Replacing mechanical testing

with ultrasonic measurements

Roman Popil, Ph.D.

Senior Research Scientist

Georgia Tech/Institute of

Paper Science and

Technology

Atlanta, GA.

Roman@gatech.edu

1

Outline of this talk

• Mechanical testing of paper/board

samples requires sample cutting, handling

leads to errors and high variability

• Ultrasonic based testing requires no

sample cutting, special fixtures, provides

an instant measure of fiber quality, is

related to all strength properties, tensile,

compression, bending

2

Outline of this talk Cont’d

• Here, it will be shown why and how this

works • Some theory: speed of sound, modulus, load

elongation curve,

• Some supporting paper physics literature: paper

moduli and strength, moduli and density

• Examples of data will be shown from

recent lab testing: • Handsheet studies of various enzyme effects

• Commercial linerboard and medium

• Corrugated boards lab-made and commercial

boxes

3

The TSO Ultrasonics Opportunity

• With basis weight times the TSI_CD or

TSI_MD , a good correlation can be

determined for a variety of mechanical

strength properties

• Large sample areas are tested at a time

not to failure, reducing variability, the test

is non-destructive

4

5

Current mechanical tests for

replacement • Ring Crush and Short Span Compression

tests for linerboard

– Ring crush is affected at low and high basis

weights

– SCT has too high a variation

• Edge Compression for corrugated board

– ECT prone to test specimen roll edging,

cutting artifacts, clamping effects

6

SCT is Replacing RCT

Photo: Mike Schaepe, IPST

7

SCT replacing RCT

SCT tracks basis weight, RCT does not and is ~1/2 SCT, but error

bars are smaller for RCT

from Popil Tappi PaperCon 2010

8

9

Short span compression has been shown in the published

literature to be a correct measure of compression strength

RCT

Christer Fellers’ slide from Inventia, Sweden

Increased buckling for lightweights decreases

RCT with density

Measuring SCT on the same sample on 2 different labs, an

example, IPST Paper Analysis lab data

Error bars in c.i. decrease as 1/√n, Tappi standard is n = 10

Overlapping error bars (n=20) indicate no significant difference, but

this high variability (cv %) is bad for quality control

n = 150 !! Mill

measured

value

A mill measured the same sample 150

times in attempt to improve accuracy !

18.42 ± 0.93 (5%)

19.41 ± 1.25 (6.4%)

10

MD

Front or

Tending Side

Back or

Drive Side

Paper from a paper machine has a variation in the MD

and the CD, CD variation arises from drying profiles,

headbox edge flows, MD are high frequency variations,

flocs, wire marks, headbox pulsations, etc.

CD

11

Sources of variation in SCT

Sources of variation in SCT

• CD and MD variation in machine made paper

Sample cutting technique, size, handling

• Paper formation

Inherent paper non-uniformity on 1mm scale

• Moisture variation

Sample conditioning, built-in moisture

compensation calibration

• Instrument issues: calibration and wear

Motor speed, clamping pressure, load cell

calibration

12

But the variation in SCT ~7%, is still too high

for quality control, why so ?

15

mm

160 mm

Strip width is about 2 flocs wide

SCT tests a length only 0.7 mm

long across 1 -2.5 flocs along the

strip length.

Mass formation cv% 6 – 8% , SCT is proportional to mass, mass is

proportional to modulus, therefore SCT variation is inevitable !! 13

Using the TSO, we can minimize

the variation in testing results

• Let’s first look at what is measured in a

mechanical tensile test

= = ?

UTM

tensile

L&W tensile L&W TSO 14

To see what modulus E and tensile stiffness

Et are, look at a tensile test:

t

l

Δl

F

y

ΔF

Δl

Tensile

strength

In a tensile test, a length of sample l, width w is stretched out a

displacement of Δl by a force increment of ΔF.

