tester

Expert Guide

Color & Quality

Heidelberger Druckmaschinen AG

Kurfürsten-Anlage 52 – 60

69115 Heidelberg

Germany

Phone +49-62 21-92 00

Fax +49-62 21-92 69 99

www.heidelberg.com

Publishing InformationPrinted: 10/06Photographs: Heidelberger Druckmaschinen AGPlatemaking: SuprasetterPrinting: SpeedmasterFinishing: StahlfolderFonts: Heidelberg Gothic, Heidelberg AntiquaPrinting in the Federal Republic of GermanyCopyright © 2006 by Heidelberger Druckmaschinen AG

Trademarks Heidelberg, the Heidelberg Logo, Prinect, Axis Control, CP2000 Center, CPC, Image Control, Speedmaster and Mini Spots are registered trademarks of the company of Heidelberger Druckmaschinen AG in Germany and other countries. Other product names used here are trademarks of their respective owners.

Subject to technical and other changes.

Contents

1 Light and Color

1.1 Light is Color 2

1.2 Color Perception 4

1.3 Color Reproduction 5

1.4 Color Systems 7

2 Color in Printing

2.1 Ink Film Thickness 10

2.2 Tonal Value 10

2.3 Relative Print Contrast 17

2.4 Color Balance/Composition 18

2.5 Ink Trapping and Color

Sequence 21

2.6 Color Control Strips 22

3 Densitometry

3.1 Reflection Densitometry 24

3.2 Densitometer Filters 26

3.3 Densitometric Values 27

3.4 Measurement 28

3.5 Evaluation 30

3.6 The Limits of Densitometry 32

4 Colorimetry

4.1 Measuring Colors 34

4.2 Standard Color Values/

Reference White 35

4.3 Standard Illuminants 35

4.4 Standard Observers/

Spectral Value Functions 36

4.5 Evaluation with a Spectro-

photometer 37

4.6 Color Distance (ΔE) 38

4.7 Munsell 46

4.8 Tristimulus Photometry 46

4.9 Spectrophotometry 47

4.10 Spectral Quality Control

with Heidelberg 48

4.11 Color Control Strips 49

4.12 Color Control with Heidelberg 49

4.13 Standardization of Printing 54

4.14 Benefits of Colorimetry for

Offset Printing 57

2 Expert Guide on Color & Quality Light and Color

1.1 Light Is Color

We live in a world of color. We use col-

ors to liven up our living space, so we

feel good in it. The shapes and colors

of things have a direct impact on our

perceptions and feelings. Properly

coordinated colors evoke a feeling

of harmony, which puts us in a good

mood.

The printing industry also uses color

to enhance its products; the aim is to

consistently supply top quality to cus-

tomers.

One of the prerequisites for this is es-

tablished standards for measuring qua-

lity. And in order to assess colors, we

have to be able to “see” them. This calls

for light.

The sun emits light – it is illuminated

from within, driven by fusion proces-

ses that take place on a vast scale.

In contrast, most of the objects sur-

rounding us do not emit any light of

their own. Consequently, we can only

see them when they are illuminated

by another light source.

Light and Color1

Light and Color Expert Guide on Color & Quality 3

Light is radiation that travels at the ex-

tremely fast speed of 300,000 kilome-

ters per second. More precisely, light

consists of electromagnetic vibrations

that propagate themselves through

space like waves. Like ocean waves, each

light wave has a crest and a trough.

Light wave crest

Light wave trough

A wave can be described either by giving

its length or by indicating the number

of vibrations per second. Wavelengths

are measured in everyday units such as

kilometers, meters, centimeters, milli-

meters, nanometers or picometers.

The number of vibrations per second –

the frequency – is expressed in Hertz.

Waves of different lengths have differ-

ent properties and uses. X-rays, for ex-

ample, are used for medical diagnos-

tics, and many households are now

equipped with microwave ovens. Other

wavelengths are used to transmit tele-

phone conversations and radio and

television broadcasts.

We only perceive a very small section

of the overall electromagnetic spec-

trum as visible light. It extends from

380 nanometers (ultraviolet light) to

780 nanometers (infrared light). With

the aid of a glass prism, light can be

split into its color constituents. Because

white light consists of a mix of colors

across the whole visible spectrum, all

of the colors of the rainbow can be seen

(see figure on page 4).

The adjacent figure shows how the

wavelengths get steadily smaller as one

moves from red across green to blue.

Red (700 nm)

Green (550 nm)

Blue (400 nm)


1.2 Color Perception

It is light that makes color visible –

but why?

Color as such is not an attribute of an

objective, such as its shape. But physi-

cal bodies do have the ability to absorb

or ref lect light of certain frequencies.

We only see those colors that corre-

spond to the ref lected wavelengths.

When white light strikes an object,

one of the following cases occurs:

• All of the light is absorbed. In this

case, we see the object as black.

• All of the light is ref lected. In this

case, we see the object as white.

• All of the light passes through the

object. In this case, the object's col-

or does not change.

• Part of the light is absorbed while

the rest is ref lected. We then see a

color whose tone depends on which

frequencies are ref lected and

which are absorbed.

• Part of the light is absorbed, while

the rest passes through the object.

We see a color whose tone depends

on which frequencies are absorbed

and which pass through.

• Part of the light is ref lected, while

the rest passes through. The color

of both the ref lected light and the

light passing through changes ac-

cordingly.

Which case occurs depends on the

properties of the illuminated object.

The light that an object ref lects or al-

lows to pass through is captured by our

eyes and converted into electrical sig-

nals that travel along nerve pathways

to the brain, which interprets them as

colors.

X-rays UV IRRadio

Gamma rays Microwaves

TVVHF HF MF LF

Wavelength

Visible light

Radar Broadcast


The retina of the eye contains light-

sensitive cells. There are two types of

cells: rods and cones. The rods dis-

tinguish between light and dark, while

the cones respond to color. There are

three different kinds of cones, each of

which is sensitive to a different range

of wavelengths. One detects light from

about 400 to 500 nanometers, or bluish

colors. Other cones “see” only green

light in the range from 500 to 600 nano-

meters. The third type is responsible

for reddish colors in the spectrum be-

tween 600 and 700 nanometers.

This design, with rods and different

cones, makes the human eye so sensi-

tive that we are able to perceive and

distinguish several million different

colors.

1.3 Color Reproduction

1.3.1 Additive Color Reproduction

In the additive color reproduction pro-

cess, light of different colors is com-

bined. Blending all of the colors of the

optical spectrum yields white light.

The additive primary colors are red,

green and blue light. Each of these

represents one-third of the visible

spectrum.

Additive color reproduction can be dem-

onstrated very well with three slide pro-

jectors, each of which casts a circle of

light of one of the three additive prima-

ry colors onto a screen.

This process is used in color television

and in the theater to generate all the

colors of the visible spectrum.

Paper

Green + Red = Yellow

Green + Blue = Cyan

Blue + Red = Magenta

Blue + Red + Green = White

No light = Black

Where the three circles of light overlap,

the following secondary colors result:


1.3.2 Subtractive Color Reproduction

In the subtractive process, different

color components are removed from

the light ref lected by the white paper.

Taking out all of them results in the

color black.

The subtractive primary colors are

cyan, magenta and yellow. Each of

them represents two-thirds of the

visible spectrum. They can be created

either by subtracting an additive

primary color out of white light (for

example, using a filter) or by superim-

posing two additive primary colors.

Printing inks are translucent substan-

ces that act like color filters. Which col-

or do you get if you print a substance

that absorbs blue light onto paper?

Blue is subtracted from the white light,

while the other constituents (green and

red) are ref lected. The additive combi-

nation of these two colorants results in

yellow: this is the color we see.

In other words, the printing ink re-

moves one-third (blue) of the white

light (consisting of red, green and

blue). Suppose that two such translu-

cent inks are printed one on top of the

other, say yellow and cyan. The inks

filter out first the blue and then the

red part of the white light. What is left

is green, which we perceive.

Paper

Cyan + Yellow = Green

Yellow + Magenta = Red

Magenta + Cyan = Blue

Cyan + Magenta + Yellow = Black

No Color = White

In subtractive color composition, overprinting cyan, magenta

and yellow yields the following secondary colors:

Paper

Paper


1.4 Color Systems

Each individual perceives colors slightly

differently. So if several people are asked

to describe certain colors, the results can

vary greatly. But printers need standard-

ized yardsticks for identifying the colors

they use in their work. To meet this need,

various evaluation systems have been

developed. Some ink manufacturers cre-

ate sample books and give each color in

them a unique name, such as Novavit

4F 434.

Others use color gamuts, like HKS and

Pantone. Color circles divided into 6, 12,

24 or more segments are also used.

Together, the two inks subtract two-

thirds of the color components from

the white light.

If cyan, magenta and yellow are all over-

printed, all of the light striking the sur-

face is absorbed – so none is ref lected.

As a result, we see black.

1.3.3 Autotypical Color Synthesis

Color images are printed using a four-

color process with cyan, magenta, yel-

low and black inks. The black improves

the definition and contrast of images.

The black that is produced by subtracti-

vely combining cyan, magenta and yel-

low is, because of the nature of the pig-

ments used in the inks, never com-

pletely pitch-black.

In classical offset printing, the halftone

dots are sized depending on the desired

color tone (see section 2.2). When over-

printed, some of the dots corresponding

to the individual colors are adjacent to

one another, while others partially or

entirely overlap. If we look at the dots

through a magnifying glass (see figure),

we see colors that – with the exception

of the paper's white – result from sub-

tractive color mixing. Without a magni-

fying glass and when looking at an off-

set-printed item from the normal view-

ing distance, our eyes are unable to

distinguish the individual dots. In this

case, the colors are additively combi-

ned.

A combination of additive and subtrac-

tive color reproduction is known as

autotypical color synthesis.


If we imagine that the primary colors are

the axes of a three-dimensional system

of coordinates, what we get is a so-called

color space.

Many experts have tackled the problem

of how to systematically organize colors,

coming up with differing ideas on how a

color space should be structured. All of

the color spaces defined so far have ad-

vantages and disadvantages.

The most important color spaces have

been standardized internationally. They

are used in a wide range of industries:

production of inks and coatings, textiles,

food production and medicine, to name

just a few. The CIE chromaticity diagram

has prevailed as the most widely used

standard (the acronym CIE stands for

“Commision Internationale

de l'Eclairage”).

This system uses the letters X, Y and Z

instead of R, G and B to designate the

axes. For practical reasons, reference is

usually made to the chromatic values x

and y and the lightness value Y (used as

a measure of brightness for body col-

ors). A color's location within the space

can be precisely defined using these

three coordinates.

Colors with the same lightness value

can be depicted two-dimensionally in

a plane. If the CIE color space is sliced

open along a lightness plane, what re-

sults is the CIE standard chromaticity

diagram (the “tongue”, see figure above).

The spectral colors are the ones with the

greatest saturation reproducible in a

given tone (wavelength). They are at the

edges of the CIE standard color system.

In the figure, their wavelengths are

given in nanometers. The straight line

connecting the wavelengths of 380 and

All these systems use samples or speci-

mens to show the individual color tones

and assign names to them. However, they

are never exhaustive and are rarely suit-

able for making calculations. As we have

seen, our color perceptions depend on

how the red-, green- and blue-sensitive

receptors in our eyes are stimulated.

