Advances and the future of grading structural timber Prof. - - PDF document

SLIDE 1

Advances and the future of grading structural timber

• Prof. Charlotte Bengtsson

SP Trätek Linnaeus University

WG 3 Strength, Stiffness and Appearance in Quality Control and Processing

• More advanced non-destructive methods and combination of

methods

• More accurate machines at reasonable cost
• Grading early in the production process
• ”Visual strength grading” by means of surface scanning techniques

‘The Future of Quality Control for Wood & Wood Products’, 4-7th May 2010, Edinburgh The Final Conference of COST Action E53 http://cte.napier.ac.uk/e53

SLIDE 2

During this COST action….

• More countries with large interest in grading
• More companies (sawmills and machine producers)
• EN14081
• Gradewood
• A lot of presentations during COST E53 activities

Production of machine graded timber in Sweden

200000 400000 600000 800000 1000000 1200000 1400000 1600000 1989 1994 1999 2004 2009 Production m3 Year Total Export Sweden

‘The Future of Quality Control for Wood & Wood Products’, 4-7th May 2010, Edinburgh The Final Conference of COST Action E53 http://cte.napier.ac.uk/e53

SLIDE 3

Gradewood, results are now coming

New physically based models for strength: WP1 Experimental research and data base: WP2&3 Novel procedures for grading: WP5 New statistically based models for grading: WP4&5 Numerical analysis and simulation Support to standardisation and CE-marking:WP5&6 New business models for wood industries: WP5&6 More accurate grading techniques: WP5

EN- standards for grading

• EN 14081 Timber Structures – Strength graded structural timber with

rectangular cross section, allowing CE-marking of graded structural timber, becomes mandatory September 1, 2012

• This standard replaces EN 519
• Four parts

– Part 1: General Requirements – Part 2: Machine Grading; additional requirements for initial type testing – Part 3: Machine Grading; additional requirements for factory production control – Part 4: Machine Grading; Grading machine settings for machine controlled systems

‘The Future of Quality Control for Wood & Wood Products’, 4-7th May 2010, Edinburgh The Final Conference of COST Action E53 http://cte.napier.ac.uk/e53

SLIDE 4

EN- standards for grading

• ”New” procedure for derivation of machine settings
• Settings applicable for one country or some countries
• Two principles

– Machine control – Output control

• The standard covers both visual and machine strength grading
• Visual override requirements as previously

Example from 14081 - 4

Source country or countries Source mark a) Species Permitted timber sizes

b)

(mm) Grade or grade combination Settings IP (machine units) Comments and additional requirements C24 4 300 000 TR26 C16 6 150 000 4 300 000 TR26 C18 6 150 000 4 300 000 C27 C18 6 150 000 4 420 000 C30 C18 6 900 000 4 300 000 C30 C24 C18 6 900 000 5 770 000 4 650 000 Finland Norway Sweden Estonia Latvia Russiad) Poland Germany Austria Czech Republic FI NO SE ES LV RU PL DE AT CZ Spruce, Picea abies Fir Abies alba 31 ≤ tn ≤ 110 63 ≤ bn ≤ 264

Actual setting, IP is given in machine units and it is not effected by the timber dimensions. Requirements for grading: − Air temperature 10°C – 50 °C − Relative humidity in the air < 85 % − Timber temperature > -10 °C − Timber mean moisture content between 10% and 16% Dynagrade − Conveyor speed ≤ 1 m/sec − Spacing between pieces ≥ 200mm − Grading speed ≤ 100 pieces/min Dynagrade HC − Conveyor speed ≤ 1,3 m/sec − Spacing between pieces ≥ 225mm − Grading speed for widths ≤ 250mm: ≤ 150 pieces/min Dynagrade XHC − Conveyor speed ≤ 1,3 m/sec − Spacing between pieces ≥ 225mm − Grading speed for width (w) ≤ 100mm: ≤ 240 pieces/min where 100 < w ≤ 250 grading speed = 78000/(w+225) pieces/min Where timber has a mean moisture content between 16% and 20% with a minimum value of 14% and a maximum value of 22% then the settings shall be calculated according to equation (1) and rounded to 3 significant digits.

