SPIE Smart Structures/NDE March 11, 2010, San Diego, CA Session 10: Civil Infrastructure Health Monitoring I Damage Inspection of Fiber Reinforced Polymer-Concrete Systems using a Distant Acoustic-Laser NDE Technique Tzu-Yang Yu a and Robert Haupt b a Assistant Professor, Ph.D. Department of Civil and Environmental Engineering University of Massachusetts Lowell b Technical Staff MIT Lincoln Laboratory
Outline Motivation and Objective Delamination/debonding problem in multi-layer fiberglassconcrete systems Distant Inspection Acoustic-Laser NDE Technique Experimental Result Summary
Motivation and Objective Deterioration and degradation of civil infrastructure
Motivation and Objective Sudden failures of civil infrastructure systems Significant impacts EX: I-35 Highway Bridge Collapse, Minneapolis, Minnesota (6:05pm, Wed., Aug. 1, 2007) (Source: Security camera by the Army Corps of Engineers) (Source: www.gettyimages.com)
Motivation and Objective Deterioration/degradation is inevitable, but sudden failure must be prevented. Among various strengthening and repairing techniques, externallywrapped strengthening technique provides a rapid and effective solution. (Source: Fyfe Co. LLC, 2005) After strengthening, a multi-layer fiberglass-concrete system is formed. à Less ductile than the original reinforced concrete system
Motivation and Objective Delamination/debonding in a strengthened reinforced concrete beam: FRP reinforcement bonded to soffit Anchorage with bolts Debonding from laminate end Debonding from flexuralshear crack FRP(fiber reinforced polymer/plastic)
Motivation and Objective Delamination/debonding in a strengthened reinforced concrete column: a) Bond delamination between plies b) Overlap joint debonding [Au (2001)]
Motivation and Objective Strengthening techniques are used For new constructions to upgrade their design capacity For damaged structures to restore their design capacity Inspection needs: Need to determine the level of strengthening Need to evaluate the quality of strengthening construction Need to monitor the long-term performance of the strengthened system Objective: To develop a distant/standoff technique for the inspection of delamination/debonding
Inspection scheme: Acoustic-Laser NDE Laser Vibrometer Backscattered acoustic waves from voids and unbonded regions cause resonances in FRP Reflected Acoustic waves Incident Air FRP unbonded area/ delamination void/ crack Rayleigh wave P-wave S-wave Concrete
Acoustic-Laser NDE Proposed acoustic-laser NDE technique Is a standoff inspection technique Has a high powered parametric acoustic array (PAA) that can excite the structure from ranges exceeding 30 meters Has a laser vibrometer that collects the surface dynamic signature of the multi-layer structure Principle: Dynamic signature of an intact multi-layer system is different from the one of an damaged multi-layer system.
Acoustic-Laser NDE Simplified models of delamination and concrete cracking: L L FRP L unbonded/ delaminated void/cracks concrete epoxy layer λ / 2 λ / 2 Rayleigh waves
Acoustic-Laser NDE Theoretical basis: Difference in natural frequencies of the damaged and intact regions Governing equation of an intact region 2D beam model: y y EI A ky x t 4 2 + ρ + = 0 4 2 where E = Young s modulus, I = moment of the inertia, ρ = density of the material (fiberglass), A = cross sectional area, y(x,t) = transverse displacement of the beam at position x and time t, and k = distributed stiffness coefficient characterizing the connection between FRP and concrete. Governing equation of a damaged region (clamped beam): y EI x y t 4 2 + ρ A = 0 4 2
Acoustic-Laser NDE Natural frequencies of the intact and damaged regions: Intact ( ) 2 2 2 K L d φ L 2 L i x dφ i i x ωi = = φ 2 ( ) 0 0 i ρ M + 0 i dx dx where M i = the generalized mass of the i-th mode, and φ i (x) = shape function. Damaged (with void) ( ) EI dx k x dx A dx ( ω ) void i ( ) L φ ( ) 2 2 2 K L d φ i i x d i x = = ρ 0 2 M 0 i dx dx EI dx A dx The Rayleigh wave over a finite length void can be described in terms of two harmonic waves traveling in opposite directions. y( x, t) = Ae j( ωt kx ) + Be j( ωt+ kx ) where A and B are complex amplitudes.
Acoustic-Laser NDE Parametric acoustic array (PAA): shutoff motion detectors 3000 Watt Power Supply positioning laser Safety Lockout Switches ceramic transducer [Courtesy of MIT Lincoln Laboratory] swivel mount
Acoustic-Laser NDE Acoustic radiation pattern of the developed PAA: microphones RF Anechoic Chamber Measurements [Courtesy of MIT Lincoln Laboratory] SPL (db) 120 100 80 60 40 7k Hz difference frequency 26.3 khz primary Acoustic Spectrum mic #3 on boresight 20 0 10 2 10 3 10 4 Frequency (Hz)
Acoustic-Laser NDE PAA radiation patterns at 7 khz and 26.3 khz: Audible Difference Frequency 7 khz Ultrasonic Primary Beam 26.3 khz Range from array (ft) Range from array (ft) SPL (db) SPL (db) Cross range (ft) Cross range (ft) à Acoustic waves by PAA is focused.