In the linear first part of load displacement curve of F vs y, we

define modulus E as:

w

15

E =

Elastic modulus and speed of sound

From the mechanical tensile

test – this is Hooke’s law

Relationship between paper density ρ ( = basis weight /t, t caliper) ,

speed of sound V and elastic modulus E is :

The tensile stiffness in a mechanical test is (LHS) :

= Et = ρt x V2 = basis weight x V2

So if we measure the basis weight BW and the speed of

sound squared V2 - we get the tensile stiffness Et

1)

2)

3)

16

Strength and modulus

connections

Tensile stiffness = E x t = Basis Weight x V2

a) Strength properties track basis weight

b) Stronger materials have a higher speed of sound

c) So Basis Weight x V2 will track compression or tensile

strength with greater sensitivity and accuracy

For linerboards and corrugated boards

want to measure VCD2 = TSI_CD on the

L&W TSO

17

Sheet compression

strength

Fiber coarseness Fiber width Fiber density

Fiber modulus Fiber collapsed

thickness

Sheet

density

18

Fig 1. Stress-strain curves for different bond

strengths from the simulated tensile test on

10 x 4 mm2, 27 g/m2 network, S. Borodulina,

KTH (2012). Fig 2 Effect of the bonding agents from D.

Page and R. Seth, lab study (1983).

Experiments and modeling show that modulus E is

principally governed by fiber quality

Seth Page Borodulina

19

• A similar predictive equation is used with continuing success for the

tensile strength “Page” equation

• For corrugated board, the edge compression strength is the summation

of the compression strength of the components, “Maltenfort” equation

• For box compressions strength, the “McKee” equation related the ECT

and bending stiffness to the failure load of a stacked box

• Bending stiffness of paper, board and corrugated board are directly

related to the modulus, “Euler” equations

Connections between modulus and

strength properties

20

Other supporting results relating modulus and

strength properties from the literature on the

following slides…

Wet pressing density increases the modulus of

paper, note that this is more sensitively detected

by sonic measurement

22

Packaging paperboard data (basis weight normalized)

show compression strength modulus relationships

This works

until starch

or other

strength

additives

are put in

the wet end, Waterhouse

Tappi J., 1994.

23

Elastic modulus can also be changed in

papermaking operations by the refining (beating

level, wood species, pulp type

• These effects can be readily monitored by on-

line TSO

24

Table 1. Pulp types and average thickness of the 300 g/m2

boards.

Type Average thickness, μm

PL (unbeaten hardwood) 451.4

PB (unbeaten softwood) 582.0

LEM (commercially beaten hardwood) 324.7

BEM (commercially beaten softwood) 443.0

BS (bleached chemical pulp) 474.6

OS (unbleached chemical pulp) 475.6

CTMP (chemo‐thermo‐mechanical pulp) 736.4

3S (three‐ply board made of OS, CTMP, BS)

Modulus and strength connection in a sheet forming study –

recent example

PACKAGING TECHNOLOGY AND SCIENCE

Packag. Technol. Sci. 2011; 24: 331–341

M. Östlund, BiMaC Innovation, Department of Solid Mechanics, KTH (Royal

Institute of

Technology), Stockholm, Sweden.

25

Tensile strength is

proportional to modulus

26

Verification of the TSO: Comparisons were made

using the IPST ultrasonic lab instrument and an L&W

TSO

1980’s IPST developed robot arm

ultrasonics tester based on a

difference method using a pair of

transducers 1 measurement takes

10 minutes

L&W TSO uses a circular array

transducers to produce a polar plot

of V2 and outputs TSI_CD and

TSI_MD, measurement takes 6

seconds

Using a variety of sample the Specific Stiffness V2 of the

1980’s IPST Robot was compared with TSI IP

ST

in

-pla

ne s

pecific

stiff

ness (

km

/s)2

y = 1.05x

R2 = 0.98

2

4

6

8

10

12

14

2 4 6 8 10 12

IPST V2 (km/s)

2

TS

I_C

D a

nd T

SI_

MD

blotter

WAM 33#

inkjet mylar

yellow copy

3m mylar

20 lb medium

OCC 42lb liner

18 lb medium

16 lb medium

56 lb liner WC

26 lb medium

33 lb medium

42 lb kraft liner

Sample ID

blotter

WAM 33#

inkjet mylar

yellow copy

3m mylar

20 lb medium

OCC 42lb liner

18 lb medium

16 lb medium

56 lb liner WC

26 lb medium

33 lb medium

42 lb kraft liner

Sample ID

CD valuesMD values

Correlation looks ok for a variety of various liners medium, copy paper and

plastic film 28

29

y = 1.7524x - 73.305

R2 = 0.9829

200

400

600

800

1000

1200

1400

100 200 300 400 500 600 700 800

Instron Tensile stiffness (N/mm)

TS

I_C

D x

BW

Ultrasonically

measured values

are always higher

than mechanical

equivalents

Comparisons of this sample set TSI_CD x BW

were made with Instron tests.