This indicates that three parameters are

needed to unambiguously describe the

set of all possible colors.

In such a system, green could be des-

cribed as follows:

Green = 0 × red + 1 × green + 0 × blue

Or even more concisely:

G = 0 × R + 1 × G + 0 × B

0.8

0.7

0.6

0.5

0.4

0.3

0.2

0.1

0.0

0.0 0.1 0.2 0.3 0.4 0.5 0.6 0.7

Visually perceivable colors

in a lightness plane of the

CIE color space (the stan-

dard chromaticity diagram,

which resembles a sail,

tongue or sole of a shoe).


780 nanometers is called the “purple

line”. The area bounded by the spectral

locus and the purple line contains all

color valences that can be created by

mixing spectral colors.

The (white) center point has the coordi-

nates x = 0.333 and y = 0.333. With pri-

mary light sources, it is indicated by an

E (for “energy-equivalent spectrum”)

and with body colors occasionally by an

A (for “achromatic”).

The saturation of every color decreases

from the center point toward the spec-

tral locus.

The Euroscale (DIN 16539) defines the

coordinates of the colors cyan, magenta

and yellow for three- and four-color

printing. Also defined are the locations

of the subtractive secondary colors red,

green and blue.

The standard color diagram illustrated

here shows the color locations defined

by DIN 16539 and the set of printable

colors.

The distribution of lightness values is

very similar. All the colors located

within the hexagon can be reproduced

in the four-color process using the

Euroscale. Colors outside this zone can

only be reproduced by adding special

colors.

The Euroscale specifies the following

values for art paper under defined

printing and measurement conditions:

The x, y and Y parameters are deter-

mined using a spectrophotometer or

tristimulus colorimeter. These are avail-

able as handheld units and central sta-

tions with online color control (for in-

stance, in Prinect® Axis Control® and

Prinect® Image Control from

Heidelberg®).

Colors reproducible

with the Euroscale

(DIN 16539).

0.8

0.7

0.6

0.5

0.4

0.3

0.2

0.1

0.0

0.0 0.1 0.2 0.3 0.4 0.5 0.6 0.7

Primary and Proportions of Lightness secondary colors standard colors value

x y Y

Yellow 0.437 0.494 77.8

Magenta 0.464 0.232 17.1

Cyan 0.153 0.196 21.9

Yellow-Magenta 0.613 0.324 16.3

Yellow-Cyan 0.194 0.526 16.5

Magenta-Cyan 0.179 0.101 2.8

10 Expert Guide on Color & Quality Color in Printing

Color in Printing2

The goal of quality control in printing is

to correctly and consistently reproduce

colors through the pressrun. Various fac-

tors affect this; besides the inks and the

color shade of the substrate, the most im-

portant parameters are the thickness

of the ink films, tonal values, color bal-

ance, ink trapping and color sequence.

2.1 Ink Film Thickness

In offset printing, for process-related

technical reasons the maximum ink

film thickness that can be laid down is

about 3.5 micrometers.

When printing Euroscale colors (as de-

fined in DIN 16539) on art paper, it is

advisable to achieve the correct color

locations with film thicknesses be-

tween 0.7 and 1.1 micrometers. The use

of unsuitable plates, substrates or inks

can prevent the standardized corner

points of the CIE chromaticity diagram

from being reached.

Less-than-optimum saturation also re-

stricts the range of reproducible colors.

In the figure, white is used to show how

insufficient saturation of all three chro-

matic colors reduces the range.

In terms of physics, the ink film thick-

ness inf luences appearance as follows:

Printing inks are translucent, not

opaque. This means that light pene-

trates them. While doing so, it strikes

particles of pigment that absorb a fairly

large share of certain light wavelengths.

Depending on the pigment concentra-

tion and the ink film thickness, the

light can strike more or less pigment,

resulting in different amounts of light

being absorbed. The light rays ultimate-

ly reach the (white) surface of the sub-

strate and are ref lected by it back

through the ink to the observer’s eyes.

2.2 Tonal Value

After the ink film thickness, halftone (or

tonal) value is the most important factor

affecting the visual appearance of a col-

or nuance. In reference to a film or digi-

tal image file, the tonal value is the share

of an area covered by halftone dots.

Brighter colors have smaller tonal values.

To reproduce different color nuances,

the conventional approach is to keep the

screen ruling (also known as screen fre-

0.8

0.7

0.6

0.5

0.4

0.3

0.2

0.1

0.0

0.0 0.1 0.2 0.3 0.4 0.5 0.6 0.7

quency) constant while varying the size

of the halftone dots as required to ob-

tain the desired tone. In frequency-mo-

dulated screening, by contrast, the dots

stay the same size while the screen rul-

ing changes. Tonal values are normally

given as percentages.

Color in Printing Expert Guide on Color & Quality 11

2.2.1 Changes in Tonal Values

When halftone dots are transferred

from film via a plate and blanket to pa-

per, various factors can affect their size

and shape, which has repercussions on

the tonal value.

Process-related changes to tonal values

(see section 2.2.3) can be compensated

for in prepress. Print samples are mea-

sured and compared with the originals,

which lets transfer curves be plotted.

Provided that the same standards are

consistently applied throughout the

process chain from the scanner to the

finished print product, true-to-original

results can be expected.

Changes to tonal values caused by print-

ing problems are unpredictable. Special

attention therefore has to be paid to

them. Here are the most important

ones:

The path of a halftone dot Factors influencing halftone dots Appearance of halftone dots

Film Film edges, adhesives

Assembly

Camerawork

Development Chemicals, development times

Two halftone dots on film

(magnified approx. 150x)

Plate Materials, wear during printing

Platemaking Exposure time, vacuum,

undercutting

Dampening Amount of dampening solution,

pH, surface tension,

water hardness, temperature

Inking Ink film thickness, consistency,

temperature

Printing Cylinder rolling Halftone dots on the plate after

inking


Dot Gain and Sharpening

• Dot gain

When halftone dots grow in size re-

lative to the film or digital image, it

is called “dot gain” or occasionally

also “dot spread”. This can be caused

in part by the printing process, ma-

terials or equipment, factors that

are relatively difficult for the ope-

rator to inf luence, and in part by

the inking, which the operator can

manipulate.

• Fill-in

Fill-in is a problem similar to dot

gain that is caused by printing ink in

the non-image areas between the

dots, narrowing the spaces until they

disappear entirely. Slurring and

ghosting can sometimes be respon-

sible for fill-in.

• Sharpening

Sharpening refers to a decrease in

the tonal value as compared to the

film or digital image. In practice, the

term is always used to describe a re-

duction in dot gain, even when the

dots are still fuller than on the film

or in the digital image.

The path of a halftone dot Factors influencing halftone dots Appearance of halftone dots

Blanket Material, condition, surface

Printing Cylinder rolling

Blanket/paper

The dots on the blanket.

Paper Surface, paper grade

Sheet transport Transfer register

Delivery Smearing

High magnification clearly shows

the first-class results on paper.

Right Wrong

• Smearing

When mechanical factors in the press

cause the deformation of halftone

dots, it is known as smearing. The

term is also used as a synonym for

offsetting.

What the operator has to pay attention to

Dot gain and its extent can be monitored

visually and with the aid of instruments.

Control strips include special patches

that are excellently suited for visually

detecting dot gain. Sharpening can be

easily monitored using measurement

targets with a high tonal value.


Halftone Dot Deformation

• Slurring

Slurring is when the shape of a half-

tone dot is distorted during printing

by relative motion between the plate

and blanket and/or blanket and sheet.

For example, a round dot can be

stretched to an oval shape. Slurring

in the direction of printing is called

circumferential slurring, and perpen-

dicular to that it is known as lateral

slurring. If both types occur to-

gether, the direction of slurring is

diagonal.

• Ghosting

In the context of offset printing,

ghosting is when a second, typically

smaller, shadow-like ink dot is unin-

tentionally printed next to the in-

tended dot. Ghosting is caused by ink

being transferred back to the blanket

out of register.

Dot gain Sharpening Slurring Ghosting Smearing

Right Wrong

Both dot gain and fill-in are usually

caused by excessively heavy inking, in-

sufficient dampening solution feed, too

much pressure between the plate and

blanket cylinders, or inadequate blanket

tension. Sometimes it can also be due to

incorrect setting of the inking and dam-

pening form rollers.

Even under normal conditions with

correctly made plates, a certain amount

of dot gain occurs. Sharpening can oc-

cur under abnormal conditions such as

plate blinding or ink accumulation on

the blanket. This can be prevented by

washing the blankets and inking units

more frequently, possibly changing the

inks and sequence of colors, and check-

ing the form rollers and cylinder pres-

sure settings.

Slurring is most conspicuous in patterns

with parallel lines. In many cases, this

also reveals the direction of slurring.

Circumferential slurring usually indi-

cates that the plate and blanket are slip-

ping slightly relative to one another as

they turn, or that the cylinders are press-

ing too hard against one another. So it’s

very important to check the printing

pressure and cylinder rolling. Frequent-

ly, the culprit can also be a blanket that

isn’t clamped tightly enough, or exces-

sively heavy inking. Lateral slurring

rarely occurs by itself. If it does, pay spe-

cial attention to the substrate and the

blanket.

Right Wrong

Right Wrong

While the coarse-screened background

has a uniform tonal value, the numerals

0 to 9 have a fine screen ruling and an

increasing tonal value. On a well-printed

sheet, the number 3 and the coarse

screened patch have the same tonal

value and the number is invisible. With

increasing dot gain, the next-higher

number disappears instead. The fuller

the printed dots get, the higher the

value of the invisible number.

This works in reverse when sharpening

occurs. Then the number 2, 1 or even 0

becomes illegible. However, the numer-

als only indicate that printing is getting

fuller or leaner. The causes must be as-

certained by examining the plate with a

magnifying glass or checking the press.


The same methods are used to check for

ghosting and slurring. A magnifying

glass should also be used to inspect the

halftone dots, because line patterns can-

not reveal whether ghosting or slurring

has occurred. Ghosting can have many

possible causes, but it is usually due to

problems with the substrate or some-

thing directly related to it.

Smearing is extremely rare in today’s

modern printing presses. If it occurs,

the parts of a sheetfed press in which

sheets are mechanically supported

on the freshly printed side should be

checked first. Stiff substrates increase

the risk of smearing. Smearing can

also occur in the delivery pile and in

perfector presses.

Printed control elements like the SLUR

strip let you quickly identify the type of

dot distortion involved. These elements

visually amplify the printing problem

so it can be easily seen.

Problems like dot gain and sharpening,

slurring and ghosting are worse with

fine screens than with coarse screen rul-

ings. The reason is that fine halftone

dots increase or decrease in size by the

same amount — i.e. in absolute, not rela-

tive terms — as larger ones. However,

many small dots together have a total

length several times that of large dots

with the same tonal value. Consequent-

ly, more ink is used to print fine dots

than large ones. Areas with fine screen

rulings therefore appear to be darker.

Control and measurement targets take

advantage of this fact.

To illustrate this, let’s look at the struc-

ture and functions of the SLUR strip (see

figure below). This strip contains both

coarse-screen and fine-screen patches.

