Mork Family Department of
Chemical Engineering and Materials Science
Nondestructive Ultrasonic Pulse-Echo Method of Testing Composite Materials
Wan Jae Lee, Wenji Victor Chang
Department of Chemical Engineering, University of Southern California, University Park, Los Angeles, CA 90089-1211, USA
Ultrasonic signal of the defected specimen:
• To study the high frequency mechanical properties of plastics, rubbers and
composite materials. Pulse echo signature of defected sample
• To use ultrasonic techniques, particularly pulse echo method, to investigate t11
the effect of material structure and processing conditions on the ultrasonic
properties of fiber reinforced composite materials, phenolic/fiberglass
laminates and unsaturated polyester/fiberglass laminates, consisted of regular t12 t13
and defected samples. 1
Introduction 0.00006 0.00007 0.00008 0.00009 0.0001 0.00011 0.00012
• Ultrasonics as nondestructive and material characterization tools have been
widely applied to semiconductors, metals, superconductors, insulators, -3
magnetic crystals, glasses, quantum liquids polymers and medical imaging.
Figure 1. A typical pulse-echo signal shape of a defected sample: polyester fiberglass.
• Unlike other analytical tools, ultrasonic techniques have the following
advantages: • Ultrasonic signal of the regular specimen:
1. It is fast, precise and non-destructive;
2. Measurement could be conducted in situ; Pulse echo signature of regular sample
3. Composite materials consisting of multi-layers can be easily evaluated; 3
4. Measurement can be made on small and complex-shaped materials and
• Applications of ultrasonic techniques to polymer materials have increased 1
recently especially because they offer speedy, reliable and economic ways to
vo lt (V )
characterize the viscoelastic properties of polymers. 0
0.00006 0.00007 0.00008 0.00009 0.0001 0.00011 0.00012
Schematic Diagram -2
Oscilloscope Figure 2. A typical pulse-echo signal shape of a normally prepared sample: polyester fiberglass.
• The wave shapes are noticeably different. All the pulse echo experiments of
composite samples produced either one of Figure 1 or Figure 2.
Pulser/ 1. Pulse echo signal of defected composite specimen shows many reflected
Water receiver echoes, signifying these deliberate defects:
bath Specimen • implanted section of tape between two plies of fabric,
• an area of a low-resin content defect,
All of these would have caused additional and more frequent echoes.
Experimental Diagram of pulse echo method
2. Pulse echo signals of regular composite specimen do not show many
Transducer echoes, representing just the two regular plies of the sample.
• Density test result produced two prevalent values:
1. Regularly processed samples yield lower density value, 1.788 ±
t12 t13 2. Defected samples yield higher density value, 1.990 ± 0.050g/cm3.
ti Water t0 Water
Zoomed up version of t11 and t12
Reflector block Reflector block t12 3
v o lt (V )
0.000064 0.000065 0.000066 0.000067 0.000068 0.0000636 0.0000646 0.0000656 0.0000666 0.0000676
The longitudinal pulse echo method -1 -1
Longitudinal wave velocity: -3 -3
time (s) time (s)
( t 0 − t 11 ) − ( t 13 − t 12 )
V = V Figure 3. Typical t11 and t12 of defected sample. Figure 4. Typical t11 and t12 of regular sample.
t 12 − t 11
t11 and t12 t0
Sample thickness: 3 3
echoe from the tape third positive peak
( t 0 − t 11 ) − ( t 13 − t 12 ) 2 2
d = V w 1 1
volt (V )
v o lt (V )
0.000064 0.0000645 0.000065 0.0000655 0.000066 0.0000665 0.000067 0.0000675 0.000068 0.000078 0.000079 0.00008 0.000081 0.000082
Vs is ultrasonic wave speed of sample -1 -1
Vw is ultrasonic wave speed in water
t is the echo time for ultrasonic wave reflected from time (s)
corresponding surface Figure 5. t11 and t12 of defected sample Figure 6. A typical pulse-echo shape of t0.
the one with tape implanted.
• Theoretically, t11 should have general shape of t0.
Materials • In Figure 4(t11 and t12 of the regular sample), t11 generally looks like Figure 6(t0),
although t12 is not defined clearly
The composite materials came in 40cm x 40cm slabs, one regular phenolic • In Figure 3(t11 and t12 of the defected sample), t11 looks like Figure 6 except its
fiberglass slab (brown), one defected phenolic fiberglass slab (brown), one third positive peak is too small. However, t12 is well defined.
regular polyester fiberglass slab (white) and one defected polyester fiberglass • Figure 5 shows one of the particular shapes of t11 and t12 of the defected
slab (white). sample. After the third positive peak, there are some more tall signals indicating
1. Phenolic/fiberglass laminates the echoe from the implanted tape.
• Regular product (cargo liner 1367A-030).
• Same material but with the following flaws:
A section of tape implanted between two plies of prepreg; Conclusion
An area where excessive pressure was applied during curing to
create a low-resin content defect;
Ultrasonic nondestructive testing was used to explore how material structure
Micro voids due to trapped volatile as result of fast heating cycles.
and processing conditions would affect the ultrasonic properties of composite
2. Unsaturated polyester/fiberglass laminates
materials by specifically studying two kinds of fiberglasses (phenolic and
• Regular product (cargo liner 1076-030).
polyester); one regularly prepared and the other prepared with deliberate flaws
• Same material containing the flaws described for phenolic fiberglass.
within each kind.
It is evident that ultrasonic signature alone does tell much about the
Procedure composite materials seeing that pulse echo signals for the regular and
defected specimens showed markedly different signal shapes;
• Pulse echo signal for the composite materials with deliberate flaws
• A 3.5 MHz longitudinal transducer was adjusted to obtain the maximum displayed many echoes coming from the defects (including voids, tapes,
amplitude position relative to the reference position. etc).
• A circulating water bath was used to control the temperature of the distilled • Pulse echo signal for the regularly fabricated composite material did not
water. show many reflected echoes coming from just the two plies in the material.
• The reference signals were collected. • Looking closely into the signals, t11 and t12 imply whether the specimen is
• Then a glass holder was put on top of the reference surface; water wave speed regular or defected and even indicate where the flaw is (Figure 5).
could be calculated at this point using this equation: Moreover, density test supports the distinction between the two materials with
2d different processing conditions and slightly different materials by attesting
t1 − t 2 with two different density values. This result is in accordance with the wave
shapes since defected samples showing many echoes are proven to be more
• All the composite slabs were cut into 3cm x 8cm sized specimens for the dense, and regular samples showing just the echoes from the plies are proven
ultrasonic pulse echo experiment. to be less dense.
• After they have been cut, all the specimens were mixed up. Hence, regular
specimens could not be told apart from the defected ones, within the same
material, in the naked eye.
• Each specimen, in no particular order, placed on top of the holder and signals
• Ultrasonic data were used to attain ultrasonic properties: sample wavespeed, • The financial support from The Los Angeles Rubber Inc. (TLARGI) is gratefully
sample thickness, attenuation and tan δ. appreciated.
• Afterwards, density of each specimens was obtained using a densimeter. • Composite samples were provided by M.C. Gill Corporation.
Centennial Celebration January 26, 2006