Internal Process Controls - Control of Color Imaging Systems: Analysis and Design - page 488

Although halftoning process is well de ned and repeatable, the quality of the

developed image on the photoconductor depends on the system s ability to reproduce

the tones in the presence of exposure and development uncertainties. The developed
mass of toner per unit area (DMA) is normally used to quantify the macro level toner
reproducibility on the photoconductor. This toner mass is a function of numerous
parameters of which many are

'

fixed by design. Parameters that contribute to the
variability of DMA within print runs can lead to print quality problems and increase
the number of misprints. The most important parameters in a two component
developer system are (1) toner concentration (TC) the ratio of the amount of

toner to the amount of carrier available in the development system and (2) the
toner charge per unit mass (Q

—

M ratio). The TC changes during engine operation
due to a varying amount of toner depletion (or addition) caused by image develop-
ment. The charge-to-mass ratio of the toner is dependent on the triboelectric prop-
erties of the toner and carrier, TC, relative humidity (RH) of the air in the
development substation, and toner agitation mechanism in the developer. Generally,
in xerographic engines, TC is maintained to some constant level using a sensor in a
separate multilevel asynchronous control loop described in Section 9.11. Although
development can be affected directly by varying the Q

=

= M ratio, a direct control of

Q M ratio cannot be implemented easily without increased hardware cost. Another

way of reducing the variability in development is to measure the DMA with sensors
and adjust the tone levels using area modulation as in the level 3 control loop
described later in Section 9.9 and

=

or by adjusting photoconductor charge, exposure,
and development bias at some periodic intervals.
A common technique for measuring DMA is to arti

=

cially create test patches of
reasonable size (e.g., 1 in. 2 on a photoconductor belt) in the IDZ at predetermined
area coverages that act as a surrogate measurement for capturing effects on the
customer image area [9,10]. These test patch areas are charged, exposed, and
developed by the laser system as necessary. The developed test patches are read
by the re ected signal from the patch area [14]. Spatial effects are captured as an

integrated signal in the sensor output and they contribute toward noise in the
measurement. The number of test patches used depends on how many tone levels
in the tone curve have to be sampled to generate the control actuation and the
approach used to implement the control function. For example, in Ref. [15], for
time-sequential sampling of tone curves, one tone level (area coverage) is proposed
during a periodic actuation instance. The patch area coverage is varied over time
using a time-sequential sampling strategy. Other more feasible implementations
(e.g., iGen3) use fixed area coverages; one at high area coverage (90% - 100%),

-

one at low area coverage (0% 20%) and one at mid-tone (around 50%), which is

considered adequate to sample the critical regions of the tone curve and affect the
entire curve through feedback to process actuators such as charge, exposure inten-
sity, and development bias [1,16]. A three patch sampling is required at reasonably
high print frequency to maintain the stability of mid-tones in image-on-image
engines. Some printers, such as the DC2060 and DC6060, which develop the
image on four individual drums, use a two patch sampling strategy.
During control, the DMA measurement vector from the sensor is

filtered
using an appropriate algorithm to remove noise components and obtain a true