#

ON-LINE MULTIFUNCTIONAL SENSORS FOR LUBRICANT AND OIL ANALYSIS

ON-LINE MULTIFUNCTIONAL SENSORS FOR LUBRICANT AND OIL ANALYSIS

Machine condition monitoring through oil analysis is a proven and cost effective maintenance management technique. Oil analysis is a diagnostic maintenance tool used to detect and quantify wear metals and contaminants in the lubricant of oil wetted systems. It is based on the fact that a representative sample of lubricant, when periodically analyzed, provides important information on the condition of the wearing surfaces. It is a test that helps determine whether a mechanical system is in a normal wear mode, or whether a potentially serious problem exists.

Currently used wear particle in oil laboratory analysis methods, such as Atomic Fourier Transform Infrared (FTIR), Atomic Absorption Spectroscopy and Inductively Coupled Plasma (ICP), Spectrochemical Analysis, Particle Count Analysis, and Ferrography employ bulky and expensive equipment, and of major importance, are limited to statistical sampling only. Portable and in-line devices developed and patented to date, allow only for measurements of very few (maximum two) specific wear particle parameters. In addition, the performance of these devices is greatly affected by harsh and hostile environment conditions in the close proximity of engines.

Optical sensors are mature instruments that are successfully employed in a variety of scientific and industrial process control and monitoring applications. Optical techniques are ideal since they are usually non-invasive, extremely sensitive, and accurate. They can be successfully applied for measurements of the physical properties of fluids. Recent achievements in the group III-Nitrides materials growth techniques permit now for the development of high-temperature Light Emitting Diodes (LEDs) with peak wavelengths ranging from Ultraviolet (UV) to Infrared (IR).

LED chips used as both wavelength-selective optical emitters/detectors, and temperature sensors can be integrated into inexpensive and rugged multifunctional sensors ideally suited for in-line fluid characterization. In these sensors, multiple (more than 8) band optical emissions in a total range from UV to IR from the LEDs interact with the analyte under test in the form of combined absorption, reflection, scattering, and fluorescence.  In addition each of the LEDs can be used as a temperature sensor by measuring the forward voltage at a fixed current. Based on specific properties of the analyte, individual signal patterns resulting from these phenomena attributed to both intensity in different spectral bands and fluorescence lifetime, as well as temperature data, are obtained. Data from test simulators, calibration samples, and standards, as well as existing available data, are used for training of the Artificial Neural Network (ANN) controlled by the same electronic circuitry. The classification (identification and quantification) of analytes is performed by the ANN based on comparing actual data with that used for ANN training.

The objective of this project is development of a miniature multifunctional intelligent sensor for complete in-line real-time engine (machine) oil analysis. The in situ measurement of oil (or other lubricant fluid) properties at elevated temperatures and pressures will yield more precise and cost effective equipment failure prevention methodologies and techniques. The project is aimed to develop a miniature multi-wavelength opto-thermo-electronic sensor laboratory prototype that employs high-temperature III nitride LED chips. The prototype will be controlled by miniature Field-Programmable Gate Array (FPGA) based circuitry with Artificial Neural Network (ANN) classification analysis. The laboratory prototype should be trained, calibrated, and tested by immersion into simulated contaminated oil samples prepared according to existing standards and data available for various types of lubricants and their contaminants.   The future goals will be focused on:  a) extension of the sensor capabilities by adding a planar MEMS-based inductance circuit for quantification of the ferromagnetic particles, b) development of a pre-commercial on-line sensor prototype, c) enabling wireless networking capability, d) testing in real working engine conditions.

Originally used by the military and the railroad as a preventative maintenance program, oil analysis has expanded today to industries from automobile shops to nuclear power plants. Each industry has its own specific needs and criteria as well as different testing methods, but lately these methods are moving toward consolidation. When selecting a sampling point, it is important to sample in several locations of the lubrication system to determine where the most representative sample can be taken. It is very difficult to accomplish this task by using current statistical sampling methods.