Right Wrong

Right Wrong


The part of the SLUR strip to the right of

the numerals mainly shows whether

slurring or ghosting has occurred. The

word SLUR is equally legible with lean,

normal and full printing; the whole

patch merely appears somewhat lighter

or darker.

It is easy to detect the directional spread

typical of slurring and ghosting in the

word SLUR, however. In the case of cir-

cumferential slurring, for example, the

horizontal lines forming the word

SLUR, which run parallel to the sheet’s

leading edge, become thicker. If lateral

slurring has occurred, then the vertical

lines forming the background of the

word SLUR appear darker.

The figure to the right illustrates how

changes in the halftone dots affect

printing, specifically when there is dot

gain. If the dots for just one color are

larger than they should be, this results

in a new shade — which naturally also

inf luences the overall appearance of

the printed image. In offset printing,

the need to transfer images from the

plate to the blanket and from there to

the paper usually results in a certain

amount of dot gain.

Color control strips can tell you whether

the results of printing are good or bad,

but they cannot provide any absolute

figures or indicate the exact nature of

the problem. An objective method is

therefore needed for assessing quality

by measuring tonal values.

Good

Fuller

Leaner

Lateral

slurring

Circum-

ferential

slurring


Like the tonal value (F), the dot gain (Z)

is normally given as a percentage (see

section 3.5 for the formulas used to cal-

culate it). It is a function of the differ-

ence between the measured tonal value

in print (FD) and the tonal value in the

film (FF) or the data. Because the dot

gain can vary depending on the tonal

value, when making statements on dot

gain it is important to also provide the

tonal value in the film. For example: 15%

dot gain with FF = 40%, or abbreviated

as Z40 = 15%. Modern measuring instru-

ments directly indicate the dot gain.

Note: The measured dot gain Z shows

the difference between the tonal value

in print (FD) and the tonal value in the

film (FF) or the data as an absolute value.

In other words, it is independent of the

film or data value.

2.2.3 Characteristic Curves

The deviation of the tonal value in print

(FD) from the tonal value in the film (FF)

or data can be clearly represented in a

“print characteristic curve”, which can

then be directly used to optimize re-

production quality.

To determine a characteristic curve,

print a step wedge with at least three

but preferably five or more tonal levels

and one full-tone (solid) patch. Use a

densitometer or spectrophotometer to

measure all of the levels, and calculate

their tonal values. Plot the obtained val-

ues on a chart against the correspond-

ing film values; the result is a “transfer

characteristic curve”. With standardized

platemaking, it is identical to the print

characteristic curve.

This curve only applies to the same com-

bination of ink, paper, cylinder pressure,

blanket and plate for which it has been

determined. If the same job is printed on

another press with different ink or on

different paper, the print characteristic

curve may differ somewhat.

In Figure 17, characteristic curve 1 is a

straight line running at an angle of 45

degrees. This line is not normally at-

tainable; it represents the ideal state in

which the print and the film are visually

indistinguishable. Characteristic curve

2 represents the tonal values actually

measured in the print. The area be-

tween the two curves is the dot gain.

The midtones are most useful for deter-

mining dot gain in print. In curve 2, it

is plain that the tonal value deviations

are greatest in that range. This charac-

teristic curve can be used to adjust the

screened film while achieving the de-

sired tonal values in print (with the

usual dot gain).

In practice, however, process-related

f luctuations will inevitably result in

minor deviations. Because of this,

tolerances are always given for the dot

gain. To keep the print quality as con-

stant as possible, it is indispensable to

continually check the tonal values in a

color control strip and with the aid of

Mini Spots® from Heidelberg.

Right Wrong

2.2.2 Dot Gain

Dot gain is the difference between the

tonal values of a screened film or digital

image on the one hand and the print on

the other. Differences can result from

(1) changes in the halftone dots or (2) the

phenomenon known as the “light trap

effect” or light gathering (see section

3.4.4).


2.3 Relative Print Contrast

As an alternative to dot gain, sometimes

the relative print contrast Krel (%) is de-

termined, mainly for monitoring the

three-quarter tones.

A print should be as contrast-rich as

possible. To achieve this, the full tones

should have a high color density but

the screen should be printed as open as

possible (with an optimum tonal value

difference). Increasing the ink feed,

resulting in a greater color density of

the halftone dots, enhances the con-

trast. But there is a limit to how far this

can be taken — too much, and the half-

tone dots will grow too full and start

filling in, especially in the shadows.

This reduces the share of paper white

and the contrast declines again.

If no measuring instrument is available

that gives a direct reading of the con-

trast, an alternative is to calculate the

relative contrast (the formulas are given

in section 3.5.3) or determine it with the

aid of the corresponding FOGRA chart.

If the contrast gets worse during the

course of a production run even though

the solid density remains constant, this

can mean that it is time to wash the

blankets. If the solid density is correct,

then the contrast value can be used to

assess various other factors that affect

the results of printing, for example:

• Cylinder pressure and rolling

• Blankets and packing

• Dampening

• Inks and additives

Because the relative print contrast,

unlike dot gain, greatly depends on the

momentary solid density, it is unsuita-

ble for use as a standardization param-

eter. In recent years its importance has

greatly diminished.

Print Print characteristic curve

Characteristic curve 2

Characteristic curve 1

DV = 1.50

Film or data

Film

Print

2.4.1 Chromatic Composition

In this approach, all achromatic shades

are created by mixing the chromatic

colors, i.e., cyan (C), magenta (M) and

yellow (Y). In other words, all gray areas

in the image, all tertiary tones, and

the shadows contain all three chromatic

process colors. Black (K) is only used to

enhance the shadows and improve

image definition there (skeleton black).

The brown shown in the figure consists

of 70% cyan, 80% magenta, 90% yellow

and 0% black. The total area coverage is

therefore 240%.

How the color components work in

combination is illustrated on the right.

The brown consists of an achromatic

(gray) portion and a chromatic portion.

70% cyan and about 58% magenta and

59% yellow (in the Euroscale) offset

one another to yield gray (an achromat-

ic color). Only the other 22% magenta

and 31% yellow combine to produce a

light-brown chromatic color. Together

with the gray portion, this yields dark

brown.


Chromatic composition results in high

total area coverage, which can theoreti-

cally reach 400%, but in practice does

not exceed 375%. These high total area

coverage levels adversely affect ink

trapping, drying and powder consump-

tion, and it is difficult to maintain the

color balance during the pressrun.

2.4 Color Balance/Color Composition

As already explained, in the four-color

process different color shades are re-

produced by mixing varying amounts

of cyan, magenta, yellow and black.

As soon as their relative proportions

change, so does the color. To prevent

this from happening, they must some-

how be kept in the right balance.

If it is only the proportion of black that

changes, the color gets lighter or darker,

which does not irritate the observer

much. The same thing happens if all

three chromatic colors change by the

same amount in the same direction. The

situation is much more critical when

the color tone itself changes. This hap-

pens when the color components

change by different amounts, and espe-

cially if the individual chromatic colors

change in opposite directions. These

kinds of changes in the color balance

are easiest to detect in gray patches. Re-

ference is therefore often made to gray

balance in this connection.

80 % M70 % C 90 % Y 240 % 0 % K

100 %

50 %

0 %C M Y K

100 %

50 %

0 %C M Y K

The extent to which the inevitable

f luctuations in each color printed

affect the results mainly depends

on the color composition approach

defined in prepress. The relevant

questions are:

• Which process colors do the gray

areas consist of ?

• What approach is used to darken

chromatic areas?

• How are the shadows created and

enhanced?

In short: how are the gray, i.e.

achromatic, parts generated, and

what is the maximum total area

coverage that results? Remember,

gray (achromatic) shades can be

produced either by combining cyan,

magenta and yellow, or by using

process black. It is also possible to

combine both approaches.


2.4.2 Achromatic Composition

In contrast to chromatic composition,

with achromatic composition all of the

achromatic colors in a multicolor image

are produced with process black. In

other words, all neutral colors consist

only of black, and black is also used to

darken chromatic colors and achieve

greater saturation. Any given color con-

sists of a maximum of two chromatic

process colors plus black. This stabilizes

the color balance. With achromatic

composition, in theory the brown dis-

cussed in section 2.4.1 can be produced

by overprinting 0% C + 22% M + 31% Y +

70% K.

However, as the figure shows, merely

replacing an achromatic shade pro-

duced with CMY by black does not yield

an identical color.

This is primarily due to the shortcom-

ings of actual printing inks. To obtain

truly similar results, it is necessary to

modify the proportions, e.g. to 62% M,

80% Y and 67% K. Achromatic composi-

tion is equivalent to 100% gray compo-

nent replacement (GCR; see section

2.4.6 below).

2.4.3 Achromatic Composition with

Under Color Addition (UCA)

Process black by itself does not always

provide sufficient definition in the dar-

ker portion of the gray axis. When this

is the case, this range and, to a lesser

extent, the neighboring chromatic

tones can be enhanced by adding CMY.

Use of this approach, called “under

color addition” (UCA) or “chromatic

color addition”, depends mainly on the

substrate-ink combination. The illus-

tration on the right illustrates UCA to

neutrally enhance the image shadows.

2.4.4 Chromatic Composition with

Under Color Removal (UCR)

The highest area coverages result from

using chromatic composition for the

neutral three-quarter tones all the way

to black. This drawback is offset by

“under color removal” (UCR). The pro-

portion of CMY is reduced in the neutral

shadows and, to a lesser extent, in the

neighboring chromatic tones, while the

amount of process black is increased.

The example below, the initial area

coverage of 98% cyan + 96% magenta +

87% yellow + 84% black = 355% is

reduced by 78% with UCR. This

favorably affects ink trapping, drying

and balance in the shadows.

22 % M0 % C 31 % Y 123 % 70 % K

100 %

50 %

0 %C M Y K

100 %

50 %

0 %C M Y K

100 %

50 %

0 %C M Y K


2.4.5 Chromatic Composition with Gray

Stabilization

Gray shades created with chromatic

composition are hard to keep balanced

in the print process. Color casts readily

occur. This can be prevented by gray

stabilization. Achromatic components

generated with C + M + Y are partially or

entirely replaced along the entire gray

axis and to a lesser extent in the neigh-

boring color ranges — i.e., not just at the

darker end of the gray axis like with

UCR —by an equivalent amount of black.

This is often referred to as “long black”.

2.4.6 Chromatic Composition with Gray

Component Replacement (GCR)

“Gray component replacement” (GCR)

involves using process black to replace

CMY components in both chromatic and

neutral image areas. GCR can be used

for all intermediate stages between

chromatic and achromatic composition

in all image areas — and is not, like UCR,

UCA and gray stabilization, limited to

the gray areas. Gray component replace-

ment is sometimes also referred to as

complementary color reduction.

The brown in sections 2.4.1 and 2.4.2,

for example, could theoretically be

produced as follows with GCR:

Like with achromatic composition (sec-

tion 2.4.2), the colors obtained with the

two methods are not identical if black is

merely substituted for part of the achro-

matic CMY without adjusting the chro-

matic components as well. Similar

colors are achieved with, for example,

49% C + 70% M + 80% Y + 30% K.

