NEW CD WARP CONTROL SYSTEM

FOR

THE CORRUGATING INDUSTRY

USING A NEW
CONCEPT IN MOISTURE SENSING
AND CONTROL

BY

DRYING TECHNOLOGY, INC

 A New CD Warp Control System
For the Corrugating Industry

Introduction: Brief history of Drying Technology, Inc

Drying Technology, Inc., (DTI) was introduced to the corrugated board CD warp
problem in the late 1980’s as a result of a request for a proposal and quote for a
warp control system. Later, the research department of a large paper
manufacturing company requested a similar proposal. We complied with each
by proposing a system that included use of our basic Delta T sensing
technology1 that requires two temperature sensors and a mathematical model
derived from first principles to sense and control linerboard moisture content
(MC). The first company soon dropped their project, and the paper company
informed us that they would develop their own system; however, they never did.

Later, we modified our basic moisture sensing technology to provide a more

practical and effective method for sensing linerboard (liner) MC exiting a
preheater. This new liner MC sensing method was incorporated into our earlier
warp control system which now offers an excellent chance for solving the
industry’s CD warp control problem. All that remains is to test it on a corrugator.

The Warp Control Problem:

It is well known that linerboard (liner) expands when wet and shrinks as it dries.
Further, when two liners are joined in a corrugated board, the difference in the
top and bottom liner MCs must be below a certain value, e.g., 1% to prevent
warp. Therefore, a control system that is capable of maintaining this “no warp”
MC difference under conditions of differing corrugator speed, liner basis weights,
and incoming MCs, should be successful in virtually eliminating CD warp in
corrugated board.

Controlling moistures for the purpose of reducing or eliminating warp is not a new
concept; however, in practice its effectiveness has been limited by the lack of a
reliable and rugged moisture sensor. Some warp control systems were designed
with conventional IR moisture sensors traveling back and forth across the liner
width to scan the liner MC. However, the harsh environment of a corrugator, the
relatively high cost of conventional IR sensors and accompanying equipment,
and the unreliability of conventional IR moisture sensors appears to have forced
use of liner surface temperature as substitute for liner MC. It will be shown below
that prediction of MC using temperature is quite poor and significantly reduces
the effectiveness of control systems using this approach.

2
Non-Conventional Moisture Sensing:

Work commenced on the Delta T moisture/dryer model conceptualization in 1975

followed by completion of the derivation of the model (equation 1) in 1979. A
moisture sensing and control system was successfully applied to eight veneer
dryers during the period 1982 to 984 using this non-conventional method for
sensing and controlling MC using two temperature sensors and a mathematical
model derived from first principles. A paper describing its derivation was
presented in November 27 – 28 at the North American Wood Symposium, MS.
State, MS., followed by two patent2 applications filed in 1984 and 1985. Patents
were granted in 1987 and 1988. The model for continuous drying is,

MC = K1 (∆T) p – K2/Sq (1)

Where the product moisture content (MC) is related to the temperature drop (∆T)
of hot air after contact with the wet product; and (S) is the production rate or
dryer speed; Ks are constants, and p & q are exponents.

The Delta T model initially was used to predict product MC exiting dryers, ovens,
kilns, and the like with a high degree of accuracy and reliability. It does not
require costly and time-consuming calibration as do conventional moisture
sensors. The controlled variable, (∆T), the difference in two temperatures, is
easily determined which enables MC sensing in very harsh environments. The
Delta T been validated by over 300 successful installations on such dryer-types
as flash, spray, fluidized-bed, rotary, and conveyor during the drying of a variety
of products. 3

Modification of the Delta T:

Later, it was found that the Delta T principle not only could predict the MC from a
drying operation, but also from webs such as textiles and liner that are cooling
after exiting a preheater. Two successive surface temperature sensors are
placed immediately after the preheater such as in figure (1) for the purpose of
sensing the top or bottom liner MC. The basic Delta T model remained the same
with the exception that the (∆T) term is determined in a slightly different manner.

If the evaporation rate from the liner is a function of the MC and the cooling rate
(T1 -T2) is a function of the evaporation rate, if follows that MC is proportional to
the cooling rate.

3
 Figure (1) – Location of Temperature Sensors
For MC Measurement of Liner

Figure (2) shows the relationship between the delta t measurement on the
cooling curve and the corresponding liner moistures. The liner temperature (red)
is shown as it enters the preheater from the left and heats up. Upon exiting the
preheater, the liner temperature follows a cooling curve. Temperatures T1 and T2
are used to predict moisture (M2) shown on the (blue) moisture curve.
.

Figure (2) – Liner Temperature & MC

To and From a Preheater

4
Development of Simulation Model Data:

A simulation program was written using the information from this article. The
simulation uses a finite time-element approach for solving heat and mass transfer
balances from the time the paper comes in contact with the preheater to a
distance after the preheater. The paper temperature and moisture content were
tracked in small increments over this distance. The results from the simulation
were compared to the data and charts found in the literature to verify the
correctness of the calculations. Such operating parameters as speed, preheater
wrap, basis weight, and incoming paper MC were varied over a wide range of
operating conditions.

The next step was to include the Delta T moisture sensor technology in the
simulation. Data was recorded for each basis weight by running a matrix of
operating conditions at various speeds, preheater wrap percentage, and
incoming moisture. Approximately 200 different conditions were run and
recorded based on the following,

Speed 150 m/s to 225 m/s in 25 m/s increments

Preheater Wrap 30% to 90% in 10% increments

Incoming Liner MC 6.0% to 9.0% in 0.5% increments

The data was analyzed and the Delta T principle applied to determine the
correlation between the delta t value and moisture of the liner at a distance
beyond the preheater. Excellent correlation exists for all four basis weights used.
These results are proof that the Delta T principal could be applied to the cooling
of the paper after the preheater to determine the moisture of the paper. It was
also noted that the paper temperature alone did not correlate well with paper
moisture as shown in figure (3) below.