On-line sensors installed at critical sampling points would thus provide for a revolutionary improvement on the methodology. Oil analysis is the evaluation of the oil itself and any contamination that is present. Oil substance analysis usually are based on physical testing of the viscosity, water content (more than 1%), and acid/base number[1].  Spectrochemichal analysis performed using Atomic Absorption Spectroscopy (AAS)[2] and Inductively Coupled Plasma (ICP) spectroscopy[3] are used to identify and quantify the metal content and additive package. Normally 19 elements of the chemical spectrum are measured in parts per million. These numbers represent elements less than 5 µm in size. The spectrometers design limits detection level to 5 µm and below. To evaluate particulates larger than 5 microns, particle count methods are used. The particle count measures all particulate in the oil larger than 5 µm. Particulate include: dirt, carbon, metals, fiber, bug parts, etc. The particle count can be done using various filtering[4] techniques or/and using a laser[5],[6] or optical methods[7]. The laser method reports the quantity, size (5-100 µm) and distribution of particulate but not what they are. The optical method gives a quantity, size, distribution and identification.

The use of both methods provide for the most representative analysis methodology available. Ferrography, a technique that separates metal particles from an oil sample for further diagnosis with a microscope has proven to provide supplemental, and often critical, information regarding the actual sizes of particles and the wear mode that generated them. Rotrode Filter Spectroscopy (RFS) is a new technique which has been shown in actual field applications to provide important additional information about large wear particles; information that may be missed with conventional techniques.

Identification of the source of the lubricant contamination and its quantification is the most important task in the prevention of wear related equipment failure. According to existing data, this task can be successfully accomplished by using combinations of various optical methods that include absorption/reflection, fluorescence, and scattering. However, such instruments are expensive, bulky, use statistical sampling, and are unable to perform the necessary measurement using a single device. The objective of this project is development of a miniature multifunctional intelligent sensor for complete in-line real-time engine (machine) oil analysis. Our approach is based on utilizing unique features and benefits provided by the following developments in the various fields of science and technology:

  1. Employment of LEDs for determination of the particle size by scattering measurements.

Portable systems for particle analysis based on solid state components have been developed lately. For example, miniature three-wavelength backscattering meters based on LEDs have been used to measure the size of particulates in turbid media[8].

  1. Employment LEDs for identification of metal particles by absorption/reflection/ fluorescence measurements.

Depending on the type of metal, specific optical signatures can be obtained using optical methods[9],[10]. Portable fluorometers and optical absorption/reflection devices based on LEDs and Si-based photodetectors are also widely used for particulate identification[11]. Two-wavelength LED based portable device for oil quality measurements was patented in January 2008[12].

  1. Effect of metal particles on the oil properties.

Properties of mineral and synthetic oils are measured by using intrinsic fluorescence from Polycyclic Aromatic Hydrocarbons (PAH) present in them[13],[14]. This fluorescence is highly sensitive to the presence of various oil contaminants that can change such critical parameters as optical properties, amount of oxygen, PH, and temperature. Optical properties determining unique fluorescence signatures of PAH contained in petroleum products are presented in the Table 1.

  1. Employment of LEDs as photodetectors.

III nitride-based LEDs have been employed as high speed wavelength-selective photodiodes[15],[16] using the p-n junction existing in the LED in a photovoltaic mode.

  1. Employment LEDs as temperature sensors.

The effect of forward voltage variation with temperature in an LED is utilized to provide a temperature sensor[17]. Integration of a temperature sensor within the same system would allow for calibration of the optoelectronic measurements at variable ambient temperatures. Temperature data will be fed into the ANN as an independent parameter, so training of the ANN will be performed by testing simulator samples or standards at various temperatures.

  1. Introduction of portable time-resolved fluorescence analysis as a complimentary method to existing steady state measurement techniques.

