Joona Eskelinen

Frequency domain low time-bandwidth product chirp synthesis for pulse compression side lobe reduction

Coded signals are commonly used in communication and radar systems. However, in ultrasonics the use of coded signals is still relatively uncommon. Linear frequency modulated chirps, employed to improve signal to noise ratio (SNR), exhibit autocorrelation sidelobes whose amplitude increase with decreasing time bandwidth product. This is important in pulse-echo ultrasonics where the chirps need to be short. The side lobes can be attenuated by windowing the chirp amplitude with a suitable window function, but this broadens the main lobe which lowers the resolution. We describe an alternative method for chirp generation in the frequency domain. The chirp is generated by applying a quadratic phase shift into a signal with predefined spectrum. The autocorrelation side lobes of these chirps are nearly identical with the original unchirped pulse. A sum of four sine modulated Gaussians were used to provide the initial spectrum for evaluating the frequency domain chirping. This method holds potential for use in NDT pulse-echo measurements and medical ultrasonics where short codes to improve SNR are needed.

Beamforming with a volumetric array of massless laser spark sources--Application in reflection tracking.

A volumetric array of laser-induced air breakdown sparks is used to produce a directional and steerable acoustic source. The laser breakdown array element is broadband, point-like, and massless. It produces an impulse-like waveform in midair, thus generating accurate spatio-temporal information for acoustic beamforming. A laser-spark scanning setup and the concept of a massless steerable source are presented and evaluated with a cubic array by using an off-line far field delay-and-sum beamforming method. This virtual acoustic array with minimal source influence can, for instance, produce narrow transmission beams to obtain localized and directional impulse response information by reflection tracking.

Laser-induced acoustic point source for accurate impulse response measurements within the audible bandwidth

Laser induced air breakdown is proposed as a sound source for accurate impulse response measurements. Within the audible bandwidth, the source is repeatable, broadband, and omnidirectional. The applicability of the source was evaluated by measuring the impulse response of a room. The proposed source provides a more accurate temporal and spatial representation of room reflections than conventional loudspeakers due to its omnidirectionality, negligible size and short pulse duration.

Discriminating pores from inclusions in rolled steel by ultrasonic echo analysis

The quality of steel is degraded by inclusions and pores which form during production. When exposed to external stress, large inclusions initiate cracks that weaken the mechanical strength of steel components. In this study, an ultrasonic immersion pulse-echo setup, which is currently used for quality control, was employed for signal-based inclusion and pore discrimination. The results from the used wavelet and short-time Fourier transform methods were verified with optical microscopy and scanning electron microscopy. The ultrasonic method was tested on 22 rolled bearing steel 100Cr6 samples featuring a total of 24 inclusions and 16 pores. Based on the differences in the echo peak frequency, it was possible to discriminate all pores from inclusions with the wavelet transform method. Sulfide and oxide inclusions also featured differences within the 9–14 MHz frequency range. The developed ultrasonic method was found to be capable of discriminating inclusions from pores in rolled bearing steel 100Cr6.

SIMULATIONS TO IMPROVE STRUCTURAL DEFECT DETECTION AND CLASSIFICATION IN SWISS‐CHEESE

Ultrasonic 2D simulations to facilitate defect detection and classification for structural quality control of Swiss‐cheese are presented. Three economically relevant structure types were modeled with different geometry parameters and the back‐scattered ultrasonic field from the structure boundary was simulated to obtain reference data for waveform analysis. Simulated waveform characteristics were evaluated and compared to the experimental ones. Two parameters were introduced to classify different defects by exploiting the frequency spectrum of the signals. Signal waveform and correlation analysis, based on the simulation results, improved defect detection probability.

High frequency broadband arbitrary coded excitation in ultrasound microscopy

We built a 200+ MHz ultrasound microscope that can employ broad band arbitrary coded signals. Our solution is based on quadrature modulation and T/R-switching. We show that the instrument works with linear and non-linear chirp signals (100 - 300 MHz). We imaged a 5 μm tall MEMS step structure with our device and with a commercial pulser, the results were comparable. Cross-talk from the local oscillator in the quadrature modulator affects the pulse compression. This limits the use of codes at the oscillator frequency.

