EXAMINATION OF THE AIR TEMPERATURE PROFILE
ACROSS LUMBER IN THE HIGH TEMPERATURE
DRYING OF SOUTHERN PINE
David M. Landoch and Professor Fred Taylor
Mississippi State University
Mississippi State, MS
ABSTRACT
The objective of this study was to examine the relationship between air
temperature and distance across the load in the high-temperature drying of
Southern Pine. Twelve charges of 2 x 6 dimension lumber were dried to an average
estimated moisture content of 12% using 4 different kiln schedules. The air
temperature was recorded at 1-foot intervals across the load during drying. Graphs
representing temperature drop across the load (TDAL) were developed for certain
time segments of drying for each schedule to examine the relationship between air
temperature and distance across the load. Values of TDAL and average charge
moisture content were examined to determine the viability of TDAL as a basis of
kiln control. The data produced by this study showed that temperature drop across
the load was not an accurate reflection of average charge moisture content at the
end of drying.
INTRODUCTION
There are many factors which affect the quality of kiln dried lumber. Drybulb temperature, wet-bulb temperature, air speed, and kiln residence time must
be carefully monitored when drying refractory species such as oak and walnut
(Haygreen and Bowyer, 1982), because a failure in properly controlling one or more
of these factors can cause product value loss due to degrade. These, however, are
not major concerns in drying southern pine. Southern pine is a highly permeable
species (Siau, 1984), meaning moisture movement produced by a pressure gradient
(as in kiln drying) is relatively high compared to other commercial species. The
main concern in drying southern pine is the determination of the drying end point.
Four popular techniques currently used in commercial operations to determine the
average charge moisture content are: The use of pre-cut, end coated sample
boards; electrical resistance measurement of wood; predetermined drying
schedules; and a measure of the temperature drop across the load (TDAL).
Temperature drop across the load is a measure of the difference between the air
temperature at the entering-air side of a stack of lumber and the air temperature
at the leaving-air side of the stack. The purpose of this study is to examine the
relationship between air temperature and distance across the load, and to determine
the viability of TDAL as a basis of kiln control, mainly in terminating the drying
process.
An increase in production from lumber mills has necessitated the
fabrication of newer and more sophisticated drying equipment. These new dry kilns
are well sealed, have powerful heating systems, efficient air speed control, and zone
temperature controls. Many of these kilns have computerized process control.
Computers monitor TDAL and use this as a basis for temperature control and
drying termination. Zones are areas inside the kiln which have their own heat
source. Temperatures can be changed from zone to zone according to a designated
computer program. TDAL, or a form thereof, such as the temperature at the
53
leaving air side of the stack, is monitored by the computer. Depending on the value
of TDAL in each zone, the computer will compensate for each zone by increasing
or decreasing the amount of heat put into each zone. A zone with a higher TDAL
would be given more heat than a zone with a lower TDAL. Theoretically, this gives
more uniformity of average final moisture content.
Methods and Materials
Green 2" x 6" x 12' southern pine dimension lumber was obtained from a
local sawmill and transported to the Mississippi Forest Products Utilization
Laboratory (MFPUL). The lumber was taken from a green chain to ensure freshly
sawn lumber for all experimentation.
Twelve packages consisting of 55 boards each were dried in the hightemperature dry kiln at the MFPUL. This kiln is well sealed, has a large heating
capacity, and is therefore capable of attaining dry-bulb temperatures above 275°F
and wet-bulb temperatures up to 212°F. Each charge was stacked 6 feet wide (11
boards) and 5 layers high. Laminated stacking stickers (1 1/2 inches wide by 3/4
inch thick) were placed evenly throughout the charge at 2 foot intervals. Two rows
of 7 thermocouples were placed 2 feet apart facing into the same conduit between
the third and fourth layers and 4 and 6 feet from the end of the charge.
Thermocouple placement was achieved by drilling 5 small holes one foot apart in
a sticker. The thermocouple wire was placed through each hole and affixed with
silicone. The thermocouples at the sides of the charge were stapled at the
appropriate location on each side of the charge. The air velocity was maintained
at 1,200 feet per minute during drying. Four one-step experimental kiln schedules
were used: dry-bulb temperatures of 245°F and 275°F and wet-bulb temperatures
of 180°F and 212°F. Three replications of each schedule were performed for a total
of 12 charges.