Current portable fluorometers are capable of measuring only few compounds (in most cases one) by utilizing steady state (intensity) measurements. Introduction of time-resolved (lifetime) fluorescence measurements in a portable system is a highly challenging task currently being under development at Integrated Micro Sensors Inc[18].

  1. Integration of a planar MEMS-based inductance sensor for characterization of ferromagnetic particles.

MEMS-based planar inductances have been successfully used in the field of RF microelectronics[19]. Integration of such inductance on the same carrying plate used for LED chip assembly would permit more precise characterization of metal particles[20]. Enabling of this feature can be considered during the next stage of the project.

  1. Employment of “smart” configurable wireless remote sensor networks.

The miniature, fast optoelectronic sensors to be developed by IMS in this project should be perfectly compatible with contextual networking platforms for modern sensing technologies. The future developments can consider implementation of short-range wireless personal area networking (WPAN), such as 802.15.4/Zigbee, which when integrated with long-range wireless telecommunication platforms (WiFi, WiMax) will provide breakthrough solutions for indoor and outdoor remote sensing.

Table 1. Fluorescence Data of Aromatic Hydrocarbons in Solution[21]
Hydrocarbon Lifetime

(ns)

Detection

(ppb)

Emission (nm) Hydrocarbon Lifetime

(ns)

Detection

(ppb)

Emission (nm)
Benzene 250-300 Chrysene 20.00 50 360-400
Naphtahlene 27.47 250 300-365 Pyrene 35.85 250 370-400
Anthracene 4.22 500 372-460 Perylene 440,470
Naphthacene 460-580 Fluorene 5.89 50 302-370
Pentacene Red Cholanthrene 400-500
Rubrene 545-623 Decacyclene 477-600
Phenanthrene 23.50150 348-407 Fluorocyclene 410-540

 

The preliminary results on a miniature multifunctional sensor system have been achieved during development of an integrated device for ultrafine particle characterization[22]  based on high-temperature GaN/InGaN based multi color LED chips assembled on a 9 mm dia Sapphire substrate also serving as a protective optical window (Figure 1). The LED chips in the circular array served as both wavelength-selective light emitters and detectors in the total spectral range from 360 to 750 nm. Detection limits in the range of few ppb have been achieved for most of the test analytes,[23],[24],[25]. The analytes were sequentially irradiated by each LED and photoresponse signals were collected by the rest of the LEDs working on a photodiode mode.  The LED array was controlled by a FDMA-based circuitry with an Artificial Neural Network (ANN)-based signal acquisition and analysis. Variable signal patterns were generated by combined effects of fluorescence, absorption, and scattering resulting from interaction of the  multi-band optical emission with the analytes.  ANN was employed for the categorization of different analytes of various concentrations using a Stuttgart Neural Network Simulator (SNNS) tool. For 8 different analytes at 4 or 5 different concentrations, totaling 35 different samples, after 2000 cycles of training the network the results were:  96% accuracy for the testing set and 100% accuracy for the training set (Figure 2). The current efforts on this project are directed towards development of an intelligent portable multifunctional bio-chemical sensor system with time-resolved capability in a ps time resolution range. The PI, Dr. D. Starikov, was awarded two patents[26],[27] on the main concepts of the above sensors. As the III nitrides technology continues to improve, multicolor LED structures, monolithically integrated on a single substrate using methods of Molecular Beam Epitaxy (MBE) or MOCVD, are envisioned in the sensor design as the ultimate goal for this research.

[1] Army Oil Analysis Program for Vehicle Testing. Accession Number : ADA312083. Final report on test operations procedure

Corporate Author : ABERDEEN TEST CENTER ABERDEEN. http://handle.dtic.mil/100.2/ (1996).

[2] Michael J. Quinn. New device for analysis of metal particles in oil. Wear. Volume 120, Issue 3, Pages 369-381 (1987).

