In-process Strain
Measurement in Roll
Forming
Florian Kern
N. Stiegler
Prof. Thomas Neitzert
Roll Forming
• Bending process
• Angle introduced
continuously along straight line
• Set of contoured rolls
• Strip motion applied by rotation
of rolls (friction)
• Alternatively, pulling of strip
• Unlimited length
Source: Schuler
Motivation
• Geometric defects a frequent
occurrence in roll forming
• Largely caused by plastic
deformation outside the
intended forming zone
• Verification of FEA-simulation
of the process
Continuous, non-contact strain
measurement in flange area
Source: Halmos
Strain Measurement Methods I
Mechanical
Strain Gauge
Source: Ferber
Mechanic longitudinal
strain device
Source:
Wikipedia
Electric conductor changes resistance
when being compressed or elongated
Strain Measurement Methods II
Photoelasticity
iso-chromatic and
isoclinic lines in a
specimen under
load
Source: Onera
Fibre Bragg Grating
Moiré Pattern
Moiré effect of two rotated
superimposed line pattern
Laser Speckle
Source:
Wikipedia
Source: Marti
 
1
1  pe
 

  B       T 
 B

Source: Thurner
Principle geometry for strain measurement
with objective Speckles
Strain Measurement Methods IV
X-ray Diffraction

Optical
1 v
v
  sin 2   
E
E
Source: Ziebs
Vickers micro hardness indentations
Source: Tönshoff
Line scan
sensor with one
pixel line
parallel e.g.
laser or etched
lines
α
α … angle between
sensor and mark
Set-up to measure strain in the
surface zone of a specimen
Line camera
Strain Measurement Methods IV
Grid Analysis
Etched (left) and laser (right) grids shown
at 50x magnification, circle radius of 2.5mm
Undeformed Ø 2.5mm laser grids
with straight laser lines
Deformation of a circle grid
emajor (%) 
a  d0
 100
d0
emin or (%) 
b  d0
 100
d0
Source: Hsu
Fitted ellipses
a: automatic grid acquisition
b: elliptic grid for a MRA
Strain Measurement - Conclusions
Balance between accuracy and cost
Compromise: Strain gauge
laborious preparation, but:
•continuous data
•inexpensive (in comparison)
•well established
•accurate (compared to other inexpensive solutions)
•delivers full set of data (3 directions)
Problem: How to acquire bending strain with one side of the strip inaccessible?
Spacer Plate – Principle I
  II   I  
 I
    I  z I II
 O z   z
z II  z I
 z II  z I  



Spacer Plate – Principle I
Strain gauge 2
Spacer plate
Strain gauge 1
rn
ra
sheet
ri + zI
ri + zII
ri
ri … inner radius
rn … neutral axis radius
ra … outer radius
b … arc length
zII … distance between lower sheet surface and gauge pattern 2
zI … distance between lower sheet surface and gauge pattern 1
  II   I  
 I
    I  z I II
 O z   z
z II  z I
 z II  z I  



Spacer Plate – Principle II
Trend of the strain in a cross-section
of sheet and spacer plate
Strain measurement from
one side of the sheet
1.1
Design of Spacer Plate
ABS spacer plate
Simulation of Spacer Plate I
tensile
bending
tensile +
bending
Simulation of the strain layout during deformation with and without
applied spacer plate on the sheet
Simulation of Spacer Plate II
Strain in x direction (xy-plane, centre)
3.5N shear force, 1625N tensile force,
14mm long spacer plate
Accuracy of Measurement
1.5E-03
5.E-03
top side of the sheet
gauge between sheet
and spacer, strain 1
1.0E-03
4.E-03
gauge on top of the
spacer, strain 2
5.0E-04
3.E-03
bending strain
bending strain
bottom side of the sheet
0.0E+00
-5.0E-04
2.E-03
1.E-03
-1.0E-03
0
2
4
6
8
10
time [s]
12
14
16
18
20
0.E+00
Bending, one strain gauge on each side
-1.E-03
0
5
10
15
20
25
30
35
time [s]
Bending, two strain gauges and spacer
plate on one side of the specimen
Bending strain can be measured with accuracy of ±3%
40
Spatial Constraints in Roll Former
2
1
Application of strain gauge device
Measured Longitudinal Strain
Simulated Longitudinal Strain
Simulated flange
strain
inrolling
rolling
Simulated flange
strain in
directiondirection
0.005
LE sheet
LE spacer
3 per. Mov. Avg. (LE spacer)
3 per. Mov. Avg. (LE sheet)
0.004
0.003
true strain
0.002
0.001
0
-0.001
-0.002
-0.003
-0.004
50
100
150
200
250
time
300
350
Longitudinal Strain
LE sheet
of simulated and measured flange strain in rolling direction
Overlay of Overlay
simulated
and measured strain in
rolling direction
LE spacer
0.005
rolling direction, between sheet and spacer
rolling direction, top of spacer
3 per. Mov. Avg. (LE spacer)
3 per. Mov. Avg. (LE sheet)
0.004
0.003
true strain
0.002
0.001
0
-0.001
-0.002
-0.003
-0.004
50
100
150
200
250
time
300
350
Principal Strain
 A  C
1
2
2
2  A   B    B   C 
2
2
  2 B   C
tan2  A
 A  C
  , 

Principal Strain - Measured
Roll forming specimen 3, principle strain
0.008
εI between sheet and spacer
εII between sheet and spacer
0.006
εI on top of spacer
εII on top of spacer
0.004
strain
0.002
0.000
-0.002
-0.004
-0.006
-0.008
0
1
2
3
time [s]
4
5
6
7
Principal Strain - Simulated
Principal strain (simulated)
3 per. Mov. Avg. (LEP2neg2)
3 per. Mov. Avg. (LEP1 pos)
3 per. Mov. Avg. (LEP1neg2)
3 per. Mov. Avg. (LEP2 pos)
0.006
0.004
0.002
Strain
0
-0.002
-0.004
-0.006
-0.008
1
37
73
109
145
181
217
253
Time
289
325
361
397
433
469
505
Summary and future work
• Review of strain measurement methods
• Design of strain gauge device that acquires all desired data
from one side of the strip
• Agreement between measurements and simulation
• Gather data under conditions that generate geometric
deviations
• Develop rules for corrective
intervention