A PC data acquisition system was used to monitor thermocouple and load
cell readings. The current temperature and charge weight were displayed on the
screen throughout the drying process. At 3-minute intervals, the temperature of
each thermocouple and the weight of the charge were stored on a diskette. Drying
was terminated when the charge reached a weight calculated to coincide with 12%
moisture content. Each board was weighed before and after drying. The moisture
content of the boards was determined by electronic moisture meter readings. A
moisture meter with 5/16-inch probes was used to measure moisture content at the
center of the board and 1 foot from each end.
Statistical analysis was performed on the temperature data by least squares
regression analysis. The statistical results obtained by regression were examined
to determine the relationship between the temperature at a location across the
stack and distance of the location across the stack. This analysis was done for
various times throughout the drying process and for each of the 4 experimental
drying schedules.
RESULTS AND DISCUSSION
Figures 1 through 4 are graphs of air temperature versus distance across the
load for various time segments of the drying process for each of the 4 schedules.
The temperatures are average values for each specified time segment. The
relationship between temperature and distance across the load does exhibit a strong
linear relationship in most cases, although this relationship is better described as
a quadratic curve because air temperature decreases at a decreasing rate (See Table
1). R-squared is a reflection of the strength of the relationship between
temperature and distance. A larger r-squared means a stronger relationship. A
54
236
233 230 227224 221 218 216 212 209 208 203
0
1
3
2
4
6
Distance Across the Load (ft)
3rd 3 Hre.
let 3 Hre. -4- 2nd 3 Hre
8
10th Hr.
Figure 1. TDAL lines for charge 1 of schedule 1. Each line represents the average
temperatures over the specified time segments.
Aug,
261
248
246 242 239 238 233 230 227 224
221
218
216
212 209 208 203
0
Temp. (F)
1
2
3
4
6
6
Distance Across the Load (ft)
1st
3
Hre
-42-- 4th 3 Hrs.
-4- 18th Hr,
Figure 2. TDAL curves for charge 1 of schedule 2. Each curve represents the
average temperatures over the specified time segments.
55
280
277
274
271
288
286
282
269
268
263
260
247
244
241
238
236
232
229
228
223
220
Avg, Temp. (F)
1
0
3
2
8
4
Distance Across the Load (ft)
—
tat 3 Hrs.
-4-
2nd 3 Hre.
7th Hr.
—12— 8th Hr.
Figure 3. TDAL lines for charge 1 of schedule 3. Each line represents the average
temperatures over the specified time segments.
280
Avg. Temp. (F)
274
271
288
2 86 282 259 268
263
260 247 244 241 238 236 232 229 228 223 220
0
1
3
2
4
6
8
Distance Across the Load (ft)
1st 3 Hrs.
--1-- 7th Hr.
--*— 9th Hr.
Figure 4. TDAL lines for charge 1 of schedule 4. Each line represents the average
temperatures over the specified time segments.
56
Table 1.
Sch. #
Chg. #
1,1
1,2
1,3
2,1
2,2
2,3
Regression analysis results of the average temperature data for all
charges of each schedule. R^2 indicates r squared.