[3] Analytical Tools to Detect and Quantify Large Wear Particles in Used Lubricating Oil – by Spectro Inc. http://www.azom.com/details.asp?ArticleID=3394

[4] Daniel Anderson. Spectroscopy for Large Particle Measurement. Spectro, Inc. http://www.noria.com/learning_center/category_article.asp?articleid=104&relatedbookgro.

[5] Mosleh Mohsen ; Blau Peter J. ; Dumitrescu Delia. Characteristics and morphology of wear particles from laboratory testing of disk brake materials. Wear, vol. 256, no. 11-12, pp. 1128-1134 (2004).

[6] C. Babu Rao and Baldev Raj. Study of engineering surfaces using laser-scattering techniques. S¯adhan¯a Vol. 28, Parts 3 & 4, pp. 739–761 (2003).

[7] Jarmo Vanhanen, Marcus Rinkiö, Jukka Aumanen, Jouko Korppi-Tommola, Erkki Kolehmainen, Tuula Kerkkänen, and Päivi Törmä. Characterization of Used Mineral Oil Condition by Spectroscopic Techniques. Applied Optics, Vol. 43, Issue 24, pp. 4718-4722 (2004).

[8] Wetlabs, Inc.  www.wetlabs.com.

[9] A. Dotsenko, S. Kuchinsky and M. Prassas. Mathematical modeling of spectral selective absorption and reflection of light by metal-dielectric composites. Journal of Non-Crystalline Solids. Volume 218, Pages 317-322 (1997).

[10] Coates, John; Rosenbaum, Neil; Abuneaj, Yosef. US Patent 6707043 – On-site analyzer (2004).

[11] Turner Designs. http://www.turnerdesigns.com/t2/instruments/instruments.html

[12] James Z. T. Liu . United States Patent 7,321,117, January 22, 2008.

[13] C.V. Ossia, H. Kong, L.V. Markova and N.K. Myshkin. On the use of intrinsic fluorescence emission ratio in

the characterization of hydraulic oil degradation. Tribology International. Volume 41, Issue 2, February 2008, Pages 103-110 (2007).

[14] T.A. Kubic, and F.X. Sheehan. Individualization of Automobile Engine Oils II. Application of Variable Separation

Synchronous Excitation Fluorescence to the Analysis of Used Automobile Engine Oils. SPIN Abstracts, Volume 28, Issue 2 (1983).

[15] Eiichi Miyazakia, Shin Itami, Tsutomu Araki. Using a light-emitting diode as a high-speed, wavelength selective photodetector

Eiichi Miyazakia, Shin Itami, Tsutomu Araki. Rev. Sci. Ins. V 69(11), (1998).

[16] Purnendu K. Dasgupta , , In-Yong Eom , Kavin J. Morris and Jianzhong Li . Light emitting diode-based detectors. Absorbance, fluorescence and spectroelectrochemical measurements in a

planar flow-through cell.

Analytica Chimica Acta Volume 500, Issues 1-2 , Pages 337-364 (2003)

[17] Patel Mohnish Kumar. An LED based temperature sensor. Publication number: GB2369437 (2002).

[18] Clement Joseph, Mounir Boukadoum, Joe Charlson, David Starikov and Abdelhak Bensaoula. High-speed front end for LED-Photodiode based fluorescence lifetime measurement system. Proceedings of IEEE Symposium  Circuits and Systems Society (ISCAS 2007), New Orleans, 27 -30 May, 2007. http://ieeexplore.ieee.org/document/4253454?reload=true

[19] Dell, J.M.; Winchester, K.; Musca, C.A.; Antoszewski, J.; Faraone, L. Variable MEMS-based inductors fabricated from PECVD silicon nitride. Optoelectronic and Microelectronic Materials and Devices, Volume , Issue , 11-13 Dec. 2002 Page(s): 567 – 570 (2002).

[20] Remmlinger, Hubert; Ingenbleek, Robert; Schuwerk, Gabriele; Schmitz, Rolf..  Method and device for machine diagnosis, especially for transmission diagnosis. US Patent 6895808 (2005).

