteered to assist in this study.

There are currently no standard characterization targets for extended color gamut printing. In addition, some of the commercial ICC profile analysis software, designed for n-color applications, does not compute the n-color gamut correctly. So, the committee followed a recommendation from Esko and each converter was asked to use the standard ANSI-CGATS IT8/7.4 CMYK test chart and run the chart three times, substituting an extended gamut color for one of the standard process inks. In addition, ESKO developed a template for Microsoft Excel¹⁴ to accept the color readings from the template and generate the gamut increase.

Representative sheets from a pressrun were sent to the committee and all color readings were made using an X-Rite EyeOne-IO scanning spectrocolorimeter. All readings were made using a single instrument by the color team at RR Donnelley. The results of those comparisons are summarized in Table I but the full data set can be obtained from the FTA.

Conclusions

There are some general conclusions that can be made without knowing the exact increase in gamut that may be achieved in any specific printing process. First, ECG is not a panacea for the costs and problems of packaging printing. Any given spot color can be tested against an ECG primary set by passing the CIELAB values through the ICC profile. If the predicted tones of the primary inks are equal to, or greater than, the full tone value, or if they are flagged by the color management method as being out of gamut; then it will not be possible to produce that spot color by process printing. Such “out of gamut” colors can be identified to the print buyer and the color may be negotiated to allow a color that is within the gamut to be substituted.

Second, process color printing is not as easy as line color printing. Publication printing has very large tolerances, not because the printers or print buyers do not care about money; but because they have learned over the decades that it is not easy to hold tight tolerances from day-to-day or even from beginning to end of a long run. Uncontrollable parameters, such as ambient temperature, machine frictional temperature, humidity, dielectric constants, solvent pickup/absorption all can affect how the ink is picked up or transferred.

Third, the ECG ink set must be capable of maintaining the full functionality of the package. So, if a spot color ink was doubling as a lamination adhesive, the ECG ink must also be able to do so. This may further limit what pigments and products can be used on that converting line. Switching between surface printing and reverse printing will be more difficult when using an ECG ink set. The inkmaker can help in selecting the right level of customization for any given ink set, but the expectation that one ink set will print everything that was printed previous with line colors is not realistic.

As a result, it has generally been observed that there is very little saving in ink consumption and costs in ECG printing over spot color printing. Cost savings have been reported when comparing job costs from cradle to grave—the total cost of converting. You will find that in multi-press shops, assigning jobs becomes more complex, as you must balance the available equipment with the available experience of the operators and support staff. The most successful ECG systems have reported utilizing dedicated presses, which are setup with 10 or more stations to handle the seven process inks plus any white ink, metallic inks and clear coatings.

When choosing extended gamut primaries the following generalizations may be applied, based upon the field trials reported here. The first ink to be added should be an orange ink, preferably a slightly yellowish orange. Examples include Pigment Orange-34 and Pigment Orange-64 (average gamut increase 23 percent). The next primary color to add would be a green (average gamut increase 22 percent). In this study, it was assumed that a yellow-shade green would provide the largest gamut, but in fact, the more traditional bluish green of Pigment Green-7 outperformed both G-36 and blends of green and yellow pigments. Finally, the last color to add would be a very reddish blue or a bluish violet (average gamut increase 16 percent). There are not very many red-shade blue pigments available and some users have reported using a primary produced by blending a violet or purple with a less reddish blue to achieve the hue that they desire. Blended or toned primaries have been successfully utilized in process printing, but it will be more difficult to maintain a stable color since the blended ink will have two sources of pigment-based variations. Based on the pressruns in this study, Both the mean and median increase in gamut, across all substrates, from a standard CMYK ink set to a CMYKOGV ink set is 60 percent.

The general recommendations on hue are such that, the hue of the orange primary should be close to being complementary, that is 180° opposite, to the cyan primary, the violet primary should be close to being complementary to the yellow primary and the green primary should be close to being complementary to the magenta primary. From the results of this study the following can be recommended:

• Magenta (PR57.1 or PR52) with metric hue angles between 0° and 8°;
• Yellow (PY14 or PY74) with metric hue angles between 93° and 98°
• Cyan (PB15:3 or PB15:4) with metric hue angles between 230° and 240°
• Orange (PO34 or PO64) with metric hue angles between 50° and 58°
• Green (PG7) with a metric hue angle between 178° and 182°
• Violet (PV23) with metric hue angle between 302° and 309°

Acknowledgements

The Committee wishes to thank the FTA and the staff for all of their support during this project. In particular, Mark Samworth, Esko, and Al Bowers, RR Donnelley invested hundreds of hours preparing the artwork, reading the supplied prints and preparing the database of summary data. This paper reports only the very pinnacle of the mountain of work that they performed. We also thank all of the converters who were willing to share the printrun experiences with the committee and by way of this report with the entire membership of the FTA.

About the Author: Danny Rich is senior color physicist at Sun Chemical Corp. He is based in Carlstadt, NJ.