23/05/2014
1
Managing Wood Quality Variation:
Is Segregation the Solution?
Scion Wood Quality Workshops – May 2014
Dr. Glen Murphy
Definitions of Quality
Scion Wood Quality Workshops – May 2014
The characteristics of a product or service that bear on its ability to satisfy stated or implied needs.
American Society for Quality
Quality in a product or service is not what the supplier puts in. It is what the customer gets out and is willing to pay for.
Peter Drucker
23/05/2014
2
Why Segregate?
Scion Wood Quality Workshops – May 2014
Segregation will lead to a greater financial net benefit for the party segregating the product than if it did not do so - the benefits outweigh the costs.
Segregate Don’t Segregate
Benefits
CostsCosts
Benefits
Roseburg Forest Products
Scion Wood Quality Workshops – May 2014
2 km
LVL, engineered wood, plywood, composite panels, decorative laminate, lumber (structural and appearance), chip, wood fuel pellets, landscape timbers.
Sources logs from its own forest plus buys stumpage and logs from other suppliers.
23/05/2014
3
Why Segregate?
Scion Wood Quality Workshops – May 2014
9%
19%
39%
10%
32%
6%
35%
45%
38%
39%
34%
31%
9%
4%
4%
11%
2%
15%
0%
10%
20%
30%
40%
50%
60%
70%
80%
90%
100%
A B C D E F
G3
G2
G1
$$$
Volu
me R
ecovery
Dryer Grade Veneer ResultsDryer Grade Veneer ResultsDryer Grade Veneer ResultsDryer Grade Veneer Results
Variability between Washington Douglas Fir sites
Measuring Stiffness with Acoustics
Scion Wood Quality Workshops – May 2014
7 Stands, 1400 stems, > 3000 logs
23/05/2014
4
Scion Wood Quality Workshops – May 2014
0
10
20
30
40
50
60
70
80
90
100
G H I J K L M
Ven
eer
Rec
ove
ry %
G3
G2
G1
Why Segregate?
Variability between Oregon Douglas Fir sites
Veneer Recovery
Scion Wood Quality Workshops – May 2014
G1G2% = 48.9V - 143.3
R² = 0.91
0
10
20
30
40
50
60
3.5 3.6 3.7 3.8 3.9 4.0 4.1
G1
/G2
Vo
lum
e R
eco
very
(%
)
Average acoustic velocity (km/sec)
G1/G2 Veneer Recovery vs Acoustic Velocity
23/05/2014
5
Return to Log Values
Scion Wood Quality Workshops – May 2014
$/MBF = 327.17x - 188.42
R² = 0.62
950
1000
1050
1100
1150
1200
3.50 3.60 3.70 3.80 3.90 4.00 4.10
Tota
l Ne
t R
eve
nu
e (
$/m
bf)
Average acoustic velocity (km/sec)
Return to Log Value vs Acoustic Velocity
Logs sorted based on velocity
Scion Wood Quality Workshops – May 2014
Dynamic MOE = f(density*velocity2)
23/05/2014
6
In-line acoustic segregation
Scion Wood Quality Workshops – May 2014
In-line acoustic segregation
Scion Wood Quality Workshops – May 2014
23/05/2014
7
Hitman PH330 on processor head
Scion Wood Quality Workshops – May 2014
● Log segregation within stands can provide significant financial benefits.
● Segregation power is improved by using an index based on TOF acoustic velocity and stem parameters.
PH330 validation in Oregon
Scion Wood Quality Workshops – May 2014
� Waratah 624 log-making on roadside trial site
� Single threshold study in 55 year old D fir stands
� Logs measured with PH330 and HM200 on site
� Results showed that a stand with average log velocity 3.72 km/sec could be upgraded by selecting the higher stiffness logs
Selected Equivalent to ‘Camp run’ with mean V
80% 3.78 km/sec
50% 3.84 km/sec
30% 3.90 km/sec ~$11/m3
23/05/2014
8
Segregate Where ?
Scion Wood Quality Workshops – May 2014
● In forest� At stump� On landings
● At super skids● At central processing yard● At mill
Segregate When ?
Scion Wood Quality Workshops – May 2014
● Prior to harvest● During harvest● After harvest● After delivery
Timing has implications for cost and market flexibility
23/05/2014
9
Who Segregates ?
Scion Wood Quality Workshops – May 2014
● Seller● Buyer● Seller or buyer’s agent
Segregate What?
Scion Wood Quality Workshops – May 2014
• Length
• Diameter
• Stiffness
• Stability
• Compression wood
• Spiral grain
• Propensity to check
• Sweep
• Knot size and location
• Resin pockets
• Density
• Juvenile wood content
• Ring width
Focus on actual performance propertiesFocus on actual performance propertiesFocus on actual performance propertiesFocus on actual performance properties
23/05/2014
10
Segregate How?
Scion Wood Quality Workshops – May 2014
LengthLengthLengthLengthDiameter over barkDiameter over barkDiameter over barkDiameter over bark
Segregate How?