[21] Michael A. Dvorak,*,† Gregory A. Oswald, Mark H. Van Benthem,‡ and Gregory D. Gillispie. On-the-Fly Fluorescence. Lifetime Determination with Total Emission Detection in HPLC. Anal. Chem., 69, 3458-3464 (1997).

[22] NIH SBIR Project „Instrumentation for Ultrafine  Particles Characterization“ grant # R43ES12513 (2003). https://www.google.com/url?sa=t&rct=j&q=&esrc=s&source=web&cd=1&ved=0ahUKEwjFnMT2-PTTAhXMJsAKHT16BvEQFggiMAA&url=https%3A%2F%2Fwww.bisti.nih.gov%2FShared%2520Documents%2FFundedProjects%2FYear2003%2FSBIRFY03.pdf&usg=AFQjCNFELBXQTw7R7uHRdBMVeqpFfnbkuQ&cad=rja

[23] Boukadoum, M., Bensaoula, A., and Starikov, D. “Neural-network-based detection of live bacteria marked with a fluorescent protein tracer”. Proc. Artificial Intelligence and Applications (AIA’2003), Benalmádena (Spain), September 2003.

[24] Boukadoum, M., Bensaoula, A., and Starikov, D. “A Portable Multi-Band Optoelectronic system for Identifying and Measuring the Concentration of Fluorophore Substances ” Proc. IEEE North East Workshop on Circuits and Systems (NEWCAS 2004), Montreal (Canada), June 2004. http://ieeexplore.ieee.org/document/1359035/

[25] Boukadoum, M., Tabari, K., Bensaoula, A., and  Starikov, D. “Comparison of the Noise Immunity of a LED-Based Multi Band Optoelectronic Sensor when Using FDMA and CDMA to Code the Excitation Source”. IEEE Asia-Pacific Conference on Circuits and Systems (APCCAS 2004), Tainan (Taiwan), December 2004. http://s3.amazonaws.com/academia.edu.documents/44447245/FPGA-based_multispectral_fluorometer_usi20160405-28158-gev0iv.pdf?AWSAccessKeyId=AKIAIWOWYYGZ2Y53UL3A&Expires=1494958827&Signature=HDr067ALvZ%2Bb8roN4eGIpySH5bM%3D&response-content-disposition=inline%3B%20filename%3DFPGA-based_multispectral_fluorometer_usi.pdf

[26] D.Starikov, I. Berishev, and A. Bensaoula. One-chip micro-integrated optoelectronic sensor. US Patent 6608360 (2003). http://www.google.ch/patents/US6608360

[27] D.Starikov, I. Berishev, and A. Bensaoula. One-chip micro-integrated optoelectronic sensor. US Patent 6881979 (2005). https://www.google.ch/patents/US6881979

 

Our preliminary results confirmed that GaN/INGaN based LED chips can perform as both emission sources and photodetectors in a temperature range up to 320°C (Figure 3). Beside high-temperature characteristics these devices demonstrate outstanding chemical and radiation resistance properties, making them superior for applications in harsh and super ambient environments. Our data also show that the forward voltage is the parameter most sensitive to the LED temperature with a TC = -2.7mV/°C (Figure 4a). The dependence of the forward voltage on the temperature in the range up to 150°C is linear and can be used to measure the lubricant temperature at fixed LED currents in the range from 20 to 90 mA.

The preliminary results above indicate that employment of arrays based on high temperature III nitride multicolor LED chips integrated on a sapphire substrate, controlled by advanced FPGA circuitry with ANN based analysis, can be employed for the development of miniature on-line sensors for characterization of contaminants in lubrication systems.  
The multi-wavelength optoelectronic sensor prototype can be fabricated using off the shelf LED chips currently available in a wide spectral range. Specifications[1] of the 12 LED chips with peak wavelength ranging from near UV to near IR that will be integrated in an array as shown in Table 2.

Table 2. LED chips that can be potentially used in the array design.