Non-destructive evaluation of the 18 th century ship wreck Vrouw Maria

We analyze the current condition of pine and oak samples recovered from a sunken, 240 years old Dutch merchant ship wreck, the Vrouw Maria, with ultrasound and other non-destructive methods. To evaluate its physical condition - mechanical properties- ultrasonic through-transmission sound velocity measurements at 0,3-4 MHz were performed along the radial and longitudinal wood fiber directions. Presence of large discontinuities (e.g. holes created by ship worms) within the samples were probed using 2 MHz ultrasonic pulse-echo measurements. DNA-based methods (community DNA extraction, amplification (PCR) of SSU rDNA, cloning, restriction fraction length polymorphism (RFLP), sequencing) probed presence of microbes, particularly fungi, in the samples. X-ray diffraction and X-ray fluorescence measurements were performed with a synchrotron to determine the extent of degradation at the nanometer-level in the wood cell walls and changes in the elemental composition of the samples. The ultrasonic measurements detected a layered structure in the pine sample featuring a thin (1-4 mm) mechanically degraded layer on top of almost intact wood (60 % decay in the stiffness modulus compared to the intact part). For the oak sample, a 49% stiffness modulus reduction was detected compared to a freshly felled reference sample. DNA-analysis detected the presence of soft-rot fungi in the degraded layer and in the underlying compact wood (to 2 cm depth). Fungal DNA was more abundant below the heavily degraded layer indicating that fungal degradation had moved deeper towards the non-degraded parts. X-ray diffraction and fluorescence indicated presence of Fe throughout the sample and heavy degradation of the crystalline cellulose in the degraded layer. The results offer insight into the condition of the entire ship wreck, and therefore support decision making regarding possible lifting of the ship.

Ultrasonic inclusion and pore classification in rolled and unrolled steel samples

The feasibility of an ultrasonic method to discriminate inclusions and pores (foreign bodies, FBs) in rolled and unrolled steel samples was studied. A 10 MHz point-focused transducer was used to scan immersed steel samples in pulse-echo mode to detect FBs. The differences in RF-echo amplitude and frequency content allowed discriminating inclusions from pores in the steel samples. Optical microscopy and scanning electron microscopy served as reference methods on cut samples to verify the classification after the ultrasonic measurement. The FB classification method was tested on 24 inclusions and 16 pores from rolled samples. All FBs were correctly classified indicating that this method is feasible for FB classification in rolled steel samples. In the unrolled samples, differences in calcium-aluminate (CaAl) inclusion and pore spectra at 16 MHz should allow CaAl discrimination from pores.

P1E-3 Ultrasonic QNDE Instrument for Quantitative Inclusion and Pore Characterization of Steel Billets

Steel fabrication suffers from inclusions and pores that are formed into steel during production. Ultrasonic characterization of unrolled steel bars is challenging due to low signal-to-noise ratio (SNR). Signal processing provides methods to characterize granular steel samples. We are developing a 9.5 MHz instrument for quantitative inclusion (manganese-sulphur) and pore characterization in immersed steel billets. The instrument is capable of quantitative foreign body classification. MnS inclusions and soft inclusions (pores) can be segregated. After the samples were characterized ultrasonically, inclusions were cut out and a scanning electron microscope (SEM) was used to validate the results. The instrumentation will allow nondestructive waste product content estimation for each waste material separately. It improves the steel fabrication quality control by giving detailed information of the inner structure and therefore allowing more precise fabrication process control.

Estimating the size of a scatterer by transmitted signal delay

Using delay in the time-of-flight (TOF) of an ultrasonic signal to estimate the lateral size of a scatterer was evaluated. A rolled steel sample featuring a 2 mm diameter side-drilled hole (SDH) was immersed in water. The maximum delay produced by the SDH was estimated from through transmission signals. To estimate SDH size the measured delay was compared to theoretical delays with straight path assumption. The size estimate agreed with the real size. This result can find relevance by enabling a simple method for scatterer size estimation.