IP IP IP 0738 ,

18

− =

(1)

a) See clause 7.3 in EN14081-1 b) Timber sizes shall be to EN 336. c) Grades prefixed by C are strength classes given in EN 338 d) Settings apply only to timber grown west of the Ural mountain range in Russia

Dynagrade ‘The Future of Quality Control for Wood & Wood Products’, 4-7th May 2010, Edinburgh The Final Conference of COST Action E53 http://cte.napier.ac.uk/e53

SLIDE 5

Machine grading principles

• Measurement of bending stiffness
• Optical detection of knots and other characteristics
• Near-Infrared-Reflection spectroscopy (NIR)
• Resonant vibrations
• Wave propagation speed
• Neural nets
• Deconvolution technique
• Radiation methods

– micro waves – x-ray – γ-radiation

• Combination of techniques

Machine grading principles

Pr e s s r

• l

l e r A i r A i r Di s pl ac e me nt L

• adc

e l l Re f e r e nc e r

• l

l e r Re f e r e nc e r

• l

l e r Comput e r L

• ad

r

• l

l e r T i mbe r

Flatwise bending

• Span ca 900 mm
• Deflection const. measure load
• Load const. measure deflection

Density

γ-ray x-ray Density profile Detectors

Resonant vibrations

• Measures
• length
• frequency
• density

f L E

A A 2 1 2

4

−

⋅ ⋅ ⋅ = ρ

‘The Future of Quality Control for Wood & Wood Products’, 4-7th May 2010, Edinburgh The Final Conference of COST Action E53 http://cte.napier.ac.uk/e53

SLIDE 6

Grading machines on the European market

Name Principle

Computermatic/Micromatic Flatwise bending, measurement of deflection CookBolinders/Techmach Flatwise bending, measurement of load Grademaster Vibration and scanning GoldenEye 702/EuroGreComat 702 X-ray EuroGreComat 704 X-ray and bending GoldenEye 706/EuroGreComat 706 X-ray and vibrations GoldenEye 80/1 X-ray and laser scanner GoldenEye 80/2 X-ray and laser scanner VM Grader 1.0 Visual grading & gravimetric density Dynagrade Vibrations in longitudinal direction Raute Timgrader Flatwise bending, measurement of load Newnes X-ray Ersson ESG-240 Flatwise bending, measurement of load Metriguard 7200 HCLT Flatwise bending, measurement of load Sylvatest Ultrasonic waves

Grading machines on the European market, cont.

Name Principle

Precigrader Vibration and density Timber Grader MTG Vibration VISCAN Vibration Triomatic Ultrasonic waves E-Scan Vibration JRT MSR Machine Bending, measurement of deflection Noesys Vibration Xyloclass T, Xyloclass F Vibration CRP360 Bending Rosgrade Vibration

• Machines with settings according to EN 14081

‘The Future of Quality Control for Wood & Wood Products’, 4-7th May 2010, Edinburgh The Final Conference of COST Action E53 http://cte.napier.ac.uk/e53

SLIDE 7

Why strength grading?

• Accurate knowledge about timber characteristics – strength,

stiffness, appearance

• Have a common classification within a market
• Obtain an engineering material, gives possibilities to develop the

timber building technique

• Optimise the yield

– Adding value – Use of resources – Optimise the use (use good enough quality)

• Timber for structural applications requires grading
• Strength, stiffness and density properties need to be known and

to be controlled to stay within desirable limits

Examples from EN 338

Strength class Characteristic property C18 C24 C30 Strength properties (MPa), 5%-percentile Bending strength

k m

f

,

18 24 30 Tension strength, parallel to the grain

k t

f

. .

11 14 18 Tension strength, perpendicular to the grain

k t

f

. 90 .

0,3 0,4 0,4 Compression strength, parallel to the grain

k c

f

. .

18 21 23 Compression strength, perpendicular to the grain

k c

f

. 90 ,

4,8 5,3 5,7 Shear strength

k

f .