100 %

50 %

0 %C M Y K

60 % M50 % C 70 % Y 200 % 20 % K

0.8

0.7

0.6

0.5

0.4

0.3

0.2

0.1

0.0

0.0 0.1 0.2 0.3 0.4 0.5 0.6 0.7

2.4.7 Five-, Six- and Seven-Color

Printing

The modern four-color process ensures

high-quality image reproduction. How-

ever, with some originals and when ex-

tremely high quality is needed, it can be

necessary to use additional special col-

ors. The use of additional colors (along-

side the four process colors) or special

process colors can extend the range of


ink that is still wet, then one speaks of

“wet on wet” printing. With multicolor

presses, it has become standard to talk

about wet on wet printing.

When inking is uniform and the colors

are accurate, this indicates that there is

good ink trapping.

In contrast, if the target color cannot

be achieved, then the ink trapping is

inadequate. This can be the case with

every tone involving overprinting of

two or more process colors. This

restricts the printable range of colors,

and certain color nuances cannot be

reproduced.

Even if the right ink film thicknesses

are printed with a given set of colors

and the primary colors cyan, magenta

and yellow are accurate, it can still hap-

pen that the secondary colors red, green

and blue can still be poor, due to over-

printing problems.

The CIE chromaticity diagram above

shows the effects of disturbed ink trap-

ping or an unfavorable color sequence

on the printed result. The white area

corresponds to the extent of the tonal

reduction caused by the ink trapping

problems.

reproducible colors. The previous fig-

ure shows the measured values for a

seven-color print entered in the CIE

chromaticity diagram.

The hexagon on the inside shows the

color gamut reproducible with the pro-

cess colors cyan, magenta and yellow

(as measured). The surrounding dod-

ecagon shows the extended color space

that can be printed with the additional

colors green (G), red (R) and blue (B).

2.5 Ink Trapping and Color Sequence

2.5.1 Ink Trapping

Another parameter that inf luences

color reproduction is ink trapping. This

is a measure of an ink’s ability to trans-

fer equally well to unprinted substrate

and a previously printed ink film. Two

different cases occur: wet on dry, and

wet on wet.

Wet on dry printing is when an ink is

laid down directly on the substrate or

onto a previously printed and dried ink

film. If the second color is printed on

0.8

0.7

0.6

0.5

0.4

0.3

0.2

0.1

0.0

0.0 0.1 0.2 0.3 0.4 0.5 0.6 0.7


2.5.2 Color Sequence

The schematic diagram shows three dif-

ferent overprints of the colors cyan and

magenta.

The second example was printed on a

multicolor press. First magenta was

printed on the dry paper (wet on dry),

then cyan on top of the still-moist ma-

genta (wet on wet). Whereas the magen-

ta was accepted well by the paper, the

cyan was accepted less well (due to the

ink splitting that occurred during over-

printing). This caused the resulting

blue to have a red cast.

The third example was also printed wet

on wet, but in the reverse order (magen-

ta on cyan). The result was red with a

blue cast.

In the four-color process, the color se-

quence black – cyan – magenta – yellow

has prevailed as the standard.

In order to reduce the effects of ink

trapping problems in special cases, the

original and the plates should be care-

fully inspected before mounting the lat-

ter on the press. When there are solid

areas, it can be advantageous to print

the lighter form before the heavier one.

This applies especially when overprint-

ing screened areas and solids. The

screened areas should be printed first

on the white paper, then the solids.

2.6 Color Control Strips

So that the print quality can be assessed

by performing measurements, color

control strips are included in the print-

ed sheets. These are available from vari-

ous research institutions and suppliers.

It is important to always use the original

strips, because copying them onto dup-

licate films results in deviations that

can falsify the results of measurements.

Color control strips are available for

four- to eight-color presses. Strips for

more than four colors have fewer tint

patches and patches for detecting slur,

and more of the elements that are need-

ed to adjust the solids and color balance.


All color control strips have multiple

elements. In the following, the most

important elements of the Heidelberg

CPC color strip and those from FOGRA

and Brunner are illustrated.

2.6.1 Solid Patches

Solid patches are used to check the

uniformity of inking. It is expedient

to use one solid patch for each ink

printed, spaced to correspond to the

width of the ink zones (in the case of

Heidelberg, 32.5 millimeters). The

solid patches can then be used for

automatic regulation of the solids.

2.6.2 Solid Overprint Patches

These elements are used to assess ink

trapping by means of visual inspection

and measurements.

K C M Y

M Y C Y C M

2.6.3 Color Balance Patches

There are solid and tint color balance

patches.

When the colors cyan, magenta and

yellow are overprinted in a solid patch,

the result should be a fairly neutral

black. For purposes of comparison, a

solid black patch is printed alongside

the overprint patch.

With the correct ink film thicknesses,

the standard color sequence, and nor-

mal dot gain, the tint patches for cyan,

magenta and yellow should yield a fairly

neutral gray when overprinted.

Color balance patches are intended to

be visually checked; they are also used

for automatic gray balance control for

the colors cyan, magenta and yellow.

In the standardized process as described

by ISO 12647-2 (identical with the stan-

dard offset process), the proper gray ba-

lance must be mainly achieved by apply-

ing an ICC color profile to generate the

separations.

2.6.4 Tint Patches

The tonal values of the tint patches on

film vary depending on the manufac-

turer.

The values measured in the tint and

solid patches are used to calculate the

dot gain and relative print contrast.

2.6.5 Slur and Ghosting Patches

Line screens with different angles are

used to check for slurring and ghosting

by visual inspection and measurement

(see section 2.2.1).

C M Y K

K C M Y

2.6.6 Platemaking Elements

Platemaking elements are used to visu-

ally check the results of platemaking.

The elements shown have microlines

and microcolumns as well as fine dots.

Today the FOGRA color control strips with 40%

and 80% patches are most widely used.

24 Expert Guide on Color & Quality Densitometry

Densitometry3

3.1 Reflection Densitometry

In ref lection densitometry, the color to

be measured is illuminated by a light

source. The light beam penetrates the

translucent ink film, which attenuates

it. The remaining portion of the light is

greatly scattered by the paper under-

neath. Part of this scattered light is re-

f lected back through the ink film, being

further attenuated in the process. What

is left then finally reaches a sensor,

which converts the light into an electri-

cal signal. The result is indicated in den-

sity units.

Lens systems are used to focus the light

to facilitate measurement. Polarizing

filters suppress the wet gloss (see sec-

tion 3.2.2). When measuring chromatic

colors, color filters are placed in front

of the sensor (see section 3.2.1).

Densitometry is an effective method for

monitoring solid density and tonal val-

ues in the print process. It works reli-

ably with black-and-white reproduc-

tions and with the process colors cyan,

magenta, yellow and black.

There are two types of densitometers:

• Transmission densitometers are used

to measure film blackening (i.e.,

with transparent materials).

• Ref lection densitometers are used

to measure light ref lected from the

surface of a print (i.e., with ref lec-

tive originals).

The technology of ref lection densito-

metry is described in detail below.

Transmission densitometer

Reflection densitometer

Densitometry Expert Guide on Color & Quality 25

The figure shows how ref lection densi-

tometry works, taking the example of a

printed chromatic color. White light —

ideally consisting of equal parts of red,

green and blue — shines onto the press

sheet. The printed ink contains pig-

ments that absorb red and ref lect green

and blue, which is why we call it cyan.

The densitometer measures the absorb-

ed light of a certain color, because den-

sity and ink film thickness correlate

well. In this example, a red filter is

therefore used; it filters out blue and

green and only allows red to pass.

The density of a printed ink primarily

depends on the type of pigment it con-

tains, its concentration, and the ink

film thickness. The density of a printed

ink reveals the film thickness but tells

us nothing about the color itself.

Color filter

Polarizing filter

Color filter

Polarizing filter

Polarizing filter

Lens systemPaper


3.2 Densitometer Filters

3.2.1 Color and Brightness Filters

The color filters in a densitometer are

optimized for the absorbed wavelengths

corresponding to cyan, magenta and

yellow.

The relevant standards, such as

DIN 16536 and ISO/ANSI 5/3, stipulate

the spectral passbands and the wave-

lengths of the pass maxima. They define

narrow- and wideband color filters (in

the case of ANSI, designated A and T,

respectively); narrowband filters (DIN

NB) are preferable because different

spectrometer makes deliver more con-

sistent measurement values with them.

Always choose a color filter that is the

polar opposite of the colors being mea-

sured. Black is evaluated with a visual

filter that is adjusted to the spectral

brightness sensitivity of the human eye.

Special colors are measured with the

filter that yields the highest measure-

ment value.

The three figures on the right show the

ref lection curves for cyan, magenta and

yellow using the corresponding color

filters as defined by DIN 16536.

Printed color Filter color

Cyan Red

Magenta Green

Yellow Blue

Cyan

Magenta

Yellow

1.0

0.5

0.0

1.0

0.5

0.0

1.0

0.5

0.0


3.2.2 Polarizing Filters

When press sheets are pulled freshly

printed from the delivery and mea-

sured, the ink is still wet and has a shiny

surface. While drying, the ink pene-

trates into the paper (absorption) and

loses its gloss. This changes not only

the color’s tone but also its density. If

the press operator wishes to use densi-

tometry to compare the wet sheets with

the reference values, which also refer

to dry ink, the results will inevitably

be wrong.

To make this possible, two linear

polarizing filters at right angles to one

another are placed in the path of the

densitometer. Polarizing filters only per-

mit light waves oscillating in a certain

direction to pass. Part of the resulting

aligned beam of light is ref lected by the

surface of the ink but its direction of

oscillation remains unchanged. The sec-

ond polarizing filter is rotated perpen-

dicular to the first, which blocks out

these ref lected light waves.

However, if the light isn’t ref lected until

after it penetrates into the ink film, ei-

ther by the ink or the paper, it loses its

uniformly aligned direction of oscilla-

tion (polarization). Consequently, the

second polarizing filter allows part of it

to pass and strike the sensor.

Filtering out the light ref lected by the

ink’s glossy surface has the effect of

making the densitometric measure-

ment values for wet and dry ink roughly

equivalent.

However, due to absorption by the pola-

rizing filter less ref lected light reaches

the sensor. Consequently, the values

measured with instruments of this kind

are usually higher than with other appa-

ratus, depending on the gloss.

Paper

Direction of scattering

Direction of oscillation

3.3 Densitometric Values

The result of measurement with a den-

sitometer is a logarithmic number: den-

sity (D). This is expressed as the loga-

rithm (base 10) of the opacity, which is

the reciprocal of the transmission or

ref lection of a tone.

Density is calculated by applying the fol-

lowing formula:

The ref lectance (also called the beta

value) is calculated as follows:

D = lg1

β

=LeP

=50 %

= 0.5LeW 100 %

LePLeW

=LeP

LeW

LeP is the light ref lected by the printed

ink, and LeW is the light ref lected by

the reference white.

The ref lectance (β) indicates the ratio

between the light ref lected by a sample

(the printed ink) and a standard white

(reference value).

The value β yields the following density:

D = lg1

= lg1

= lg2 = 0.30β 0.5

β

β


There is a close correlation between the

ink film thickness and the ink density.

The diagram shows that ref lection di-

minishes and density increases with

thicker ink films.

Please see page 27 for the calculation

formulae.

The diagram shows how the ink film

thickness and density correlate for the

four process colors used in offset print-

ing.