Scion Wood Quality Workshops – May 2014
Branch size aids
Knot Size
23/05/2014
11
Segregate How?
Scion Wood Quality Workshops – May 2014
Whorl location
detected by laser
scanning
Knots detected from scanning
of digital images
3D Sweep
Scion Wood Quality Workshops – May 2014
Why DBH and height are not good enough metrics for some stands!
23/05/2014
12
Terrestrial laser scanning
Scion Wood Quality Workshops – May 2014
Stand Value
$$$
Log Product Yields
Return to Log Values
Scion Wood Quality Workshops – May 2014
Value based on lumber plus chip sales minus mill costs.Value based on lumber plus chip sales minus mill costs.Value based on lumber plus chip sales minus mill costs.Value based on lumber plus chip sales minus mill costs.
Sweep reduced value by 10 to 30%
23/05/2014
13
Segregate How ?
Scion Wood Quality Workshops – May 2014
Finnish Forestrix Finnish Forestrix Finnish Forestrix Finnish Forestrix Project Project Project Project tested GPS, IMU, and 2D and 3D laser sensors for determining where “harvester” and trees are, tree size and tree
form.
Also provide some measures of tree quality ?
Real-time sorting with NIR
Scion Wood Quality Workshops – May 2014
23/05/2014
14
NIR Principle
Scion Wood Quality Workshops – May 2014
Density measurement with NIR
Scion Wood Quality Workshops – May 2014
350 400 450 500
NIR basic density (kg/m3)
350
400
450
Mea
sure
d ba
sic
dens
ity (
kg/m
3)
Green ChipsGreen ChipsGreen ChipsGreen Chips
Density predicted from Density predicted from Density predicted from Density predicted from NIR measurements of NIR measurements of NIR measurements of NIR measurements of green chainsaw chipsgreen chainsaw chipsgreen chainsaw chipsgreen chainsaw chips
23/05/2014
15
Chips are internal wood sample!
Scion Wood Quality Workshops – May 2014
Segregate How ?
Scion Wood Quality Workshops – May 2014
0° 10° 20° 30° 40°
Mean microfibril angle M
Stif
fnes
s of
the
cell
wal
l, G
Pa
50
40
30
20
10
14
11.2
8.4
5.6
2.8 Stif
fnes
s (G
Pa)
of w
ood
of d
ensi
ty 4
20 k
g m
-3
Good wood
Poor wood
0° 10° 20° 30° 40°
Mean microfibril angle M
Stif
fnes
s of
the
cell
wal
l, G
Pa
50
40
30
20
10
0° 10° 20° 30° 40°
Mean microfibril angle M0° 10° 20° 30° 40°
Mean microfibril angle M
Stif
fnes
s of
the
cell
wal
l, G
Pa
50
40
30
20
10
14
11.2
8.4
5.6
2.8 Stif
fnes
s (G
Pa)
of w
ood
of d
ensi
ty 4
20 k
g m
-3
14
11.2
8.4
5.6
2.8 Stif
fnes
s (G
Pa)
of w
ood
of d
ensi
ty 4
20 k
g m
-3
Good wood
Poor wood
Longitudinal shrinkage vs MoE
0
0.2
0.4
0.6
0.8
1
1.2
1.4
2 4 6 8 10 12 14 16
Bending MOE, GPa
Leng
thw
ise
shrin
kage
, %
Unstablewood
Stablewood
■Crook
○ Bow
X Twist
Acoustic velocity and MOE are indicators of STABILITY
– strong correlations with microfibril angle, longitudinal shrinkage, and distortion
Speed (ft/s)
Board
warp
(in
)
23/05/2014
16
Segregate How ?
Scion Wood Quality Workshops – May 2014
• Internal� heart wood / sapwood� internal pruned knot whorls
(internal checks?)� stiffness and strength� density� moisture content� stability� resin blemishes� propensity to check
Radar, CT Radar, CT Radar, CT Radar, CT log scannerslog scannerslog scannerslog scanners
NIR, NIR, NIR, NIR, AcousticsAcousticsAcousticsAcoustics
Visual Visual Visual Visual (external)(external)(external)(external)
Increment coresIncrement coresIncrement coresIncrement cores
Staying Segregated
Scion Wood Quality Workshops – May 2014
Radio FrequencyRadio FrequencyRadio FrequencyRadio Frequency
ID tagsID tagsID tagsID tags
Source: Korten & Kaul 2008
Tracking logs through the supply chain
23/05/2014
17
Staying Segregated
Scion Wood Quality Workshops – May 2014
Concluding Comments
Scion Wood Quality Workshops – May 2014
● Segregation will occur when the financial benefits outweigh the costs.
● When, where, and who segregates stems and logs will affect what is segregated and how segregation occurs.
● What qualities need to be segregated is likely to be customer dependent.
● New sensor technologies are providing a range of new tools for determining how segregation is done.
● The earlier in the supply chain segregation is carried out the greater is the need for standardised tools and procedures – one or a few tools will have to do it all.
● Significant research is required to answer these who, what, why, when, where and how questions.