ν

2,0 2,5 3,0 Stiffness properties (MPa) MoE parallel to the grain, mean value

mean

E ,

9 000 11 000 12 000 MoE parallel to the grain, 5%-percentile

05 ,

E

6 000 7 400 8 000 MoE, perpendicular to the grain, mean value

mean

E

, 90

300 370 400 Shear modulus, mean value

mean

G

560 690 750 Density (kg/m³) Density, 5%-percentile

k , 12

ρ

320 350 380 Density, mean value

mean , 12

ρ

380 420 460

• Grades can also be defined elsewhere, ex. LS, L, LD for

glulam laminations

‘The Future of Quality Control for Wood & Wood Products’, 4-7th May 2010, Edinburgh The Final Conference of COST Action E53 http://cte.napier.ac.uk/e53

SLIDE 8

Basics

• Visual grading

–Grading rules, with maximum knot size etc –Trained graders –Scanners

• Machine strength grading

–Prediction of strength properties –Classification into strength classes, i.e C18, C24 etc

Visual grading

• Visual strength grading has a long tradition
• Formal grading rules were established beginning of 20th century (US)
• 1930 and onwards visual grading rules were introduced in Europe

‘The Future of Quality Control for Wood & Wood Products’, 4-7th May 2010, Edinburgh The Final Conference of COST Action E53 http://cte.napier.ac.uk/e53

SLIDE 9

Visual grading

• Basis

– Rules with criteria for each grade – Trained graders

• Advantages

– Simple and cheap – Easy to check the quality

• Disadvantage

– Low capacity – Low yield

• Only visually recognisable characteristics can be used
• Only simple combinations possible

Machine grading

• NDT for grading timber introduced in US and Australia late

1950s

• The ”Machine control” system was developed in Europe in the

late 1960s

• EN 519 was published 1995
• Focus on softwood species
• Machine strength grading is

usually complemented with visual

• verride requirements
• More ”complicated” combinations

possible, also high grades

‘The Future of Quality Control for Wood & Wood Products’, 4-7th May 2010, Edinburgh The Final Conference of COST Action E53 http://cte.napier.ac.uk/e53

SLIDE 10

Principles for machine strength grading

Strength

IP measured by a machine

Två system finns:

• Maskin kontroll
• Out-put kontroll

(resultatstyrd metod)

Strength

IP measured by a machine

Strength

IP measured by a machine

Två system finns:

• Maskin kontroll
• Out-put kontroll

(resultatstyrd metod) R2≈0.45-0.60 for common machines Two systems:

• Machine control
• Output control

Clear Wood – Correlation between physical properties

Strength MOE Density

R²=0,76 R²=0,66 R²=0,64

Density 1500 kg/m³

Clear wood

‘The Future of Quality Control for Wood & Wood Products’, 4-7th May 2010, Edinburgh The Final Conference of COST Action E53 http://cte.napier.ac.uk/e53

SLIDE 11

Timber – Correlation between physical properties

R

2 = 0.63

CV

r = 18%

N = 407 20 40 60 80 100 5000 10000 15000 20000 25000 Modulus of elasticity [MPa] Bending strength [MPa]

40x145 mm Norway spruce Clear wood: R2=0,76

Timber – Correlation between physical properties

R2 = 0,1759 10 20 30 40 50 60 70 80 90 100 100 200 300 400 500 600 700 Density (kg/m³) Bending strength (MPa)

Clear wood: R2=0,66

‘The Future of Quality Control for Wood & Wood Products’, 4-7th May 2010, Edinburgh The Final Conference of COST Action E53 http://cte.napier.ac.uk/e53

SLIDE 12

Timber characteristics

• Clear wood with varying properties

–Low density –High density

• Strength reducing characteristics

–Knots

• 95 % of failures in redwood

(Sequoia S.)