R^2
(line)
R^2
(curve)
1st 3H
2nd 3H
3rd 3H
10th H
1st 3H
2nd 3H
3rd 3H
10th H
1st 3H
2nd 3H
3rd 3H
10th H
.9852
.8736
.9753
.9446
.9795
.9806
.9716
.8960
.9601
.9775
.9914
.9257
.9939
.9269
.9804
.9553
.9862
.9864
.9789
.9610
.9603
.9935
.9915
.9614
1st 3H
2nd 3H
3rd 3H
4th 3H
5th 3H
16th H
1st 3H
2nd 3H
3rd 3H
4th 3H
5th 3H
16th H
1st 3H
2nd 3H
3rd 3H
4th 3H
5th 3H
16th H
.9389
.9076
.8931
.8224
.8938
.6621
.9902
.9339
.9302
.8496
.8455
.6607
.9968
.9831
.9955
.9921
.8975
.8934
.9741
.9685
.9308
.9326
.9480
.9584
.9957
.9897
.9859
.9833
.9475
.9268
.9992
.9879
.9981
.9925
.9325
.8944
Time
Sch. #
Chg. #
3,1
3,2
3,3
4,1
4,2
4,3
57
Time
R"2
(line)
R"2
(curve)
1st 3H
2nd 3H
7th H
8th H
1st 3H
2nd 3H
7th H
8th H
1st 3H
2nd 3H
7th H
8th H
.9625
.9517
.9254
.9181
.9811
.9254
.9634
.9620
.9783
.9865
.9551
.9515
.9949
.9653
.9778
.9607
.9847
.9906
.9717
.9628
.9960
.9911
.9797
.9701
1st 3H
2nd 3H
7th H
8th H
9th H
1st 3H
2nd 3H
7th H
8th H
9th H
1st 3H
2nd 3H
7th H
.9565
.9013
.9131
.9122
.9146
.9779
.9875
.9664
.9684
.9778
.9645
.9807
.8830
.9869
.9852
.9733
.9122
.9411
.9827
.9931
.9832
.9737
.9831
.9702
.9833
.9763
value of 1 (which is the highest possible r-squared) denotes a perfect relationship
between the temperature data and the line or curve. Notice that the r-squared
values are always higher for the curve than for the line.
Temperature drop across the load is currently being used commercially with
computer process control to monitor, control, and terminate drying. To use TDAL
effectively as a basis of kiln control there needs to be a strong relationship or direct
proportionality between TDAL and average charge moisture content. Figures 5
through 8 are graphs of TDAL and average charge moisture content versus drying
time. Temperature drop across the load tends to decrease as moisture content
decreases, but its behavior is erratic and there does not appear to be a close
relationship between TDAL and average charge moisture content. Table 2 shows
a comparison between average charge moisture content and average TDAL at the
end of drying for each charge of the 4 schedules. Schedule 3 had a TDAL value of
39.0°F relative to a moisture content of 11.9% and a TDAL value of 31.8°F with an
average final MC of 16.0%. Clearly TDAL is not a good reflection of the average
moisture content of lumber drying in a kiln.
SUMMARY
Temperature drop across the load does not accurately depict the moisture
content of lumber. Therefore the reported success with TDAL may be due to the
quality of the drying equipment and not the computer process control with TDAL.
Computer process control used to terminate the drying process when TDAL drops
to a certain level must operate on conditions that allow TDAL to fluctuate.
Temperature drop across the load varies considerably throughout drying.
Therefore if a computer were programmed to terminate drying when TDAL
dropped to a specific value the lumber could be under-dried (See Figures 5 through
8).
TDAL is a good indicator of the drying rate (Oliveira, 1986). Although
drying rate is dependent of the moisture content of lumber as well as temperature,
relative humidity, and air velocity (Taylor and Mitchell, 1987), Figures 5 through
8 and the values of TDAL relative to MC in Table 2 show that TDAL alone is not
an efficient method of determining the drying end point.
58
TDAL
110
106
100
96
go
86
80
76
70
66
80
66
60
46
40
36
30
26
20
16
10
6
U
Avg. MC
TIDAL (F) / MO (%)
0.6 1 1.6 2 2.6 3 3.6 4 4.6 6 6
8 8.6 7 7.6 8 8.6 0 9.6
10
Drying Time (Hrs.)
Figure 5. A graph of TDAL and average moisture content vs. drying time for
charge 1 of schedule 1.
TIDAL
110
106
100
86
00
86
80
76
70
66
Nk,
-'*
..-4.•
..N.
.--4•:,,.6.
_
80 - •-,_
66
60
46
40
36
30
26
20
16
10
6
0
Avg. MO
TIDAL (F) / MO (%)
-
0
--1
1
_
\\ /.---. ----.-- \ ./..--"-----■- •
/
-4,6
---6.4.
\
-.I.-
V.---\/•
2 3 4 6 8 7 8 9 10 11 12 13 14 16 18
Drying Time (Hrs.)
Figure 6. A graph of TDAL and average moisture content vs. drying time for
charge 1 of schedule 2.