No. Model Peak  wavelength (nm) Color Material Optical power

(at 20mA)

1 UV355CHIP 355 nm n/a GaN/InGaN typ. 40 µW
2 C375MU3 375 nm n/a GaN/InGaN 0.8-1.2 mW
3 C385MU4 385 nm n/a GaN/InGaN 1.2-1.8 mW
4 C395MU6 395 nm n/a GaN/InGaN 2.5-3.5 mW
5 C405MU9 405 nm Violet GaN/InGaN 5.5-6.5 mW
6 C480LB8 460 nm Blue GaN/InGaN 10.8-13.0 mW
7 NSPGF50AS 520 nm Green GaN/InGaN 16.6-20.2 mW
8 C565-30 565 nm yellow-green GaP 0.12 mW
9 C590-30V 590 nm Yellow InGaAlP/GaAs 1.2 mW
10 C630-30W 635 nm Red InGaAlP 5.0 mW
11 C700-40 700 nm n/a GaAlAs 2.0 mW
12 C800-40P 800 nm n/a GaAlAs/GaAlAs 4.0 mW


[1]
http://www.roithner-laser.com/LED_chips.htm

The LED chips spectral range from 355 to 800 nm allows for combined fluorescence, absorption, and scattering measurements in a very large number of compounds. This multifunctional multi-wavelength optoelectronic array will employ only LED chips working as both light sources and phototodetectors.  Employment of LEDs as photodetectors has been disclosed in a publication that describes a Zn-doped InGaN-based blue LED having the emission peak at 450 nm and the extinction peak at 380 nm and a GaAlAs red LED having the emission peak at 660 nm and the extinction peak at 620 nm. The LEDs had a nanosecond response when a reverse bias was applied to the junctions as for p-i-n photodiodes. Most LEDs are based on p-n-junctions, which allows for their employment as wavelength-selective photodiodes without application of a reverse bias. The spectral range of such photosensitive LEDs is determined by the energy gaps of the p-n junction layers and waveguide materials used for the LED fabrication.

The signals from the LEDs will be collected under irradiation with each LED resulting in signal patterns characteristic to each specific compound, even when mixed with other unknowns.  These patterns will be entered into an artificial neural network based program, which will identify the compound and its concentration.  For this purpose a multi-channel control/acquisition circuit with a power supply, signal amplification, and a microcontroller interface will be designed and assembled.

The LED chips will be mounted in a circular geometry using an optical epoxy Epo-Tek 301-2 from Epoxy Technology onto a metallized and pre-patterned sapphire substrate used as a carrying and heat dissipating substrate and a protective optical widow. The metallization will be used to provide electrical connections to the array elements by using microbonding with a 30mm gold wire. Sapphire base plates with a thickness ranging from 0.2-0.5 mm and a diameter <10 mm will be used for LED bonding and packaging. First, a 0.1-0.2 µm thick titanium layer and a 0.2-0.5 thick µm Au layer will be deposited using e-beam evaporation. Such a Au thickness is required to provide quality thermocompressive bonding. The Au layer will be covered with spin-coated photoresist. A photolithography process will be performed to pattern this layer and a solution of 3HCl+1HNO3 will be used to etch it. After removal of the photoresist in acetone Au features will remain on the substrate surface.

LED chips used for the sensor fabrication have already Au-capped contacts suitable for the thermocompression process and most of them are resistant to temperatures over 300 °C, as shown by our preliminary data.