Acoustic-Laser NDE Acoustic radiation patterns at different frequencies: 200 Hz 500 Hz 1 khz Range from array (ft) Range from array (ft) 3 khz 7 khz SPL (db) SPL (db) Cross range (ft) Cross range (ft)
Acoustic-Laser NDE Acoustic power from PAA: Low power amplifier close near field near field end near field 120 Primary ultrasonic carrier 26.34 khz 100 SPL db (re 20 µpa) 80 60 Audible difference Frequencies (Hz) 7000 Hz 1000 Hz Meaured data Modeled prediction based on ultrasonic self-demodulation in air 40 500 Hz 20 200 Hz 0 10 20 30 40 50 Range from transducer array (ft)
Specimen description Experimental Result 15.7 cm 3.8cm 15.2 cm (Concrete core) 7.6 cm 30.4 cm 38.1 cm 2.5 cm Small void 3.8 cm 0.5 cm Large void 7.5 cm (2 in.) Intact/solid region Damaged/void region Concrete mix ratio = water : cement : sand : aggregate = 0.45 : 1 : 2.52 : 3.21 Glass FRP (GFRP) mix ratio = epoxy : glass fiber = 0.645 : 0.355 GFRP type = Tyfo SHE-51A by Fyfe Epoxy = Tyfo S epoxy by Fyfe GFRP sheet thickness = 0.25 cm (0.1 in)
Experimental Result Low frequency acoustic response using a loudspeaker source: 300 Velocity (µ/sec) 250 200 150 100 Solid region Large void region 50 0 0 500 1000 1500 2000 Frequency (Hz) (Measurements were made at a distance of 30 m in an open area in Lexington, MA.)
Experimental Result High frequency acoustic response using the PAA: 8000 Velocity (µ/sec) 6000 4000 2000 solid region large void region 4322 Hz 0 2000 3000 4000 5000 6000 7000 Frequency (Hz) à Using a sound speed of 340 m/s shows that the ½ wavelength of the resonance is approximately 2 inches which is the width of the large void.
Experimental Result High frequency acoustic response: 10 4 Velocity (m/sec) 10 3 10 2 large void small void 10 1 solid region 2000 3000 4000 5000 6000 7000 Frequency (Hz)
Experimental Result Approximate 3D solution: λa Uniform circular plate with fixed edge supports in free vibration ω mn 3 2 where D= Eh 12 1 υ = flexural rigidity of the plate, E = Young s modulus, h = thickness of the plate, ν = Poisson s ratio, ρ = density of the material, w= w( r, θ, t) = transverse displacement in cylindrical coordinate as the function of spatial variables and time t. (E = 21.5 psi (148 GPa); ρ = 1.4 lb/in 3 (1.5 kg/m 3 )) is found from the frequency equation; a = the radius of the circular plate, λ = eigenvalue of the frequency equation. where = ( λa) 2 a mn 2 D ρh ( ) di dj Jn a a In a a dr dr n n ( λ ) ( λ ) ( λ ) ( λ ) = 0 2 k n a ( λa λ ) Jn ( λa) = k! Γ ( n+ k+ 1) 2 k = 0 4 2 k n a ( λa λ ) In ( λa) = k! Γ ( n+ k+ 1) 2 k = 0 4 = Bessel function of the first kind = modified Bessel function of the first kind
Experimental Result Approximate 3D solution: Fundamental frequency, ω 1, (Hz) 16000 14000 12000 10000 8000 6000 4000 2000 0.3 0.4 0.5 0.6 0.7 λa λ a = 0.532 à 4512 Hz Velocity (µ/sec) 8000 6000 4000 2000 0 High Frequency Source (PAA) solid region resonance large void region 2000 3000 4000 5000 6000 7000 Frequency (Hz) 4322 Hz à Difference is attributed to i) non-perfectly shape of the delamination, and ii) the variation in boundary condition.
Summary The proposed acoustic-laser technique is capable of remotely exciting a fiberglass-concrete system and collecting the surface dynamic signature from the system. Surface dynamic signature of the intact (solid) region in a multi-layer fiberglass-concrete system is different from the one of the delaminated (void) region. à A database relating surface dynamic signature and delamination/debonding characteristics can be established. High velocity measurements are remotely observed at the debonding location and at the resonant frequency relating to the debonding geometry. Possible use for detecting surface concrete cracking and steel corrosion
Thank you for your attention. Questions?