InnoTesting 2015 Active Thermography Application of active thermography in production processes Christiane Maierhofer, Mathias Ziegler, Rainer Krankenhagen, Philipp Myrach, Florian Jonietz Bundesanstalt für Materialforschung und –prüfung FB 8.7 Energy Infrastructure Environment Material
This document is posted to help you gain knowledge. Please leave a comment to let me know what you think about it! Share it to your friends and learn new things together.
Non-stationary heat conduction process in anisotropic solids is described by the proportionality of heat flux and temperature gradient using the heat equation (diffusion equation):
T: temperature: position vector
t: time
Heat equation
Solution of parabolic DE:• Spatial boundary conditions
(temperature or heat flux) have to be set
• Temporal starting conditions, i.e. temperature distribution at t = 0, have to be set
• Strong attenuation
Comparison to solution of hyperbolic DE (wave equation):• Spatial boundary conditions have to be
set• Two temporal starting conditions:
temperature distribution at t = 0 and first derivative to time
Cooperation with BAM 8.5 and Benteler SGLEC project Thermobot (www.thermobot.eu)
11
2 cm∆x = 81 µm
∆x = 240 µm
InnoTesting 2015 Active Thermography
Comparison of flash thermography and CT, specimen R 3/4
3rd layer
x = 1.28 mm
Set-up:2 plates glued with 4 beadsDefects:Broken fibre bundleBeads:1: release agent2: without sand blasting3: o.k.4: wrong mixture
TT
CT
R
Phase image from mid
x = 2.1 mm
T3
24
1
3
2
1
C. Maierhofer et al., Composites Part B, 64, 2014, 175-186
12
2 cm
Validation at CFRP test specimen
InnoTesting 2015 Active Thermography
Theory of lockin excitation
13
Solution for periodic heating(lockin excitation, thermal waves)
Penetration depth (or thermal diffusion length ) of thermal wave: • decreases with increasing frequency• Increases with increasing diffusivity
Phase velocity of thermal wave: • Increases with increasing frequency and
with increasing diffusivity
>> thermal waves have a high dispersion and high at tenuation>> broadening of heat impulses (multi frequency)
)4//(/0),( πω −−−− ⋅= µztiµz eeTtzT
ωα2=µ
ωαω 2=⋅= µv
1D, for semi infinite isotropic bodies
effusivity
diffusivity
cρλε =
cρλα =
InnoTesting 2015 Active Thermography
Comparison of flash/lockin thermography
14
time
A
ϕ
FT(fA)
time
FT(f)A
ϕ
Flash thermography Lockin thermography
• Thermograms at raising or max. contrast• Pulse-Phase-Thermography (PPT)• Thermal signal reconstruction (TSR)
• Online FFT or 4 point method at excitation frequency
• Offline FFT at excitation frequency
f
InnoTesting 2015 Active Thermography
Comparison: 4.4 J impact damage in CFRP
15
Frequency in Hz
1 Hz 0.1 Hz 0.05 Hz0.45 mm 0.63 mm 2 mm
Depth in mm
• Phase images at 1 Hz are similar• Phase images at lower frequency have less contrast for lockin
excitation• Possible reason: Material separation due to thermal expansion
is larger for flash excitation>> Comparison to laser speckle interferometry/shearography
C. Maierhofer, ECNDT 2014, Keynote, www.ndt.netCFRP samples from ZFL Haldensleben, Prof. Dr. J. Häberle, ZIM Kooperationsnetzwerk
InnoTesting 2015 Active Thermography
Comparison: Cu-Shunts am LHC, CERN
2 mm Cu sheet
3 mm Cu sheet
Flash
Lockin1 Hz20 P.
Flash
Lockin1 Hz20 P.