The LED chips can be aligned using an optical microscope, clamped together with their Au layers touching in the desired final configuration, and placed in a vacuum furnace chamber. Such chamber, already available in the company, is used for high-temperature characterization of multi-layer ceramic capacitors being developed on an ongoing DOE SBIR Phase II project. The chamber should be evacuated to a pressure below 10-5 Torr. While maintaining the clamping force and the vacuum, the parts will be heated to a temperature in the approximate range of 100 to 350 °C for about 1 hour. The combination of heating and clamping pressure in vacuum will cause atoms to diffuse and mix between the Au layers of the two parts resulting in a single gold layer that will bond the two parts together. The array is to be sealed with a high-temperature, thermally-conductive, insulating, highly corrosion-resistive silicone compound. This compound will protect the structure from the ambient and will provide for more effective heat dissipation.   In this project we will also consider development of an approach allowing auto alignment of the LED chips with the base plate contacts configuration, as well as simultaneous thermocompression of multiple chips on a single base plate or multiple base plate/LED chip pairs.

One LED chip will be selected on the basis of a high forward voltage temperature dependence at a fixed current. After calibration this LED chip will be used for precise temperature measurements by providing a constant current and measuring the forward voltage. A separate channel in the control and acquisition system as well as a separate input in the ANN will be dedicated to collect and process the temperature data.

For effective time-resolved (TR) acquisition of the fluorescence signals generated in the LED, as a result of the interaction of the excitation light, the analyte should be sampled continuously or with a very high rate. Our ultimate goal is combining the two different types of measurements (SS and TR) measurements in a single portable design. The system’s output will be based on data from both measurements that will be classified by ANN in real-time. In steady-state photocurrent measurements, the common method of capturing and digitizing a DC signal is to use a switched integrator in combination with an analog to digital converter (ADC). The principle is based on collecting the signal to an integration capacitance for an integration period selected by the user (Figure 5) followed by digitizing. Current ADCs are very accurate in digitizing low-level currents. In our photo-detection set up, we envision currents in the range from picoamperes for very low-emission levels to a few hundred nanoamperes. To measure such a wide dynamic range of current-inputs, an ADC with a high dynamic range will be necessary.

The measurements will also require necessary ADC control signals from a portable setup. There are many different ways of controlling signals and acquiring data by using a portable design based on employment of microcontrollers, FPGAs, PLDs and other programmable devices. Utilization of time-resolved measurements will require employment of FPGA based control capable of providing fast response-time and stable operation. A suitable FPGA will be selected depending on the requirements for time resolution in the TR measurements, and to have a sufficient capacity for performing ANN and other related tasks.

Based on our preliminary studies, a TR system with a resolution in the nanosecond range will be required for measuring the lifetimes of the PAH components22. Amplification of the photodetector signals (in pA or nA) plays an important role in these measurements. We have already demonstrated a method of amplifying such signals

using a bootstrapped-cascoded technique[1]. This technique was successfully employed for amplification of current pulses with 5-10ns fall/rise time.

In order to train the optoeltectronic sensor system to detect metal particles in contaminated lubricants, simulator samples of most important machine oils used in the automobile and aerospace industries should be used. The first step in this task is to select the sample lubricants and specific contaminants to be used in these samples. This selection can be based on the data already available in the literature[2],[3],[4],[5]. Then we can disperse particles of various metals and sizes into the oil samples at different concentrations added by using sonisication methods.  We can also disperse small carbon particles in order to simulate soot in the oil samples. Dispersion of small amounts of water should be also considered.

The preliminary selected target metals are: iron, nickel, cobalt, vanadium, and zinc. It has been found that the most useful information is obtained by determination of the iron content in the oil with a desired detection level of about 5 ppm. Potential engine malfunctioning will show a trend-type increase to the 15-20 ppm level[6]. A preliminary list of selected lubrication materials used in different military systems that can be employed in the Phase I project for preparation of calibration contaminated oil samples is given below:

  1. MIL-L-2104 -Lubricating Oil, Internal Combustion Engine, Gr.10w30, 40, 15w40
  2. AMS-G-4343 – Grease Pneumatic Systems
  3. MIL-PRF-5606-Hydraulic Fluid, Petroleum Base, Aircraft, Missile & Ordinance
  4. MIL-PRF-6081-Lubricating Oil, Jet Engine, Gr. 1010
  5. MIL-PRF-7808-Lubricating Oil, Aircraft Turbine Engine, Synthetic Base
  6. MIL-DTL-17111C-Fluid Power Transmission
  7. MIL-PRF-46176-Brake Fluid, Silicone, Automotive, All Weather Operational & Preservative
  8. MIL-L-83767-Lubricating Oil, Vacuum Pump, Mechanical Types I, II, III & IV