The dotted vertical line indicates the

usual ink film thickness in offset print-

ing, or about one micrometer. The dia-

gram shows that the density curves do

not f latten until considerably higher

values are reached. Then, at even higher

thicknesses the density barely increases

any further. Even if you were to mea-

sure a full can of ink, the value obtained

would be only negligibly higher. Of

course, ink films that thick have no rele-

vance to the standard four-color process.

Black

Cyan

Magenta

Yellow

Ink film thickness

Den

sity

2.5

2.0

1.5

1.0

0.5

0.0

0 0.5 1.0 1.5 2.0 2.5

3.4 Measurement

3.4.1 Calibration to Paper White

Before any measurements are made, the

densitometer is calibrated to the applica-

ble paper white (reference white) in or-

der to eliminate the inf luence of paper

coloring and the paper’s surface when

assessing the printed ink film thickness.

To accomplish this, the density of the pa-

per white is measured in reference to

“absolute white” and this value then set

to zero (the reading is D = 0.00).


3.4.2 Solid Density

The values measured in a solid area indi-

cate the solid density. It is measured in a

color control strip that is printed on the

sheet perpendicular to the direction of

travel and has a number of patches, in-

cluding solid patches for all four process

colors (and special colors if required).

The solid density can be used to monitor

and ensure a uniform ink film thickness

across the entire width of the sheet and

throughout the pressrun (within cer-

tain tolerances).

3.4.3 Halftone Density

Halftone density is measured in the tint

patches of the color control strip. These

round patches, which are typically three

to four millimeters wide, contain a mix

of halftone dots and paper white, corre-

sponding to the inner structure of the

human eye.

The measured value is the halftone den-

sity. This value increases with the tonal

value of the halftone dots and the ink

film thickness.

Paper

3.4.4 Optically Effective Area Coverage

(Tonal Value)

When using densitometry to measure

halftone images, it is not the geometri-

cal area coverage (the percentage of the

patch’s surface covered by halftone

dots) that is measured but rather the

“optically effective area coverage”.

The difference between the geometrical

and optically effective area coverage is

that, regardless of whether they are

assessed by a visual check or measure-

ment with a densitometer, part of the

light shining onto the sheet penetrates

into the paper in the blank areas be-

tween the halftone dots, and part of

what is ref lected strikes the rear of the

dots and is absorbed by them.

This effect is known as “light trapping”.

It makes the halftone dots appear larger

than they actually are. The optically

effective area coverage thus consists of

the geometric area coverage plus the

optical magnification effect.


3.5 Evaluation

The values measured for the solids and

halftones can then be used to calculate

the tonal values, dot gain and contrast.

A prerequisite for doing this is that the

densitometry must be calibrated to the

paper white beforehand.

3.5.1 Tonal Values

The measured solid and halftone densi-

ties (DV and DR) can be used as follows

with the Murray-Davies equation to de-

termine the printed tonal value (FD):

3.5.2 Dot Gain

The dot gain (Z) is the difference be-

tween the measured printed tonal value

(FD) and the known tonal value in the

film (FF) or data.

3.5.3 Relative Print Contrast

The relative print contrast is also calcu-

lated from the measured solid density

(DV) and the halftone density (DR). The

DR value is best measured in the three-

quarter tones.

3.5.4 Ink Trapping

Ink trapping is calculated from the den-

sities measured in single-color solid and

two- and three-color overprint patches,

while taking the color sequence into

account.

The ink trapping calculated with the

following formulae tells us what per-

centage of a color is overprinted on

another. It is compared with the first-

down color, the trapping of which is

assumed to be 100%.

3.5.4.1 Overprinting Two Colors

Where

D1+2 is the density of the over

printed colors,

D1 is the density of the first-down

color, and

D2 is the density of the second-

down color.

Note: All density values must be meas-

ured using the color filter that is dia-

metrically opposite the second color.

3.5.4.2 Overprinting Three Colors

Where

D1+2+3 is the density of all three over

printed colors and

D3 is the density of the last-down

color.

Note: All density values must be mea-

sured using the color filter that is dia-

metrically opposite the last-down color.

The formulas given here are also used

by the quality control systems Prinect

Axis Control and Prinect Image Control

from Heidelberg. Other methods also

exist for determining ink trapping. All

of the methods are controversial, so the

results they give should not be taken too

seriously. However, they are useful for

comparing pressruns with one another

(and especially for comparing sheets

pulled from the same run).

FD (%) = 1–10–DR· 100

1–10–DV

Z (%) = FD–FF

Krel. (%) =DV – DR

· 100DV

FA21

(%) =D1+2– D1

· 10D2

FA3(%) =

D1+ 2 + 3 – D1 + 2

· 100D3

21


Densitometer Colorimeter

Tristimulus colorimeter Spectrophotometer

Mixing of special colors •

Inking setup

• By standards x (•) x • x •

• Using color control strips in test prints x (•) x • x •

• Based on specified values (x) (•) x •

• Using proofs x • x •

• Based on a specimen x • x •

• Based on image data (repro) (x) (•) x •

• Assessment of color suitability (x) (•) x •

Inking adjustment (by comparing) x • x •

Pressrun control

• Based on solid patches x (•) x • x •

• Based on single-color tint patches x (•) x • x •

• Based on multicolor tint patches x • x •

• Based on in-image measurements x • x •

• Detection of ink soiling x • x •

• Detection of changes in substrate x • x •

Measurement values

• Solid density x (•) (x) (•) x •

• Tonal values and increases x (•) (x) (•) x •

• Relative ink trapping x (•) (x) (•) x •

• Absolute ink trapping x • x •

• Metamerism (x) (•) x •

• Subjective impressions x • x •

x = suitable for process colors

• = suitable for process colors

( ) = limited suitability


3.6 The Limits of Densitometry

Similar to the method used to create col-

or separations, densitometers work with

filters geared to the four process colors.

They provide a relative measure of the

ink film thickness, but do not reveal

anything that directly correlates with

human color perception.

This fact limits their usefulness. The

table on page 31 shows their typical

applications as compared to tristimulus

colorimeters and spectrophotometers.

One major constraint on densitometry

is that the same ink densities do not nec-

essarily create the same visual impres-

sion. This is always the case when the

colorants being compared differ, which

is why proofs, test prints on different

paper and/or with different ink than

will be used in the production run, or

other samples cannot serve as reliable

references for setting the inking.

The restriction to red, green and blue

color filters is also significant. As soon

as color sets comprising more than the

four process colors come into play, prob-

lems arise for measuring the additional

colors. Usually no suitable filters are

available for them, which leads to exces-

sively low ink density and incorrect dot

gain values.

The use of densitometers is also difficult

for regulating inking based on multi-

color tint overprint patches (e.g., gray

patches). Measuring a gray patch with

all three color filters yields different ink

densities than if each color were mea-

sured by itself. Each of the three colors

makes a more or less substantial contri-

bution to all ink densities. This is be-

cause the process colors are not gen-

uinely pure primary colors with each

representing two-thirds of the spec-

trum; they also absorb light from other

wavelengths. Densitometers are useful

for monitoring quality in pressruns

using the four-color process. In all

other cases, their suitability is limited.

The color tone shown here (Pantone

Warm Gray 1) has — as can be seen in

the adjacent diagram — relatively high

remission, which drops off slightly in

the blue spectrum (380 to 500 nanome-

ters). So the highest density value (0.27)

is measured with a blue filter.

Color specimen: Pantone Warm Gray 1

1.0

0.5

0.0


The special colors HKS 8 and HKS 65,

shown in the second and third exam-

ples, have radically different tones. This

is also evident in their remission curves.

However, both colors have the greatest

absorption in the blue spectrum (380 to

500 nanometers), for which reason the

highest density value (1.6) is measured

with a blue filter in both cases. This illus-

trates the fact that there is no correla-

tion at all between density values and

color tones.

Only colorimetric measurements can

tell us something about a color’s appear-

ance.

Color specimen: HKS 8

1.0

0.5

0.0

Color specimen: HKS 65

1.0

0.5

0.0

34 Expert Guide on Color & Quality Colorimetry

This stimulation results in electrical sig-

nals being sent via the optical nerve

to the brain, which interprets them as

colors.

This natural process is emulated in col-

orimetric instruments.

To perform a measurement, a printed

sample is illuminated. The ref lected

light passes through one or more lenses

and strikes a sensor. The sensor mea-

sures the received light for each color

and relays the results to a computer.

There the data is weighted using algo-

rithms that simulate the action of the

three types of cones in the human eye.

These algorithms have been defined by

the CIE for a standard observer. They

yield three standardized color values:

X, Y and Z. These are then converted

into coordinates for the CIE chromatic-

ity diagram or some other color space

(e.g., CIELAB or CIELUV).

As explained in the section on color sys-

tems, three parameters are needed to

unambiguously describe a color. Color-

imetry tells us how to obtain these val-

ues and how they are interrelated —

provided, that is, that colors are mea-

surable. So there is a direct connection

between color measurement and col-

orimetry.

4.1 Measuring Colors

Tristimulus colorimeters and spectro-

photometers are used to measure col-

ors. They are described in sections 4.8

and 4.9 below.

The principle of operation of all colori-

metric instruments is based on how hu-

man beings see and perceive color (see

figure).

A color (specimen) is illuminated by a

light source. Part of the light is absorb-

ed by the specimen while the rest is re-

f lected. The ref lected light is what our

eyes register, because it stimulates the

red-, green- and blue-sensitive cones

(color receptors) in the retina.

Colorimetry4Light source

Radiation

Measurementinstrument

Person

Spect

ral r

efle

ctan

ce

Spectral reflectance

Eye

Cones

RedGreenBlue

Stimulation

Color perception Color coordinates

Standard color values

Spectral value

algorithm for standard

observer

Lenses with

sensor

Colorimetry Expert Guide on Color & Quality 35

4.2 Standard Color Values/

Reference White

Before colors can be measured, it is ne-

cessary to determine standard color val-

ues based on measured ref lectance and

emissions under standardized condi-

tions. Most instrument manufacturers

have fixed or applied these so that the

user doesn’t need to worry about them.

However, three factors are usually vari-

able when measuring body colors, and

therefore have to be set by the user: re-

ference white, the type of light (illumi-

nant), and the observer.

Normally, colorimetric values are based

on “absolute white”. They are calibrated

to the measuring instrument’s white

standard, which is in turn calibrated to a

(theoretical) absolute white. In contrast

to densitometry, measurements are only

referenced to the paper in special cases.

4.3 Standard Illuminants

Without light there is no color. The type

of light also codetermines how we per-

ceive a color. The color of the light itself

is defined by its spectral composition.

The composition of natural sunlight is

inf luenced by the weather, the season

and the time of day.

The spectral composition of artificial

light also varies. Some lamps emit red-

dish light, while others give off slightly

greenish or bluish light.

Lighting conditions affect spectral re-

f lectance and thus color perception.

Standard color values therefore have to

be based on standardized light, which

is called an illuminant.

For standardization purposes, the spec-

tral distribution (intensity) of different

illuminants has been defined within the

wavelength range from 380 to 780 nano-

meters. The figure above shows the spec-

tral distributions of the standardized

illuminants A, C, D50 and D65.