www.cern.ch
Cu-Shunts are supporting the connection of two cable heads
Customer: CERN, C. ScheuerleinC. Maierhofer et al, NDT&E International, 52, 2012, 103-111
16
InnoTesting 2015 Active Thermography
New excitation sources: LED -Array
17
Stand atControl 2014
LEDcontrol Computer
IR camera
Photo-diode
Sample
M. Ziegler, MNPQ project, 2013-2014
Advantages of LED excitation source
� Can be modulated up to 1 kHz
� Narrow spectrum, no overlap with IR camera
InnoTesting 2015 Active Thermography
Current: Pores in aluminium
18
Al pressure casting: rear axle carrier of the chassis of BMW i3
http://www.springerprofessional.de/, Hengst, BMW
Cooperation with Bergakademie TU Freiberg
Thermogram after flash
Phase image at 4.6 Hz
Section of a crankcase, Length150 mm, thickness 2 mm
KCT fromBAM 8.5
Lockin excitation at 8 HzLaser widening
U. Richter, C. Maierhofer, R. Mischke, M. Röllig, Int. Foundry Research, 2015, accepted
InnoTesting 2015 Active Thermography 19
Flash thermography Lockin thermography
Advantages:• Very short measurement time• Multi spectral
>> all penetration depth until Lmax
Disadvantages:• Strong thermal strain• less SNR especially at higher
penetration depths
Advantages:• Less thermal strain• Higher SNR at larger penetration depth• Higher frequencies (LED, LASER)• Depth selective
Disadvantages:• Much long measurement times• Each measurement is only optimized for one
penetration depth
time
A
ϕ
FT(fA)
time
FT(f)A
ϕ
f
Comparison of flash/lockin thermography
InnoTesting 2015 Active Thermography
Cracks and local excitation
20
Characterisation of vertical cracks by heating with a laser spot
Joachim Schlichting, Dissertation, TU Berlin, 2012Adolf-Martens-Award 2013
Static laser spotDetermination of crack depth and crack angle is possible
Moved laser spotFast detection of shallow and
narrow cracks is possible
time
InnoTesting 2015 Active Thermography
Test parameters:• Scan speed: 12 mm/s
• Trace distance: 0.5 mm
• Scan time: 40 s• Power density: ~0.7 kW/cm² (9 W, 1.3 mm Spot Ø)
• Camera resol.: 39 µm/pixel
Calibration block 1 (aka MTU Nr. 3) Ø 50 mm:• Reference for MT (fluorescent) EN ISO 9934-2 Annex B • 90MnCrV8 steel, hardened, browned• Stress corrosion cracks and grinding cracks:
width: 0.1-5 µm, depth: 2 µm-1 mm, approved by microscopy
Reference specimen for magnetic particle testing
Sub-µm cracks are recognized! Below diffraction limit (~5 µm)
21
Natural cracks
P. Myrach, M. Ziegler, C. Maierhofer, M. Kreutzbruck, AIP conference proceedings, QNDE, 1581, 2014, 1624-1630
INS Project Laser-Thermography M. Ziegler
InnoTesting 2015 Active Thermography
Testing of large scaled structures
22
Objective:Testing and
inspection withoutscaffolding and
access aids
Building facades
Funded by
Rotor blades of windmills
VIP-Project IKARUS, BMBF R. Krankenhagen
InnoTesting 2015 Active Thermography
Testing of rotor blades
23
Cross section of a rotor blade
leading edge
trailing edge
beam
foamFull GFRP
Adhesive defects
Which structures and which defects and inhomogeneities can be detected with active thermography?
18 to 22 °C
Project IKARUS (Funding program VIP of BMBF, FKZ 03V0135, 2011-2015):Infrared-Kameratechnology for non-touching Analysis of Rotor blades under Offshore conditions
InnoTesting 2015 Active Thermography
10 p.m.
~ 180 cm
~ 20
0 cm• Segment of a rotor blade with natural and
artificial defects and inhomogeneities• Comparison of measurements and numerical
simulations under different environmental conditions (solar radiation, air temperature, background radiation)
• Culture house in Cobbelsdorf, Coswig• Mural painting Industrialized agriculture of
Erich Enge 1970/71• Beam construction• Sun radiation
from 150 W/m2 to850 W/m2 on 9.9.2014
• Time: 10:45 to 15:50 o’clock
2nd phase image of the sequence
10th phase image of the sequence
steel beams
cracks
26
InnoTesting 2015 Active Thermography
Standardisation
27
DIN 54192Active thermography
Draft standardflash thermography
Structure of draft standardlockin thermography
DIN NA 062-08-27 AA Visual and thermographic testing
CEN TC 138/WG 11 Infrared and thermographic testing
prEN 16714-1Part 1 Gener-al principles
prEN 16714-2Part 2 Equip-ment
prEN 16714-3Part 3 Terms and definitions
Active thermography
EMRP VITCEA: Validated inspection methodsfor composites in energy applications
Pre Normative EMPIR-Proposal of EURAMET Focus Group on Standardization of Thermal Imagers (from 2016): Calibration of infrared cameras, crack detection