Post on 28-Mar-2018
transcript
Recovery of Simulated Recovery of Simulated Sawn Logs with SweepSawn Logs with Sweep
and Ovalityand Ovality
Robert A. MonserudPNW, Portland, OR
Christine Todoroki FRI, Rotorua, NZ
The Problem
• Todoroki 1998: “Not all logs are straight.”
• If curve-sawing not available, need to quantify expected product loss due to sweep
• Difficult to obtain a balanced sample of logs with sweep
• Unable to break confounding between sweep and other factors
The Solution: Sawing Simulation
• Digitize a representative sample of logs– Location & size of all knots, defects
• Systematically bend digitized logs (parabola)
• All logs retain original branching structure– Number, size, shape, location at pith
• Saw digital logs into boards with a sawing simulator: AUTOSAW
Benefits of sawing simulation• Sawing parameters can be held constant• Log variables, such as sweep, can be
examined in isolation of other confounding factors
• Logs can be repeatedly sawn in different ways
• Able to explore the full range of variation• Experimental Design is balanced
Material
• 52 Western Hemlock logs (Tsuga heterophylla)– All knots and defects
measured and mapped
• Add sweep in 1-inch increments (16 times)– Bend in center of 16-ft
logs (uniform)– Bend 4-ft from end
(non-uniform)– 33 sets of 52 logs =
1716 observations
Effect of sweep on conversion
0 5 10 15 20 25
Small end diameter (inch)
0
20
40
60
80
100
Con
vers
ion
(%)
0" sweep4"8"16"
Total Lumber Value($ per Log)
0
20
40
60
80
100
120
0 5 10 15 20 25
Log small end diameter (inch)
Tot
al lo
g va
lue
($U
S)
Straight
Sweep
Proportion of Select Structural
0
0.1
0.2
0.3
0.4
0.5
0.6
0 5 10 15 20 25
Log small end diameter (inch)
Prop
orti
on S
elec
t Str
uctu
ral
StraightSweep
Average Lumber Value($ per MBF)
250
275
300
325
0 5 10 15 20 25
Log small end diameter (inch)
Ave
rage
Lum
ber
Val
ue($
US/
MB
F)
StraightSweep
Log value & volume due to rotation: Straight vs Swept
100110120130140150160170
0 60 120 180 240 300 360
Log rotation (degrees)
Tot
al L
og V
alue
($U
S)
400
450
500
550
Lum
ber
Vol
ume
(BF)
Straight $ Curved $ Straight BF Curved BF
Log 6211091 SED = 20 in.
Results• Recovery of straight logs = 59 %
– (Volume of boards = 59% of log volume)
• Recovery declined 2.4% for each 1-inch of sweep per 16-foot log
• Declined 10% for each 4-inch of sweep• Trend was linear• Intercept increases with diameter• Variation was large and constant
– (CV = 25%)
Results: Ratio of Sweep to small-end Diameter: s/d
• Recovery declined nonlinearly with s/d– Nearly linear when s/d < 1– Slope is -3.2% for each 0.1 s/d
(-7% and -5% in two other studies)
– Rather tight relationship (R2=89%)
– No additional variation due to diameter
0
10
20
30
40
50
60
70
80
90
100
0.0 0.5 1.0 1.5 2.0 2.5 3.0 3.5
Recovery % by Sweep, Grade, & Sweep Location
0.0
25.4
50.8
76.2
101.
6
127.
0
152.
4
177.
8
203.
2
228.
6
254.
0
279.
4
304.
8
330.
2
355.
6
381.
0
406.
4
Sweep deflection (mm)
-75
-50
-25
0
25
50
75
Con
vers
ion
(%)
Uniformly sweptGrade 4Grade 3Grade 2Grade 1Nonuniformly sweptGrade 4Grade 3Grade 2Grade 1
Conclusions
• Expected trend of decreasing recovery % with increasing sweep was found– Trend was linear– Variation largely due to log size (diameter)
• Relation between recovery % and s/d(sweep/diam) was exponential decay, not a constant rate.
• Value loss ($/Vol) was also exponential decay
Conclusions
• Straight logs have higher value than swept logs
• Volume recovery is the main reason
• Differences in grade yield are a secondary reason
• More wane from curved logs is probably the cause
Conclusions
• Sawing discrete boards is a step function– Very sensitive to small changes in
initial set-up– Large and essentially constant
amount of variation always present
• Sawing simulation a useful tool for analyzing variation
Citation:
• Monserud, R.A, Parry, D., Todoroki, C.L. 2004. Recovery of Simulated Sawn Logs with Sweep. New Zealand Journal of Forestry Science 34(2): 190-205.