© copyright FACULTY of ENGINEERING ‐ HUNEDOARA, ROMANIA 91 1. Iva ŠARČEVIĆ, 2. Dubravko BANIĆ, 3. Diana MILČIĆ COLORIMETRIC DIFFERENCES ON WOOD SUBSTRATE DUE TO VARNISHING INFLUENCE 1-3. DEPARTMENT FOR GRAPHIC MACHINES, UNIVERSITY OF ZAGREB, FACULTY OF GRAPHIC ARTS, ZAGREB, CROATIA ABSTRACT: Digital printing machines with ink jet technology allow printing on wood substrate and varnish can be used as final process in wood finishing. Although varnish is protecting printed ink and increasing mechanical properties of wood surface it is also changing hue and saturation of printed color. The aim of this study is to quantify that difference in color value printed on a wood substrate with and without varnish layer. For that purpose, standardized colorimetric methods were taken based on CIE L*a*b* values using the equation for color differences CIEDE2000. KEYWORDS: wood substract, varnish, ink jet, CIALAB space, color gamut, color differences CIEDE2000 INTRODUCTION Digital ink jet printers can print directly on: glass, metal, stone, wood, plastics, etc. They are using solvent ink based on alcohol gel. Wood products as print substrates are characterized by great thickness and specific structure that achieves different surface properties (great absorption, surface roughness and possibility of different surface textures). For easier maintaining the wood surface the varnish is used. Varnish is a transparent and protective film with glossy or semi-glossy reflection. It is used for increasing mechanical properties of wood surface and for protecting printed ink. However, in the contact with wood surface, varnish is changing the hue and saturation of the printed color. Colorimetric differences, with and without varnish on the printed wood surface, can be measured with standardized colorimetric method based on CIEL*a*b* values. L*a*b* (CIELAB) and L*u*v* (CIELUV) color space were designed to be device independent and perceptually uniform. They were introduced in the 1976 by the Commission Internationale de l´Eclairage (CIE - the primary organization responsible for standardization of color metrics and terminology). The CIELAB color space is widely used in color imaging and printing industry while CIELUV is commonly used in the display industry. These spaces are defined in terms of transformations from CIE XYZ tristimulus values to these spaces [1]. Based on CIE L*a*b* values the gamut of reproduction can be also constructed. Color gamut is the range of a set of colors and can be represented as location in a three-dimensional color space. For the gamut of reproduction the ICC profiles are necessary. The ICC (International Color Consortium) is a consortium of those vendors founded in the year 1993 with the aim of developing a universal color management solution. The ICC profile format, defined by the ICC Profile Specification, consists of various data structures, which provide a mechanism for color transforms [2]. METHODOLOGY The research was carried out on an ink jet printer DTS (direct to substrate).The printer driver takes RGB values as inputs. A standard X – Rite profile 343 Patches test chart was printed on three different wood surfaces –chipboard, MDF board and spruce board. After the chart was printed and dried on each board, L*a*b* values were measured using a spectrophotometer i1 Pro with 45º/0º measuring geometry, under conditions 50D illumination and 2º observer.The obtained data consisted of values of RGB inputs and their corresponding spectral reflectance. As a wood finishing the varnish was applied, two layer of basic varnish and one waterborne. When varnish was dried the L*a*b* values were measured again under equal conditions.The evaluation was carried out by the values with and without varnish for each of the 343 test chart patches using the equation for color difference CIEDE2000 [3]. 2 2 2 2 2 0 ⎟ ⎟ ⎠ ⎞ ⎜ ⎜ ⎝ ⎛ ′ Δ ⎟ ⎟ ⎠ ⎞ ⎜ ⎜ ⎝ ⎛ ′ Δ + ⎟ ⎟ ⎠ ⎞ ⎜ ⎜ ⎝ ⎛ ′ Δ + ⎟ ⎟ ⎠ ⎞ ⎜ ⎜ ⎝ ⎛ ′ Δ + ⎟ ⎟ ⎠ ⎞ ⎜ ⎜ ⎝ ⎛ ′ Δ = Δ H H C C T H H C C L L O S k H S k C R S k H S k C S k L E (1) The obtained ΔE errors are Euclidean distances in the L*a*b* space. The minimum, mean, median and maximum of errors were calculated and the results are displayed in Table 1.