 

The LED chip array based on a sapphire carrying plate can be sealed from the back using a high-temperature high-chemical resistance liquid-proof protective silicone compound. This array will be used to perform measurements by submersion into calibration oil samples held in light-tight containers. Advanced housing design would provide for on-line installation of the developed LED chip arrays. Latest achievements in on-line sensor packaging will be modified and implemented to meet the required specifications for on-line optoelectronic sensor usage.

The housing can be provided with a standard high-temperature sealed (liquid and light-tight) connection, which will permit sensor installation in various lubrication systems with minimum modifications necessary, i.e. incorporation in a “tee” configuration. Another requirement is high-temperature wiring that should also be compatible with conditions for low-level signal transmission. Measures also should be taken to provide for low maintenance sensor designs. Additional features can include periodic cleaning of the sapphire window using a planar MEMS-based heater or micro-mechanical brushing mechanism integrated within the external sapphire window surface. The addition of the sensor to the lubrication system should also cause minimal lubricant turbulence.

Next, design for integration of data and analysis, the selection and optimization of ANN algorithms and architecture should be performed. The final algorithm with all the coefficients derived from the trained ANN is implemented in an FPGA to allow for a completely portable device. Reduction of noise background is also a key criterion, which should be investigated. Optimization of photodiode amplifier circuit parameters should be performed for amplification of low-level signals. Another key task is to combine the TR and SS measurement controls in a single FPGA chip, overcoming noise from different sources.  A sample setup for combining the steady-state and time-resolved measurements is shown in Figure 6[7]. An EEPROM is required to start the system in a portable mode. The SS and TR measurements are made separately and stored in the FPGA, where the ANN algorithm is executed.

[1] C. Joseph, M. Boukadoum, J. Charlson, D. Starikov, and A. Bensaoula, “High-speed front end for LED-Photodiode based fluorescence lifetime measurement system”, Proc. IEEE International Symposium on Circuits and Systems, May 2007 (ISCAS 2007), pp. 3578-3581. http://ieeexplore.ieee.org/document/4253454/

[2] G. T. Fei and L. D. Zhang. Preparation and optical absorption peak of small metal particles dispersed in oil. Journal of Alloys and Compounds. Volume 245, Issues 1-2, Pages 116-118 (1996).

[3] S. Bucak, A. Pugh-Jones, C. Lewis and D.C. Steytler.  Metal nanoparticle formation in oil media using di(2-ethylhexyl) phosphoric acid (HDEHP). Journal of Colloid and Interface Science

Volume 320, Issue 1, Pages 163-167 (2008).

[4] K. Sayo, S. Deki1, S. Hayashi. A novel method of preparing nano-sized gold and palladium particles dispersed in composites that uses the thermal relaxation technique. Journal The European Physical Journal D. Volume 9, Numbers 1-4 (1999).

[5]Masaharu Tsuji, Masayuki Hashimoto, and Takeshi Tsuji. Fast Preparation of Nano-sized Nickel Particles under Microwave

Irradiation without Using Catalyst for Nucleation. Chemistry Letters. Vol. 31, No. 12 p.1232 (2002).

[6] Packer, Louis L., Bruton, William A., Woody, Bernard A. Engine Oil Inspection System Using X-Ray Fluorescence. United States Patent 3751661(1973).

[7] Data sheet for DDC101 from Texas Instruments. http://www.alldatasheet.com/datasheet-pdf/pdf/523635/TI/DDC101.html

Figure 6. Schematic for a possible setup of a portable data-acquisition system combining steady-state (SS) and time-resolved (TR) measurements.