The standard illuminants C, D50 and D65

resemble average daylight, with the

greatest radiation intensity in the blue

region. The figure below shows the

spectral composition of D65. The stan-

dard illuminant A is most intense in the

red spectrum and therefore appears

reddish (like evening light and the light

from light bulbs).


4.4 Standard Observers/

Spectral Value Functions

We are all equipped with three spectral

value functions for interpreting red,

green and blue. In persons with normal

color vision, they are approximately the

same. Consequently, only borderline

colors are perceived differently from

person to person. For example, what

one individual sees as bluish green may

be perceived by another as greenish

blue.

For colorimetric purposes, it is there-

fore indispensable to define a theoreti-

cal average person, the “standard

observer”. In the 1920s a series of experi-

ments was carried out with subjects

having normal color vision. The find-

ings were used to derive the standard

spectral value functions x, y and z,

which the CIE specified in 1931 in a

number of national and international

standards including DIN 5033 and

ISO/CD 12647.

The experiments were conducted using

a circular split screen 2 degrees across

(see the diagram on the right). This cor-

responds to a screen 3.5 centimeters in

diameter at a distance of one meter.

In 1964 the tests were repeated with a

screen 10 degrees across and the results

were also standardized, giving rise to

the “10-degree standard observer”.

1 m

10° = 17.5 cm^2° = 3.5 cm^


4.5 Evaluation with a Spectrophotometer

The standard color values are calculated

based on the spectrum of the S(λ) illumi-

nant, the measured spectral ref lectance

of the color b(λ) and the standardized

spectral value functions x(λ), y(λ) and z(λ)

for the standard observer.

The lambda in brackets (λ) indicates that

the calculation depends on the wave-

length l of the light. The first step is to

multiply the radiation function of the

standard S(λ) illuminant for each of its

wavelengths (i.e., for each spectral col-

or contained in it) by the ref lectance val-

ues β(λ) measured for the color. This

yields a new curve, the color stimulus

function ϕ(λ).

The second step is to multiply the values

of the color stimulus function by those

of the standard spectral value functions

x(λ), y(λ) and z(λ). This yields three new

curves.

Finally, integral calculus is applied to de-

termine the areas beneath these curves,

which are then multiplied by a standar-

dization factor to obtain the standard

color values X, Y and Z, which precisely

describe the measured color.

Illuminant

Reflectance

Color stimulusfunction

Standard spectral

value function

Integral andstandardization

factor

Standard color

values

times

yields

times

and

yields


4.6 Color Distance (ΔE)

The color distance (ΔE) is a measure of

how far apart two colors are within a

color space (for example, between an

original and a printed reproduction).

The CIE color space is explained in sec-

tion 1.4 on color systems. But this color

space has one serious drawback: equal

distances in the chromaticity diagram

do not correspond to equal perceived

visual differences between different

color tones.

The American D.L. MacAdam studied

this in many experiments and devel-

oped what are known as MacAdam el-

lipses that define regions of color in a

chromaticity diagram that are indistin-

guishable to the observer. The figure

shows them enlarged by a factor of ten.

Because the CIE diagram is actually

three-dimensional, in reality they are

ellipsoids. It turns out that these regions

vary widely in size depending on the

color.

As a result, the CIE color space is not

suited for evaluating color distances.

Using it would mean accepting differ-

ent tolerances for every color tone. In

order to reliably and usefully calculate

color distances, a color space is needed

in which the distance between two col-

ors actually corresponds to the perceiv-

ed difference between them. Two such

systems are CIELAB and CIELUV, which

are mathematically derived from the

CIE chromaticity diagram.

The transformation applied maps the

MacAdam ellipsoids onto spheres of

nearly identical size. As a result, the

numerical distances between colors

matches the perceived distances be-

tween them.

In 1976 the CIELAB and CIELUV color

spaces, which are now the ones most

widely used in the printing industry,

were internationally standardized.

0.8

0.7

0.6

0.5

0.4

0.3

0.2

0.1

0.00.0 0.1 0.2 0.3 0.4 0.5 0.6 0.7


The figure shows the a* and b* axes of

the CIELAB color space in the x-y color

chart.

Other color spaces are also used in the

United States, such as the CMC system

and the Munsell color space.

0.8

0.7

0.6

0.5

0.4

0.3

0.2

0.1

0.00.0 0.1 0.2 0.3 0.4 0.5 0.6 0.7


4.6.1 CIELAB

The CIELAB color space is the one most

often used to measure body colors (for

instance, for developing color mixing

formulae and measuring printed col-

ors). The hue and saturation are plotted

on the a* and b* axes. The a* axis runs

between –a* (green) and +a* (magenta),

and the b* axis runs from –b* (blue) to

+b* (yellow). The lightness axis, L*, runs

from 0 (black, below) to 100 (white,

above).

The figure on the right shows the

CIELAB color space for body colors.

Because it was derived by a mathemati-

cal transformation, it is shaped differ-

ently from the CIE chromaticity dia-

gram. The shape of the individual

lightness levels also changes with L*.

The figure below shows a cross-section

of the CIELAB color space for body col-

ors at a lightness value L* of 50. The

smaller green and enlarged blue ranges

are clearly visible.


It is very practical to use a schematic

diagram for this.

Example:

L* = 75.3 means that it is a light color,

the position of which is situated be-

tween yellow and red with a* = 51.2 and

b* = 48.4. So in this example, it is a light

yellowish red or orange.

Conclusion: The actual measured color

deviates from the specified reference

color.

The color distances are calculated using the following formulae:

ΔL* = L*actual – L*target

Δa* = a*actual – a*target

Δb* = b*actual – b*target

ΔE*ab = √ ΔL*2 + Δa*2 + Δb*2

White

Yellow

Red

Black

Blue

Green

Specified Actual

target color measured color

L* 70.0 75.3

a* 55.0 51.2

b* 54.0 48.4


In terms of their visibility, color f luctua-

tions can be classified as follows:

Because the transformation used is not

linear, the CIE chromaticity diagram

and the CIELAB color space are not in-

terchangeable. The fact that the latter

is widely used around the world speaks

in its favor, however.

White

Actual

Target

The calculation yields:

ΔL* = 75.3 –70.0 = 5.3

Δa* = 51.2 – 55,0 = – 3.8

Δb* = 48.4 – 54,0 = – 5.6

ΔE*ab= √5.32+(–3.8)2+(–5.6)2= 8.6

ΔE between 0 and 1 Deviation that is not normally

visible

ΔE between 1 and 2 Very small deviation, only visible

to a trained eye

ΔE between 2 and 3.5 Medium deviation, also detectable

to an untrained eye

ΔE between 3.5 and 5 Obvious deviation

ΔE over 5 Significant deviation


4.6.2 CIELUV

The CIELUV color space is another at-

tempt to linearize the perceptibility of

color differences. It is also derived from

the standard CIE chromaticity diagram

but using other formulae. Its three coor-

dinates are designed L*, u* and v*.

Because the CIELUV and CIELAB color

spaces result from different transfor-

mations, they differ in shape. Both are

used for body colors.

The figure shows a cross-section of the

CIELUV color space for body colors with

a luminance value L* of 50. The green

colors are located further inward than

in the CIELAB color space, and the blue

range is enlarged by comparison (see

section 4.6.1).

The CIELUV color space is frequently

used to assess the light colors of televi-

sion screens and computer monitors. Its

advantage is that it is derived from a

linear transformation, so that color re-

lationships are the same as in the master

CIE space (this is not the case with

CIELAB).


4.6.3 CIELCH

CIELCH is not a color space in its own

right; it merely refers to the use of the

cylindrical coordinates C (chroma, as

the distance from the center) and h

(hue, as an angle) instead of Cartesian

coordinates in the CIELAB or CIELUV

color space.

The calculations involved correspond

to those in CIELUV.

Here is a schematic diagram with the

same measured color locus as in section

4.6.1:

4.6.4 CMC

CMC, a system based on the CIELAB col-

or space for evaluating color distances,

was developed in Britain in 1988 by the

“Colour Measurement Committee of the

Society of Dyers and Colourists” (CMC).

Unlike CIELAB and CIELUV, it describes

how well differences in color are accept-

ed by an observer, not how they are per-

ceived.

It addresses the fact that, generally

speaking, color f luctuations near the

lightness axis are perceived as much

more irritating than deviations in more

saturated colors. It is also true that

f luctuations in chroma (saturation) are

tolerated much better than in the hue

angle.

The figure below illustrates application

of the CMC principle to assess color dif-

ferences in the CIELAB color space. Each

ellipse shows colors with acceptable de-

viations around the target locus based

on the CMC formula.

Measured color: L* = 75.3

C* = 70.5

h* = 43.4°

The lightness L* remains unchanged.

The chroma C*ab is calculated with C*ab = √a*2 + b*2.

The hue angle h*ab is equal to h*ab = arctan ( b*).a*


It can be clearly seen that the ellipses

(representing tolerances in the CMC

color space) are smaller near the central

lightness axis than in regions of greater

saturation. They are also shaped to re-

f lect the fact that the permissible devia-

tions in the hue angle are smaller than

in the chroma value. They allow for

f lexible adjustments for assessing light-

ness and color tone deviations; these

adjustments are made using two weight-

ing factors, l and c (where l is the weight-

ing factor for lightness; the weighting

factor c for the color tone is usually left

equal to 1). The textile industry often

uses weighting factors with a ratio of l :

c = 2 : 1, meaning that lightness devia-

tions are more readily accepted than

color tone deviations by a factor of two.

This relationship can be adjusted to suit

each application. This means, however,

that color distance values are only infor-

mative and comparable in conjunction

with the same weighting factors.

Δ Lightness

Δ LightnessΔ Chroma

Δ HueΔ Hue

Δ Chroma


4.7 Munsell

In 1905, Alfred Munsell developed a sys-

tem for quantitatively and objectively

representing color distances as they are

perceived. He used the terms hue, chro-

ma (saturation) and value (lightness) to

describe the attributes of color. Five

basic hues make up the notation system:

red, yellow, green, blue and purple. It

was published in 1915 as the “Munsell

Book of Color” for 40 color tones with

the C illuminant, including both glossy

and matt samples.

Each of the five basic hues is subdivi-

ded into up to 100 even-numbered color

tones, each of which has a grid com-

prising 16 chroma and 10 lightness lev-

els. The figure shows a cross-section of

the Munsell color tree with 40 color

tones. Because not all of the slots in

each grid are occupied, the result is

an irregular color space.

Munsell coordinates cannot be mathe-

matically converted into CIE coordi-

nates.

Other color systems are the DIN color

atlas (defined by DIN 6164), the Natural

Colour System (NCS), the OSA system

(from the Optical Society of America),

and the RAL design system (RAL-DS).

4.8 Tristimulus Photometry

Tristimulus photometers resemble den-

sitometers. However, instead of three

color filters for red, green and blue plus

a visual filter, they use filter combina-

tions that reproduce the three standard

spectral value functions x, y and z.

The absolute measurement precision

of tristimulus photometers is less than

that of spectrophotometers, typically

because they fail to accurately model

the standard spectral value functions

and the required standard light source

is unavailable. They are useful, however,

for determining color distances, an ap-

plication in which absolute precision is

not critical.

Tristimulus instruments are also consid-

erably cheaper than spectrophotome-

ters.

