Nathan Sharp

Thermal and mechanical response of PBX 9501 under contact excitation

The thermal and mechanical responses of a cyclotetramethylene-tetranitramine-based explosive (PBX 9501) and two non-energetic mock materials (900-21 and PBS 9501) under high-frequency mechanical excitation are presented. Direct contact ultrasound transducers were used to excite samples through a frequency range of 50 kHz to 40 MHz. The mechanical response of each sample was approximated from a contact receiving transducer and trends were confirmed via laser Doppler vibrometry. The steady-state thermal response of the samples was measured at discrete excitation frequencies via infrared thermography. A maximum temperature rise of approximately 15 K was observed in PBX 9501, and the mock materials exhibited similar thermal characteristics. Temperature gradients were calculated to estimate the total heat generated within the samples due to the mechanical excitation. The active heating mechanisms were found to be highly dependent on the frequency of excitation. Possible mechanisms of heating at frequencies bel...

A bio-inspired asynchronous skin system for crack detection applications

In many applications of structural health monitoring (SHM) it is imperative or advantageous to have large sensor arrays in order to properly sense the state of health of the structure. Typically these sensor networks are implemented by placing a large number of sensors over a structure and running individual cables from each sensor back to a central measurement station. Data is then collected from each sensor on the network at a constant sampling rate regardless of the current timescales at which events are acting on the structure. These conventional SHM sensor networks have a number of shortfalls. They tend to have a large number of cables that can represent a single point of failure for each sensor as well as add significant weight and installation costs. The constant sampling rate associated with each sensor very quickly leads to large amounts of data that must be analyzed, stored, and possibly transmitted to a remote user. This leads to increased demands on power consumption, bandwidth, and size. It also taxes our current techniques for managing large amounts of data. For the last decade the goal of the SHM community has been to endow structures with the functionality of a biological nervous system. Despite this goal the community has predominantly ignored the biological nervous system as inspiration for building structural nervous systems, choosing instead to focus on experimental mechanics and simulation techniques. In this work we explore the use of a novel, bio-inspired, SHM skin. This skin makes use of distributed computing and asynchronous communication techniques to alleviate the scale of the data management challenge as well as reduce power. The system also periodically sends a ‘heat beat’ signal to provide state-of-health updates. This conductive skin was implemented using conductive ink resistors as well as with graphene-oxide capacitors.

Lithium-ion battery electrode inspection using flash thermography

Pulse thermography was used to experimentally evaluate lithium-ion battery electrode quality. Lab manufactured electrodes with gross defects, thickness variation, and composition variation all were detectable with this method. Thickness variation was shown to have a one to one ratio percent change in thickness to percent change in thermal response. A thickness difference of 4 μm (4 % of total) was detectable with the method. Lab electrodes were compared with commercial electrodes with comparable results. Both types of electrodes showed a significant thickness oscillation that has not previously been reported regarding lithium-ion battery electrodes.

Pulse thermography for quantitative nondestructive evaluation of sound, de-mineralized and re-mineralized enamel

Current limitations for diagnosing mineralization state of tooth enamel can lead to improper surgical treatments. A method is investigated by which the tooth health state is characterized according to its thermal response, which is hypothesized to be sensitive to increased porosity in enamel that is caused by demineralization. Several specimens consisting of previously extracted human teeth a re prepared by exposure to Streptococcus mutans A32-2 in trypticase-soy-borth supplemented with 5% sucrose at 37°C for 3 or 6 days to de-mineralize two 1×1mm2-windows on each tooth. One of these windows is then re-mineralized with 250 or 1,100ppm-F as NaF for 10 days by pH-cyclic-model. Pulse thermography is used to measure the thermal response of these sections as well as the sound (healthy) portions of the specimen. A spatial profile of the thermal parameters of the specimens is then extracted from the thermography data and are used to compare the sound, de-mineralized, and re-mineralized areas. Results show that the thermal parameters are sensitive to the mineralization state of the tooth and that this method has the potential to accurately and quickly characterize the mineralization state of teeth, thereby allowing future dentists to make informed decisions regarding the best treatment for teeth that have experienced demineralization.