MC Vs Liner Temperature Shows Poor Correlation:

Although numerous attempts have been made over the years to use temperature
to predict liner MC, correlation between MC and temperature is quite poor as
seen in figure (3). Therefore, temperature of the liner should not be used as a
surrogate for liner MC in a warp control system, which is in agreement with
comments found in the literature5.

5
 Figure (3) – MC Vs Temperature
For 42 lb linerboard

MC Vs Delta T Value shows excellent Correlation:

As mentioned above, the evaporation rate from the liner is proportional to the
MC, and the cooling rate is proportional to the evaporation rate. Therefore, delta
T is proportional to MC. A plot of MC vs delta T for 42 lb liner at the same
simulation conditions as used in figure (3) shows delta T to be an excellent
predictor of MC as seen in figure (4) below.

Figure (4) – MC Vs Delta T

For 42 lb linerboard

6
Figure (5) shows excellent correlation between MC and delta T for four basis
weights; therefore, liner MC should be easily predicted and controlled based on
delta T. This presents excellent proof that use of the Delta T method of MC
sensing and control can be used to improved CD warp control in the corrugating
industry.

Figure (5) – MC Vs Delta T At

Various Basis Weights

Development of the Delta T Warp Control System:

Recognizing that warp might be eliminated or significantly reduced by

maintaining a difference between the top and bottom liner MCs, simulations
were run at various differences in top and bottom liner MCs that were assumed
to be at “no warp” conditions. Both temperature and the delta T method of
providing relationships at “no warp” conditions were compared. The results are
shown in figures (6) and (7) below.

Figure (6), a plot using the simulation data, shows poor correlation between
bottom and top temperature for the “no warp” conditions. These results reinforce
our earlier conclusion that the use of temperature as a surrogate for MC would
not provide an adequate basis for a warp control system.

7
 Figure (6), Bottom Vs Top Liner Temperatures

Figure (7), a plot using the same simulation data, shows that the use of the delta
T method for sensing MC produces a “no warp” relationship with excellent
correlation.

Figure (7), Bottom Vs Top Delta T Values

At “No Warp” Condition

8
A warp control system can now be designed that maintains the “no warp” delta T
relationship. Similar relationships are generated for various differences in MCs
at the “no warp” condition. This “no warp” relationship holds even for board
made from liners at different basis weights.

Warp Control Algorithm Development:

The next step in the process was to determine the relationship between the top
and bottom delta t values at the conditions of ‘no warp’. The application of the
Delta T had to be made without knowing the actual moisture of the liners, since
liner moisture would be difficult to obtain on a production line. Assumptions were
made concerning the moisture relationship between the top and bottom liners
that would result in “no warp.” Four assumptions for “no warp” differences
between top and bottom moisture were tested. These were “no warp” with:

(1). The bottom liner moisture 0.5% greater than top moisture.
(2). The bottom liner moisture 0.7% greater than top moisture.
(3) The bottom liner moisture 1.0% greater than top moisture.
(4) A varying difference based on the bottom moisture.

Data was gathered by running the simulation with varying incoming liner
moistures and wrap arm positions. The wrap arms were adjusted to satisfy the
assumption of a specific difference between the top and bottom liners and then
the data was recorded. These data points would represent the ‘no warp’
conditions.

The ‘no warp’ data was analyzed and the Delta T principal was applied. There
was an excellent relationship between the top and bottom delta t values during
the ‘no warp’ conditions. An excellent relationship was found for all four “no
warp” conditions (figure (7)). The initial test was made with the same liner on top
and bottom. Additional test were made with two different liners and the results
were the same. The control algorithms would be set up to maintain the
relationship between the delta t values by adjusting the wrap arms.

The control algorithms were programmed into the simulation to test the automatic
control of the wrap arms to maintain ‘no warp’ conditions. The error in the
relationship between the top and bottom delta t values was used to initiate a
control response. The basic scheme was to control the wrap arms in a way to
maintain an equal distance from 50% warp. This would help to eliminate a
preheater warp from reaching a minimum or maximum value too quickly.
Another constraint is the minimum temperature required for gluing. Moisturizers
could also be used to adjust moisture if needed.

9
 Figure (8) – Simulation Program Screen Display

Conclusions and Recommendations:

It is commonly known that if the difference in the top and bottom liner MCs could
be easily measured and controlled, a warp control system for virtually eliminating
warp in the finished board might be a possibility. Based on work described
above, we propose a warp control system based on a new, simple, non-
conventional MC sensing for use in controlling the moisture of each liner.

This new CD warp control method can be supplied in the following listed
packages:

1. If allowed, Integrate the Delta T warp system into an existing wet end
control system.

2. A stand-alone Delta T warp control system.

References:
1. Robinson, John, “Improve Moisture Control,” Chemical Engineering Progress, vol. 88, no. 12, pp. 28-33,

2, US Patents 4, 701, 857 & 7,777,604

3. Robinson, John & Douglas, Roger, “Improve Moisture Control for Profit,” The Process Engineer, April 2005, 3 pgs.

4. Taylor, Bruce, “The Interaction of Paper Moisture and Temperature and its effect on Corrugated Board Quality,”
Corrugating International, Vol. 1, No. 3, pp. 39-50.

5. ibid.

Rev. 12.23.09

10