Using a lamp that emits light with a spec-

tral composition approximating that of

a standard illuminant, a control patch

is illuminated. In the example shown on

page 47, the color cyan is being mea-

sured.

The spectral ref lectance is measured

using three different filters for x, y and

z. Behind filter x (red), the standard col-

or value X is measured, behind filter y

(green) the standard color value Y, and

behind filter z (blue) the standard color

value Z. These standard color values can

then be converted to a system that lin-

earizes perceived color distances

(CIELAB or CIELUV).


4.9 Spectrophotometry

Spectrophotometry measures the visi-

ble spectrum, for example from 380 to

730 nanometers. The light ref lected by

a printed color is split into its spectral

constituents, for instance using a dif-

fraction grid, and these are captured

by a large number of sensors.

The measured ref lectance values are

used to calculate the standard color val-

ues X, Y and Z. This is done on a com-

puter using the standard spectral value

functions. Because these functions do

not need to be modeled with filters, the

absolute precision of spectrophotome-

ters is very high.

A major advantage of spectrophotome-

try — besides its high absolute precision

— is the fact that spectrophotometers

can output the standard color values for

all standardized illuminants and obser-

vers, provided that the corresponding

values have been stored. They can also

calculate densities for any desired filter

standards.

The principle of operation of a tristimulus photometer

Paper


Ink manufacturers have to make their

products to precisely match specifica-

tions. This is very important for stan-

dardized colors (as per DIN ISO 2846-1),

but also for all HKS and special colors.

They achieve this by measuring a speci-

men with a spectrophotometer and

then using an appropriate formulation

program to calculate the proportions

for mixing the ink.

Previously, it was impossible to optimal-

ly use spectrophotometers in print

shops. They were expensive and awk-

ward to use, and the measurements they

provided were not directly applicable to

the process colors. They therefore tend-

ed to be used only for one-off measure-

ments of special colors and for checking

materials (e.g. substrates and inks).

They played no role in quality control.

4.10 Spectral Quality Control with

Heidelberg

At drupa 1990, Heidelberg became the

first manufacturer to exhibit a spectral

measurement system for offset printing

that could be directly interfaced with

the press via the CPC®1 automatic re-

mote inking control system: the CPC 21.

It was joined at IPEX 98 by the spectral

image measurement system Prinect

Image Control. The CPC 21 was replaced

at IPEX 2002 by Prinect Axis Control.

puter. There the measured values are

colorimetrically evaluated and output

as the standard color values X, A and Z

and also as the standard color value

components x, y and Y.

Light source

Deflecting mirror

Paper

Annual mirror

Fiber-optic cable

Spectral remission

Diffraction grid

Press

CP2000 Center

Computer

Sensor

Diodes

At IPEX 2006, Heidelberg unveiled the

world’s first system that measures in-

line in the press: Prinect® Inpress Con-

trol.

A measuring head travels over a color

control strip and/or the printed image

and spectrally measures all of the con-

trol elements or image pixels. This can

be done as desired with the standard

illuminants A, C, D50 or D65 and either

the 2° or 10° standard observer. The

principle of operation of a spectropho-

tometer is shown below.

First the light source is directed via an

annular ref lector at a 45° angle onto the

printed specimen. The light ref lected at

an angle of 0°is relayed from the mea-

suring head to the spectrophotometer

via a def lecting mirror and a fiber-optic

cable. There it is split into its spectral

color by a diffraction grid (similarly to a

prism).

Photodiodes then measure the radiation

distribution across the entire visible

spectrum (between 380 and 730 nano-

meters) and pass the results to a com-


After comparing the measured values

with previously entered target values,

the system calculates adjustment rec-

ommendations for the various colors

involved and relays these to the Prinect®

CP2000 Center® press control system.

There the data is converted into precise

values for controlling the individual ink

zone motors and sent to them.

4.11 Color Control Strips

Heidelberg also offers a library of digi-

tal print control elements (Dipco) for all

Prinect products used to monitor and

control inking and color. This compre-

hensive package contains all the digital

elements needed to check and control

the results obtained at each stage of the

print process, from prepress to print-

ing. The Heidelberg color measurement

systems Prinect Axis Control and

Prinect Image Control measure and eval-

uate all color control strips included in

the Dipco package, provided that they

are aligned with the ink zones of

Heidelberg presses. The results of mea-

suring every element of a color control

strip are compared with stored refer-

ence values. Based on this comparison,

the Heidelberg color measurement sys-

tems calculate adjustment recommen-

dations for the individual ink zones in

each of the printing units.

How to mount the Color Control Strips

• Do not place diagonally on the sheet;

mount parallel to a sheet edge.

• Mount the strip so it is pointing

toward the center of the sheet.

• Mount all parts of the strip together

in one row, without separating

them.

• Select the correct strip for the print

job (process colors only, process and

special colors, special colors only).

• Select the correct strip for subse-

quent measurement and control

with color measurement systems:

– Full-tone/gray-patch control: use

4GS, 6GS, 6GS99 or 8 GS

– Full-tone control: use 6S or 6S+

• Select the correct strip for the half-

tone tint patches to be evaluated:

– 70%: Prinect strips

– 40% and 80%: Prinect/FOGRA strips

• Do not cut strips so the patches are

smaller than 6 mm high by 5 mm

wide.

• Position strips so they will not be

where the grippers grab the sheet.

• Strips can be placed at the leading or

trailing edge or in the middle of the

sheet (B&W printing).

• When working with Prinect Image

Control, do not position strips

directly adjoining the print image

(space about 1 mm away).

The individual patches in the strips

measure 6 mm high by 5 mm wide. The

ink zones are 32.5 mm wide in all

Speedmaster® presses, so there is room

for 13 patches across two ink zones.

4.12 Color Control with Heidelberg

4.12.1 Color Measurement and Control

Systems from Heidelberg

Prinect Axis Control measures color

control strips, doing so along one axis,

which explains its name. Measurements

are performed on the press control

console.

Prinect Image Control measures the

entire image and can then control

inking accordingly, which is also ref-

lected in its name. Measurements are

made on a separate console.

Prinect Inpress Control measures color

control strips in the press and controls

inking inline.


4.12.2 Colorimetric Control Methods

Color measurement and control systems

from Heidelberg let you select from

three different control modes:

• Colorimetric based on full-tone

(solid) patches

• Colorimetric based on gray patches*

• Colorimetric based on in-image

measurements**

Originally only two colorimetric control

modes were available: full-tone control

using a color control strip (for process

and special colors), and gray-patch con-

trol using an autotypical gray patch

(CMY) and additional solid and tint

patches for the chromatic colors CMY.

Heidelberg has added a third mode that

is based on measurement of the print

image itself. Prinect Image Control from

Heidelberg is the world’s first system

able to measure the entire printed

image and control the ink zones on

the basis of the data obtained. It is ideal

for making sure that sold product

measures up.

All three control modes use colorimet-

ric reference values. The ink zones are

adjusted to bring the print results into

optimum alignment with these refer-

ence values. In other words, the goal is a

perfect color match between the press

and reference sheets. The colorimetric

approach underlying color measure-

ment systems from Heidelberg means

that a technology is used that emulates

the color perceptions of the human eye

to minimize the detectible color dis-

crepancies between the OK sheet and

the press sheets.

4.12.3 Prerequisites for Measurement

and Control on Printing Presses

Before looking at how different mea-

surement systems work, it is important

to describe the most important pre-

requisites that must be met to ensure

reliable measurement and control. The

focus is on presetting the ink keys and

priming the ink train. How the inking

is preset depends mainly on the job's

area coverage values and the material

parameters (these are characteristic

curves stored in the press control sys-

tem). Ideally, the area coverage values

are ascertained using CIP4 PPF data

from prepress that is transferred to the

press either online or with the aid of

a memory card. The purpose of pre-

setting the ink zones is to get the colors

as close as possible to the target values

right from the start. The ink keys and

ink stripe widths are appropriately set

in every zone of each inking unit. To

determine this, the characteristic curves

are applied to convert the area coverage

values into presetting values. This

makes sure that the fountain rollers

supply exactly the amount of ink that

will be accepted by the substrate. A fre-

quently underestimated factor is prim-

ing of the ink train. Before the first

sheet is printed, the amount of ink is

introduced to the ink train that will

later result during the production run

under stable conditions. It's best if the

first sheet pulled is very close to the tar-

get colors. Experi-ence has shown that

larger deviations from the reference

values necessitate a larger number of

adjustments. If inking is set well to

begin with, there's no need to adjust it.

When starting to print, the steps

described here determine the point at

which color measurement and control

can begin.

*Not with Prinect Inpress Control

**With Prinect Image Control only

Prinect Inpress ControlPrinect Image ControlPrinect Axis Control


4.12.4 How Color Measurement and

Control Systems from Heidelberg Work

Heidelberg uses spectrophotometers for

all of its modern color measurement

systems, regardless of whether they cap-

ture ink density or L*a*b* values. During

measurement, the generated spectra

are relayed to an integrated computer,

where special software uses them to cal-

culate the required values. The spectral

color values are the basis for colorimet-

ric control, in other words, the recom-

mendations for adjusting the ink zones

are calculated directly without taking

an indirect route via the density.

For control purposes, it is vital for the

spectral values to be stored in the mea-

suring instrument as reference or target

values. In the cases of Pantone and HKS

colors, this requirement has been met in

all Heidelberg equipment. No spectral

values are stored for process colors (4C),

highly pigmented and other colors.

There are two reasons for this: the large

number of ink makes and types used in

practice, and the fact that the colors of

process inks often vary considerably.

This makes it necessary for the press

operator to determine the spectral val-

ues of these inks by measuring a (full-

tone) print specimen to ascertain a new

target color locus. This only takes a few

minutes and has the advantage of gener-

ating target values that can realistically

be achieved with the ink used in the

print shop. Quality control by monitor-

ing color deviations is also feasible, for

instance between different batches of

ink.

The colorimetric control process. Both the relevant values and the adjustment

recommendations are directly derived from the spectrum of the measured color.

4.12.5 Determining Target Values:

A Practical Example

Let’s say that you want to print accord-

ing to Media Standard Print 2004 (Medi-

enStandard Druck 2004). This standard

defines dot gain as well as colorimetric

reference values expressed as CIE-L*a*b*

coordinates. Due to the inf luence of var-

ious factors, the CIE-L*a*b* can never be

perfectly matched, so tolerances are

also given for the individual process col-

ors under production conditions. It is

important for the press operator to

know how closely he can approximate

the reference values to the inks he is

using.

There are two practical approaches for

determining the achievable target value

(= pressrun standard): 1. Make a series of

prints ranging from underinked to over-

inked and measure them. The sheet in

which the color is closest to the target

value within the permissible tolerances

is suitable for use as the standard for the

measurement system. 2. Have the ink

manufacturer prepare a laboratory test

print on the same paper that will be

used for the job. Scan it to serve as the

standard for the measurement system.


Flow chart showing the conversion of spectral values into control signals

4.12.6 Inline Measurement and Control

After the target values have been deter-

mined, measurement of the pressrun

can begin. The first sheet pulled pro-

vides the first actual measured values,

which — as mentioned before — should

not be too far from the target values.

The goal is now to adjust the ink zones

and thus the ink film thicknesses to

achieve the target values in a minimum

of steps.