Recent IPST client handsheet study, shows correlation of TSO

with SCT, CMT

L&W TSO V2 (aka specific stiffness) x Basis wt = Tensile

stiffness, correlates with SCT and CMT :

CMT =

30

Handsheet lab series strength properties:

CMT & SCT vs BW x V2

31

Sample Caliper Basis Weight TSI-CD CD SCT Ring Crush

(mm) (g/m2) (km/s)2(lb/in) (lb)

Liner A1 0.232 150.1 4.59 18.6 61.5

Liner A2 0.268 171.7 4.88 22.4 76.1

Liner A3 0.305 195.3 5.46 22.1 101.2

Liner A4 0.392 249.5 4.7 30.8 118.4

Liner A5 0.420 275.7 4.9 31.2 122.4

Liner A6 0.476 293.9 4.69 33.5 129.6

Liner B1 0.600 370.4 4.43 46.1 158.4

Liner B2 0.427 277.1 5.13 36.3 126.5

Liner B3 0.725 437.1 3.78 51.2 161.5

Liner B4 0.327 206.3 5.51 26.2 98.5

Liner B5 0.663 393.1 3.96 40.8 160.2

Liner B6 0.260 167.0 5.24 21.7 74.3

Liner B7 0.524 323.8 3.97 34.4 125.4

Liner B8 0.229 149.6 4.55 16.1 62.0

Liner B9 0.322 195.8 5.43 28.6 98.8

Liner B10 0.319 205.4 4.68 22.2 87.8

Liner B11 0.203 128.9 5.13 14.3 39.0

Liner B12 0.259 177.6 4.97 17.2 65.7

medium 1 0.146 78.5 3.58 7.5 15.2

medium 2 0.192 88.3 3.76 8.6 21.9

medium 3 0.197 95.7 4.12 10.0 25.5

medium 4 0.208 126.9 4.21 15.0 45.0

medium 5 0.251 163.7 4.57 22.2 74.1

Example: Collection of commercial linerboards

and medium from US southeast mills (data

collected for Tappi PaperCon 2010 conference paper)

SCT and RCT correlate with L&W (TSI_CD x Basis wt.)

TSI_CD x Basis wt = elastic modulus x caliper = tensile stiffness 32

SCT and RCT correlate with BW x V2 for US southern

medium and linerboards

33

Another recent example: 8 linerboard and

medium samples from a Mexican box plant

Can they replace SCT with TSO ?

Note : average SCT cv% = 7.4 %

SCT trends basis weight as expected

34

• VCD2 shows quality of paperboard, has only 2 %cv, does not trend with

basis weight

• Basis wt x VCD2 tracks SCT

Instead of SCT, TSO and basis weight can be used to

screen incoming materials for best box performance

35

Using TSO to predict ECT of

corrugated boards – a proposal:

• TSI_CD = ECD x t = basis weight x VCD2

• Wavelength λ of ultrasonic sound waves of 100

kHz exceeds thickness t of boards:

• λ = 2.7 km/s/100 KHz ≈ 2.7 cm

• Sound waves propagate along the board through

the whole board

• Therefore ECT ≈ (basis weight x TSI_CD)

36

37

MD

CD

ZD

Direction assignments

for corrugated board

Linerboard single-face

Linerboard double-face

Fluted medium

Corrugated board and ECT basics

ECT test clamping fixtures

“Neck down” sample cutter

ECT is prone to artifact

depending on type of

board and method of

sample prep Rolled edge OK, middle

crease

Bowing

liners

ECT of lab-made A flute boards with different weights of

medium all the same liners weights– TSO detects

changes in medium strength

39

Another example: Copy Paper

Boxes from Southeast Asia

40

Can Ultrasonics Replace ECT ?

(maybe !)

• Shipped boxes submitted by client were

reassembled, tested for BCT, bottom flaps for ECT

• Side panels were ultrasonically tested after BCT

• Used VCD2 x BW for those boards where vertical

loading of the side panels is in the CD

• VCD2 x BW correlated with ECT

41

Correlation of Copy Paper box ECT with

ultrasonics

42

Summary

1. Mechanical testing introduces high

variability, is labor intensive: • SCT is ok, but high cv%

• RCT is popular but is misleading

• ECT is also prone to sample prep artifact

2. Ultrasonic testing (basis weight x V2) can

replace or supplement mechanical testing

with advantages: • No sample prep required

• Fast

• Less variability

43

Thank you ! Send questions, comments

to: Roman@gatech.edu

44