59
TDAL
Avg. MO
110 TDAL (F) / MO (%)
106
100
96
90
86
80
76
70
66
60
66
60
46
40
36
30
26
20
16
10
6
0
0 0.6 1 1.6 2 2.6 3 3.6 4 4.6 6 6.6 8 8.6 7 7.6 8
Drying Time (Hrs,)
Figure 7. A graph of TDAL and average moisture content vs. drying time for
charge 1 of schedule 3.
TDAL
-4-
Avg. MO
76
70
86
80
66
60
46
40
36
30
26
20
16
10
6
0
0 0.6 1 1.6 2 2.6 3 3.6 4 4.6 F 6.6 8 8.6 7 7.6 8 8.6
Drying Time (Hrs.)
Figure 8. A graph of TDAL and average moisture content vs. drying time for
charge 1 of schedule 4.
60
Table 2. Comparison of average final percent MC and average TDAL during
the last hour of drying.
Schedule #
Charge #
MC %
TDAL
1
1
1
1
2
3
10.2
11.0
20.0
17.7
20.5
22.5
2
2
2
1
2
3
11.6
12.0
16.1
10.0
8.1
9.1
3
3
3
1
2
3
11.9
11.8
16.0
39.0
23.3
31.8
4
4
4
1
2
3
11.8
10.2
12.2
24.5
18.9
16.1
LITERATURE CITED
1. Bai, G.L. and P. Garrahan. 1984. The temperature and moisture content in
lumber during preheating and drying. Wood Sci. Technol. 18:121-135.
2. Bird, R.B., W.E. Stewart, and E.N. Lightfoot. 1960. Transport Phenomena.
John Wiley and Sons, Inc., New York. 780 pp.
3. Herzberg, B.L., F.W. Taylor, and H.N. Rosen. 1985. Factors that affect the
time required to high-temperature dry pine dimension lumber. Forest Prod.
J. 35(7/8):34-36.
4. Oliveira, L.C. 1986. Mathematical Model for Control of High Temperature
Drying of Southern Yellow Pine Dimension Lumber. Ph.D. Dissertation.
Virginia Polytechnic Institute and State University. 178 pp.
5. Panow, S., and M.R. Nester. 1985. Drying end point determination during
high temperature drying. Forest Prod. J. 35(1):51-55.
6. Panshin, A.J. and Carl deZeeuw. 1980. Textbook of Wood Technology.
McGraw-Hill. New York. 722 pp.
7. Pratt, Willard E. 1956. Estimating the moisture content of lumber during the
drying process. Forest Prod. J. 6(9):333-337.
8. Price, E.W. and P. Koch. 1981. A note on effects of kiln stick thickness and
air velocity on drying time of southern pine 2 by 4 and 2 by 6 lumber. Wood
and Fiber Sci. 13(2):115-119.
61
9. Shekhtov, K. 1973. Investigation of the distribution of moisture and
temperature in a kiln lumber charge during the drying process. Mekhanichna
Tekhnologiva na D'rvesinata. (No. 19):83-90.
10. Siau, John F. 1984. Transport Processes in Wood. Springer-Verlag. New
York. 245 pp.
11. Steinhagen, H. 1974. Effect of kiln air velocity, at various moisture content
levels on drying rate of 4/4 yellow-poplar sapwood. Forest Prod. J. 24(4):4547.
12. Stevens, W.C., and D.D. Johnston. 1957. Temperature drop across a pile of
green timber and its effects on drying rates. Timber Technology. February.
13. Taylor, F.W., and P.H. Mitchell. 1987. Drying pine at very high temperatures
and air velocities. Wood and Fiber Sci. 19(3):239-245.
14. Wengert, W.M. 1984. Southern pine kiln temperature control with
temperature drop across the load (TDAL) measurements. In: Computer
Automation for Sawmill Profit. Forest Products Research Society Proceedings
7333, Madison, WI.
15. Wengert, E.M. and P.G. Evans. 1971. Automatic programming and control
for steam-heated lumber dry kilns. Forest Prod. J. 21(2):56-59.
16. William, L.J., E.T. Choong, D.G. Arganbright, D.K. Doucet, M.R. Gorvad,
W.L. Galligan and W.T. Simpson. 1984. Forest Prod. J. 34(11/12):59-63.
62