This approach may seem simple at first

glance, but it is based on a complex

color model that describes how changes

in the ink film thickness affect the color

of the ink used. Colorimetry by itself

can only tell us where the color achieved

so far is located within the color space

(the actual measured value) and where

we need to get it to (the target or refer-

ence value); what it doesn’t tell us is how

to accomplish this. But this is not the job

of colorimetry. That is what the color

model used is for. It can be used to work

out how the color changes if, for exam-

ple, the ink film thickness is increased

by five percent. If the film thickness on

the paper is changed by applying more

or less ink, its visual appearance also

changes by a certain amount, as we

know. Imagine a series of prints ranging

from very light inking to full saturation

within the CIE-L*a*b* color space. They


lie along a line that varies not only in

terms of lightness, but also in its posi-

tion along the a and b axes. This is called

a color line. When using full tones to

control inking, the achievable color

tones are fixed by the ink’s pigmenta-

tion, color intensity and variable film

thickness. This color model can be used

in this example to calculate which film

thickness comes closest to yielding the

target value, and where the target value

is located within the color space.

4.12.7 How Colorimetry Helps

In practice, this means that the operator

sees at a glance whether or not he can

attain the desired color results. If all

parameters of the print process are opti-

mally coordinated, he can expect to

achieve them. If the printing conditions

change, for instance resulting in black-

ening of the chromatic colors in the

pressrun, the colors can deviate signifi-

cantly from the targets. Colorimetry

can then be a major help by revealing

whether the desired color results can

continue to be reached within the speci-

fied tolerances under these conditions,

or whether it is necessary to take steps

such as washing the inking rollers.

When using an ink whose target value is

not stored, the color measurement sys-

tem also shows, right from the very first

pull, whether the colors can be kept

within the tolerances or not. This can be

the case, for example, when working

with a different make or type of ink and

a previously stored reference value. This

is a situation in which one of the mea-

surement system’s most important

functions comes into play: the ability to

determine and display the smallest

achieve color deviation (ΔE0). It can also

happen that different batches of the

same type of ink let you attain the same

CIE-L*a*b* but with different densities.

If you only printed based on the refer-

ence densities, the visual appearance of

the prints can diverge afterward. This is

why the ISO standard does not provide

any reference densities.

The operator sees at a glance where the inking needs to be corrected. The black line shows the reference

colors. The bars show the recommended adjustments, expressed as percentages, for each ink zone.

Colorimetric control always shows two results: the distance (ΔE) to the target color that still needs

to be overcome to get as close as possible to the reference value, and the remaining discrepancy

(ΔE0) between the actual and reference values that cannot be eliminated. The color line is shown in

red.

(Magenta)

b

Actual ΔE Reference

Color line

Target

ΔE0

CIE-L*a*b* color space a


4.12.8 Summary

The biggest advantage of colorimetric

control is that it lets you consistently

bring the results of printing as close as

possible to the visual appearance of the

original, letting you know quickly if the

deviation becomes too large. Color-

imetric evaluation corresponds to the

color perceptions of the human eye,

with the advantage of being free of sub-

jective inf luences and variable environ-

ment inf luences and therefore able to

deliver objective readings. The measure-

ment data can be stored and docu-

mented and used for quality certifi-

cates. Measurement results can also be

automatically evaluated with the Qual-

ity Monitor software from Heidelberg,

which is part of two Prinect products:

the Prinect® Profile Toolbox und

Prinect® Calibration Toolbox.

4.13 Standardization of Printing

The standards of the graphic arts indus-

try discussed below play key roles.

The ProzessStandard Offsetdruck (German Offset Printing Process Standard)

of the German Printing and Media Industries Federation (bvdm)

Screen ruling 60 lpc

Screen angle Nominal angular difference between C, M, K = 60°

(chain dots), = 30° (circular or square dots) Y = 15°

from another color, dominant color at 45° or 135

Halftone dot shape Color control strip: circular dots, image: chain dots

with 1st dot touch ≥ 40%, 2nd dot touch ≤ 60%

Total area coverage ≤ 340 %

Gray balance Cyan Magenta Yellow

Quarter tones 25 % 18 % 18 %

Midtones 50 % 40 % 40 %

Three-quarter tones 75 % 64 % 64 %

ISO-Compliant Inks

The Euroscale, originally defined by

DIN 16539 in 1975, has been improved

since then. In 1996, ISO 2846 suc-

ceeded in establishing a common pro-

cess standard that incorporated the

ideas of the U.S. SWOP and the Japanese

TOYO standards. Part 1 of this standard

defines tolerances for colorimetric

properties and transparencies for pro-

cess inks for four-color and web offset

printing that may not be exceeded

when making test prints on APCO paper

with a defined reference ink film thick-

ness. However, the color values given in

this standard are binding only for ink

manufacturers, not for printers.

ISO 12647-2 and ProzessStandard Off-

setdruck (German Offset Printing Pro-

cess Standard)

In 1981, the German Printing and Media

Industries Federation (bvdm) issued its

first publication on standardizing sheet-

fed offset printing. The practical experi-

ence gained and the relevant scientific

research findings made subsequently

were incorporated into the interna-

tional standard ISO 12647-2 “Process

control for the production of half-tone

color separations, proof and production

prints”. A revised edition of ISO 12647

was published in November 2004. ISO

12647-2 provided the basis for the

ProzessStandard Offsetdruck (the Ger-

man Offset Printing Process Standard)

that the bvdm published in 2003. It can

be ordered from the bvdm as a binder

(193 A4 pages with supplementary

sheets). Because its scope goes beyond

that of ISO 12647-2, many printing com-

panies in Germany and elsewhere are

taking advantage of it as the basis for

achieving faithful color reproduction.


Paper type 1/2 3 4 5

L*/a*/b* L*/a*/b* L*/a*/b* L*/a*/b*

Black backing

Black 16/0/0 20/0/0 31/1/1 31/1/2

Cyan 54/–36/–49 55/–36/–44 58/–25/–43 59/–27/–36

Magenta 46/72/–5 46/70/–3 54/58/–2 52/57/2

Yellow 88/–6/90 84/–5/88 86/–4/75 86/–3/77

Red 47/66/50 45/65/46 52/55/30 51/55/34

Green 49/–66/33 48/–64/31 52/–46/16 49/–44/16

Blue 20/25/–48 21/22/–46 36/12/–32 33/12/–29

Substrate backing

Black 16/0/0 20/0/0 31/1/1 31/1/3

Cyan 55/–37/–50 58/–38/–44 60/–26/–44 60/–28/–36

Magenta 48/74/–3 49/75/0 56/61/–1 54/60/4

Yellow 91/–5/93 89/–4/94 89/–4/78 89/–3/81

Red 49/69/52 49/70/51 54/58/32 53/58/37

Green 50/–68/33 51/–67/33 53/–47/17 50/–46/17

Blue 20/25/–49 22/23/–47 37/13/–33 34/12/–29

Paper types 1 2 3 4 5

115 gsm 115 gsm 65 gsm 115 gsm 115 gsm

Glossy Matt-coated LWC Uncoated Uncoated

Coated Art reproduction Web offset White offset Yellow offset

Art reproduction

L*a*b* reference values for the five paper types


The Media Standard Print

(MedienStandard Druck)

The Media Standard Print (MedienStan-

dard Druck) first appeared in 2004 at

the initiative of the German Printing and

Media Industries Federation (bvdm). In

addition to technical guidelines for digi-

tal data for printing, based on ISO 12647,

it defined specifications and tolerances

for digital contract proofs. This estab-

lished a set of rules for agencies, pre-

press studios and printing companies,

providing a basis for improving commu-

nication and optimizing workf lows. In

2004 the fourth, revised edition of the

Media Standard Print was issued. It pri-

marily established the following rules:

• A proof must simulate one of the five

reference print conditions defined

by the ProzessStandard Offsetdruck

(German Offset Printing Process

Standard).

• A proof must include a line of text

indicating the file name, the output

date, and the color management

settings used.

• A UGRA/FOGRA media wedge must

be included.

• The conditions for measurement

and evaluation must be defined.

AF (%) Dot gain ΔA (%) for paper type

1 + 2 3 4 + 5

40 09 – 13 – 17 12 – 16 – 20 15 – 19 – 23

50 10 – 14 – 18 13 – 17 – 21 16 – 20 – 24

70 10 – 13 – 16 12 – 15 – 18 13 – 16 – 19

75 09 – 12 – 15 10 – 13 – 16 11 – 14 – 17

80 08 – 11 – 14 08 – 11 – 14 09 – 12 – 15

Media Standard Print

Specifications and tolerances for digital proofing

ΔE

Mean ΔE for all L*a*b* color distances of color patches 4

Maximum ΔE for all L*a*b* color distances of color patches 10

Tolerance for primary colors 5

Maximum deviation of substrate 3

Reference dot gain values for the five paper types

Specifications and tolerances for digital proofing

4.14 Benefits of Colorimetry for

Offset Printing

To sum up, here is an overview of the

principal advantages that colorimetry

offers to offset printing:

• The measurement values very close-

ly match visual perception of the

colors.

• Colorimetry is a process-indepen-

dent color evaluation method that

can be used throughout the print

process from prepress across all

kinds of proofs to final quality

control of finished products.

• Colorimetric reference values can

also be expressed as figures. Linking

to prepress is possible.

• Colorimetric reference values can be

taken from specimens.

• Colorimetry is the only way to en-

sure objective evaluation.

• Colorimetry makes image-relevant

color control possible (for instance

using gray patches) without cali-

bration of individual colors and

stored conversion tables.

• All colors, including very light spe-

cial colors, can be correctly and

dependably controlled with

colorimetry.

• Dot gain is precisely captured by

spectral measurement, also with

special colors.

• Control of the production run is

more reliable, because changes in

the substrate, ink soiling and

metamerism can be captured and

taken into account.

• Halftone printing with more than

four colors can also be correctly

controlled.

• Print quality can be described and

documented more effectively.

There is a color-tone-independent

measure of color deviations: ΔE.

• Spectral measurement enables the

development of better color models

possible.

• Colorimetry lets the printing in-

dustry move into line with all other

industries in which color plays an

important role.

• Densitometry is an integral part of

spectral color measurement.

• Print image fragments can also be

compared with originals.


Expert Guide

Color & Quality

Heidelberger Druckmaschinen AG

Kurfürsten-Anlage 52 – 60

69115 Heidelberg

Germany

Phone +49-62 21-92 00

Fax +49-62 21-92 69 99

www.heidelberg.com

Publishing InformationPrinted: 10/06Photographs: Heidelberger Druckmaschinen AGPlatemaking: SuprasetterPrinting: SpeedmasterFinishing: StahlfolderFonts: Heidelberg Gothic, Heidelberg AntiquaPrinting in the Federal Republic of GermanyCopyright © 2006 by Heidelberger Druckmaschinen AG

Trademarks Heidelberg, the Heidelberg Logo, Prinect, Axis Control, CP2000 Center, CPC, Image Control, Speedmaster and Mini Spots are registered trademarks of the company of Heidelberger Druckmaschinen AG in Germany and other countries. Other product names used here are trademarks of their respective owners.

Subject to technical and other changes.

tester

Documents

color quality light

color balancecomposition

color sequence

color perception

color reproduction

color systems

world of color

color control strips