• 91 % of failures in Norway

spruce –Slope of grain –Top failure serious but very frequent –Compression wood not so serious hard to detect –Cracks –Rot

Typical within member variability

150 mm

Clear Wood Area Ratio

‘The Future of Quality Control for Wood & Wood Products’, 4-7th May 2010, Edinburgh The Final Conference of COST Action E53 http://cte.napier.ac.uk/e53

SLIDE 13

Between members variability

5000 10000 15000 20000 1000 2000 3000 4000 5000 Length position (mm) Flatwise MOE (N/mm²)

Less stiff with large knots Stiff with small knots

Radial variation

A B C D E F G H I J K L M N O P Q 6 4 2

200 400 600 Density (kg/m³)

A B C D E F G H I J K L M N O P Q 6 4 2

5000 10000 15000 20000 25000 MOE (N/mm³)

‘The Future of Quality Control for Wood & Wood Products’, 4-7th May 2010, Edinburgh The Final Conference of COST Action E53 http://cte.napier.ac.uk/e53

SLIDE 14

Prediction of strength

Coefficient of determination R² Bending strength Tensile strength Characteristics that can be measured non-destructively Source 1 2 3 4 1 5 6 Knots 0.27 0.20 0.16 0.25 0.36 0.42 0.30 Annual ring width 0.21 0.27 0.20 0.44 0.36 0.33 0.28 Density 0.16 0.30 0.16 0.40 0.38 0.29 0.38 MoE, bending or tension 0.72 0.53 0.55 0.56 0.70 0.69 0.58

• MoE. flatwise, short span

0.74 Knots and annual ring width 0.37 0.42 0.39 0.49 Knots and density 0.38 0.38 0.55 0.61 0.64 Knots and MoE 0.73 0.58 0.64 0.70 0.76 0.78

Grading accuracy

• How well the measured characteristic predicts strength

(R2)

• How accurate the characteristics can be measured

(CoV)

10 20 30 40 50 60 70 80 90 100 5 000 10 000 15 000 20 000 25 000 MOE (MPa) Bending strength (MPa) Central South North inland North Coastal

‘The Future of Quality Control for Wood & Wood Products’, 4-7th May 2010, Edinburgh The Final Conference of COST Action E53 http://cte.napier.ac.uk/e53

SLIDE 15

Effect of grading accuracy

Pr e s s r

• l

l e r A i r A i r Di s p l a c e me n t L

• a

d c e l l Re f e r e n c e r

• l

l e r Re f e r e n c e r

• l

l e r Co mp u t e r L

• a

d r

• l

l e r T i mb e r

Grading C30 and better

10 20 30 40 50 60 70 80 90 100 0,0 0,2 0,4 0,6 0,8 1,0 Coefficient of determination, r2 Yield (%)

CoV=0.4 CoV=0.3 CoV=0.2 CoV=0.1 CoV=0.05 Prediction using MOE Prediction using knots

Output Control

• Used in USA and Australia
• Preliminary setting based on limited testing (60 pieces per

species, grade and dimension)

• Proof-loading of limited number of pieces per shift (5 pieces

per grade and dimension per shift)

• CUSUM control procedure to follow for controlling the grading
• If the grading is “in control” adjustment of the settings is

allowed to increase the yield (requires extra proof-loading)

‘The Future of Quality Control for Wood & Wood Products’, 4-7th May 2010, Edinburgh The Final Conference of COST Action E53 http://cte.napier.ac.uk/e53

SLIDE 16

Example

R2 = 0,43 0,0 10,0 20,0 30,0 40,0 50,0 60,0 70,0 80,0 90,0 100,0 2 000 000 4 000 000 6 000 000 8 000 000 10 000 000 Grading Parameter Bending strength (MPa) C18 4 300 000 C30 6 480 000

Future development

• ”Input control” (=machine control+output control)
• Introduction of scanner systems for automatical visual grading

and for visual override requirements

• Grading in raw condition
• Grading of logs
• Pregrading of logs (scanning)
• ”Industrial tomogaphy”
• Combination of measurements
• Development of the new standard 14081 (raw material regions

and methodology)

• Proof loading
• New machines

‘The Future of Quality Control for Wood & Wood Products’, 4-7th May 2010, Edinburgh The Final Conference of COST Action E53 http://cte.napier.ac.uk/e53

SLIDE 17

Challenges

• Find funding for new R&I-projects
• Make a ”better” standard för graded timber
• Find better grading methods and procedures so that

timber can be used in a safe way for demanding structures

Thanks for listening!

charlotte.bengtsson@sp.se

‘The Future of Quality Control for Wood & Wood Products’, 4-7th May 2010, Edinburgh The Final Conference of COST Action E53 http://cte.napier.ac.uk/e53