by clicking the arrows at the side of the page, or by using the toolbar.
by clicking anywhere on the page.
by dragging the page around when zoomed in.
by clicking anywhere on the page when zoomed in.
web sites or send emails by clicking on hyperlinks.
Email this page to a friend
Search this issue
Index - jump to page or section
Archive - view past issues
FLEXO Magazine : November 2010
Technologies & Techniques A signal can be approximated using a Fourier series, con- sisting of sine and cosine functions, i .e. harmonics. Each har- monic in the series contributes to the sum of the whole signal. A larger number of harmonics in the series gives a smaller difference between the original signal and the approxima- tion. As a result of the Fourier transform, the information that describes the signal is changed into the frequency domain, so that the information consists of a finite series of sine and cosine components, all making different contributions to the signal. The sine and cosine components can be written in expo- nential notation and they are then divided into an amplitude part and a phase part. The amplitude part indicates the contribution which each frequency component makes to the whole signal. The phase part describes where in the signal this contribution is located. The amplitude part seems to be most important for print mottle assessment (Palmer 1999). The amplitude part of a signal can be presented in a power spectrum. A power spectrum corresponding to an image is two-dimensional, since a printed or projected image is a two- dimensional signal. The power spectrum is often presented as two-sided, due to symmetry, where each component oc- curs on both sides of the origin. The value at the origin of the power spectrum represents the accumulated magnitude of the whole signal. The frequency of the components increases with increasing distance from the origin. The further from the origin, the higher is the frequency of the component. The two-dimensional power spectrum also contains informa- tion about the orientation of the components. The direction from the horizontal axis of the power spectrum to a component represents its orientation. Fahlcrantz (2003a) has reported that, if there is a major variation in an image mainly along one direc- tion, the components along the same direction of the power spectrum will be the main contributors to the variation. Filtration using a band-pass filter in the frequency domain makes it possible to extract frequencies. The band-pass filter only allows frequencies within its bandwidth, the distance from the cut-off frequencies, to pass through. Filtration in the frequency domain can also be done with regard to orienta- tion. Orientated filtration can give useful information concern- ing variations in a specific direction. Perception of print mottle. The types of textural feature which are important in texture perception have been investigated by Ravishankar Rao and Lohse (1996). They reported that repeti- tive non-random regular locally orientated uniform textures were significantly important. Fahlcrantz (2003b) has reported that systematic noise in printed pictures, such as stripes, bands and texture, was considered to be more disturbing than random noise. The reason why systematic variations are more important than random variations is primarily that it is valuable for humans to be able to detect boundaries of objects which dominate natural scene images (Fahlcrantz 2003a). The result of the instrumental measurements of print mottle must be processed to take into account the way in which an observer rates the quality. The correlation between the instru- mental measurements and the visual ratings is affected by the band-pass filter range used and the orientation (Johans- son 1993; Fahlcrantz 2003a). Fahlcrantz (2003b) has presented a method which showed a high correlation between instru- mental evaluations and visual ratings when printed samples with systematic mottle appearance were investigated. Ink Transfer & ConTaCT MeChanICs Optimizing ink transfer and ink immobilization into the substrate may be the most crucial step to reach high print qual- ity. Since the basic step in any mechanical printing process is the contact between the ink-covered printing plate and the substrate, it is easy to see that the contact mechanics of the printing nip are important for ink transfer and print quality. Hsu (1963) stated that when a uniform liquid ink film is pressed against the surface, the depth of the deepest part of the surface that can be coated is equal to the initial ink film thickness. However, further research has shown that ink transfer is more complex. The dynamic printing process, including print- ing press conditions, interaction between paper and printing plate and ink properties, also influences ink transfer. In this section, a brief overview of ink transfer is presented, followed by sections addressing the concepts of contact area and con- tact pressure, compression of paper and printing plate, effect of nip mechanics on ink transfer, and finally, a section address- ing nip mechanics and ink transfer in the post-printing process. Ink transfer. The primary step affecting print quality using a mechanical printing method is the transfer of ink from the printing plate to the substrate. Several researchers have investigated this ink transfer. Walker and Fetsko (1955) devel- oped an equation describing the manner in which a fraction of the ink on the plate is transferred to the paper in the print- ing nip. They suggested that there was incomplete contact at low ink levels and that the transferred amount of ink was in- sufficient to satisfy the ink immobilizing capacity of the paper. In addition, they suggested that the non-immobilized free ink between the plate and the substrate was subject to splitting. The splitting depends on the interaction between the ink and paper and on the printing press geometry. From this viewpoint, they proposed a transfer equation as follows: where y is the amount of ink transferred to the paper [g/m2] or [μm], x the initial amount of ink on the printing plate [g/m2] or [μm], k a constant related to the smoothness of the paper [m2/g or 1/μm]. This parameter describes the rate at which cover- age increases with increasing ink quantity on the plate, b the immobilization capacity of the paper surface for the ink [g/m2] or [μm], i.e . the maximum amount of ink that can be immobilized in the paper. It depends on the surface rough- ness and on the absorptivity of the paper and f the fraction of the ink transferred to the paper (0 ≤ f ≤ 1) during splitting of the free film between the plate and the surface. Walker and Fetsko reported that the value of k increases with increasing printing pressure and that it increases slightly with decreasing printing speed. Their results also showed that the extent to which k increases depends on the mechanical properties of the paper such as its compressibility. It was also reported that both b and f increase with increasing pressure and decreasing speed. Mangin et al. (1981) used different curve-fitting methods to compare, estimate and interpret vari- ous ink transfer equations. It was found that the Walker-Fetsko equation was superior, but by introducing one more parameter the exactness of the model increased according to Karttunen et al. (1971). Zang (1993) later reported that the Walker-Fetsko equation is satisfactory for coated pa- pers but not for uncoated papers. According to Nordström and Grön (1998), the ink-substrate interaction can be divided into three regions, I, II and III, when film transfer is plotted versus ink film on the printing plate. In region I, the percentage ink transfer increases with increasing ink film thickness on the form, but the filling in of pores and the coverage of the surface is incomplete. In this region, the surface structure has the greatest influence on the ink transfer. In region II, the percentage ink transfer decreases with increasing ink film thickness. Within this region, the film immobilization, i.e. the absorptive and volumetric proper- ties of the substrate, determines the ink transfer. In region III, the percentage ink film transfer remains constant, and in this region the fluid properties, rheologi- cal and fluid dynamic force, determine the ink film transfer. It has been reported that the sur- face structure affects both ink transfer and print evenness and that a smooth surface is necessary to receive a uniform film of ink from the printing plate (Zang and Aspler 1995). Kapoor and Wu (1978) reported that ink transfer depends on the geometric properties of the paper, such as voids, non-contact areas, and sharpness of asperities. These voids and asperities cause small spots which have a color or texture different from the rest of the paper. Barros et al. (2004) have shown that uncovered areas in full-tone flexo-printed paperboard are associ- ated with depressions in the surface topography. Jensen (1989) found that the print density decreased with increasing Parker Print-Surface (PPS) roughness 82 FLeXO november 2010 www.flexography.org FLX_Nov10_mech.indd 82 11/1/10 2:26 PM
Sustainable Fall 2010