Crack detection sensor layout and bus configuration analysis

In crack detection applications large sensor arrays are needed to be able to detect and locate cracks in structures. Emerging graphene-oxide paper sensing skins are a promising technology that will help enable structural sensing skins, but in order to make use of them we must consider how the sensors will be laid out and wired on the skin. This paper analyzes different sensor shapes and layouts to determine the layout which provides the preferred performance. A ‘snaked hexagon’ layout is proposed as the preferred sensor layout when both crack detection and crack location parameters are considered. In previous work we have developed a crack detection circuit which reduces the number of channels of the system by placing several sensors onto a common bus line. This helps reduce data and power consumption requirements but reduces the robustness of the system by creating the possibility of losing sensing in several sensors in the event that a single wire breaks. In this paper, sensor bus configurations are analyzed to increase the robustness of the bused sensor system. Results show that spacing out sensors in the same bus as much as possible increases the robustness of the system and that at least 3 buses are needed to prevent large segments of a structure from losing sensing in the event of a bus failure. This work is a preliminary effort toward enabling a new class of ‘networked materials’ that will be vitally important for next generation structural applications. ‘Networked materials’ have material properties related to information theoretic concepts. An example material property is ‘bandwidth’ per unit of material that might indicate the amount of information the material can provide about its state-of-health.

Lithium-Ion Battery Cell Health Monitoring Using Vibration Diagnostic Test.

With their superior advantages of high capacity and low percentage of self-discharge, lithium-ion batteries have become the most popular choice for power storage in electric vehicles. Due to the increased potential for long life of lithium-ion batteries in vehicle applications, manufacturers are pursuing methodologies to increase the reliability of their batteries. This research project is focused on utilizing non-destructive vibration diagnostic testing methods to monitor changes in the physical properties of the lithium-ion battery electrodes, which dictate the states of charge (SOC) and states of health (SOH) of the battery cell. When the battery cell is cycled, matter is transported from one electrode to another which causes mechanical properties such as thickness, mass, stiffness of the electrodes inside a battery cell to change at different states of charge; therefore, the detection of these changes will serve to determine the state of charge of the battery cell. As mass and stiffness of the electrodes change during charge and discharge, they will respond to the excitation input differently. An automated vibration diagnostic test is developed to characterize the state of charge of a lithium-ion battery cell by measuring the amplitude and phase of the kinematic response as a function of excitation frequency at different states of charge of the battery cell and at different times in the life of the cell. Also, the mechanical properties of the electrodes at different states of charge are obtained by direct measurements to develop a first-principles frequency response model for the battery cell. The correlation between the vibration test results and the model will be used to determine the state of charge of the cell.Copyright

An asynchronous sensor skin for structural health monitoring applications

In crack detection applications large sensor arrays are needed to be able to detect and locate cracks in structures. This paper analyzes different sensor shapes and layouts to determine the layout which provides the optimal performance. A “snaked hexagon” layout is proposed as the optimal sensor layout when both crack detection and crack location parameters are considered. In previous work we have developed a crack detection circuit which reduces the number of channels of the system by placing several sensors onto a common bus line. This helps reduce data and power consumption requirements but reduces the robustness of the system by creating the possibility of losing sensing in several sensors by a single broken wire. In this paper, sensor bus configurations are analyzed to increase the robustness of the bused sensor system. Results show that spacing sensors in the same bus out as much as possible increases the robustness of the system and that at least 3 buses are needed to prevent large segments of a structure from losing sensing in the event of a bus failure.

Endowing Structures with a Nociceptive Sense Enabled by a Graphene-Oxide Sensing Skin

In this work we use bio-inspired strategies to move towards a low bandwidth, robust, asynchronous structural health monitoring approach which mimics biological nociceptive sense. A novel sensing skin is developed with graphene-oxide capacitive sensors which act as structural damage receptors.

Inspection for kissing bonds in composite materials using vibration measurements

Improper bonding of composite structures can result in close contact cracks under compressive stresses, called kissing bonds. These bond defects are very difficult to detect using conventional inspection techniques such as tap testing or local ultrasonic scanning and can lead to local propagation of damage if the structure is subjected to crack opening stresses. A method is investigated for identifying kissing bonds in composite material repairs based on vibration measurements. A damage feature of the kissing bond is extracted from the response of the input-output measurement that is a function of the structural path. This path exhibits local decoupling associated with the close contact cracks. Experimental vibration measurements from sandwich composite materials are presented along with the results of the damage detection algorithm for the healthy sections of the material and the kissing bond sections. A vibration based inspection technique could increase the ability to detect kissing bonds in composite material repairs while decreasing inspection time. Benefits of this method of identification over conventional techniques include its robust, objective damage detection methodology and the reduced requirement for specimen preparation and surface texture when compared to ultrasonic scanning.