·Home ·Table of Contents ·Terrestrial and Marine Transportation

Mobile Inspection System for Rail Integrity Assessment A.CHAHBAZ, M.BRASSARD AND A. PELLETIER Tektrend International Inc. 2113A St. Regis Boulevard, Dollard, Montreal, Quebec, H9B 2M9, Canada Email: research@tektrend.com Contact

ABSTRACT

Current methods of automated rail inspection make use of ultrasonic methods and in some cases eddy current techniques for defect assessment in railroads. However, high inspection reliability and cost effectiveness are essential in the operation of railroad maintenance.

In this work, we demonstrate the development of a rail test vehicle with an EMAT probe carriage designed for real-time testing at high speed. This system inspects the rail via Electro-Magnetic Acoustic Transducers (EMATs) instead of piezoelectric transducers. In this paper, the principles of operation of EMATs and the basis of this choice of transduction methods are described and experimentally investigated.

The development of this mobile inspection system including methodology involved in the system design, software implementation and defect etectability are reported. This paper describes also test results obtained with the designed vehicle where the detection capabilities of the system were evaluated over an evaluation track at Canadian National (CN) Taschereau Yard in Montréal. Results are presented as B-scan images and A-scan waveforms, which are observed in the inspection cabin of the vehicle on an LCD display.

INTRODUCTION

Within the scope of this research work, a mobile rail test vehicle with two EMAT(Electromagnetic Acoustic Transducer) probe carriages were designed for real-time testing of rails. A non-destructive inspection procedure, including hardware and software, were also developed, with the ability to detect, locate and validate rail defects. The developed inspection system uses bulk and surface waves generated by couplant-free EMAT probes.

Rail defects can be classified as transverse head defects, longitudinal head defects, damaged rail, surface defects, and web defects. Transverse defects are those which developed perpendicular to the axis of the rail and longitudinal defects, along the axis. To identify the appropriate EMAT probes suitable for detection of the aforementioned defects, the AREA Rail Inspection Specifications were analyzed and the ultrasonic inspection methods suitable for detection of some specific defects were identified. The target defects and the EMAT transducers used to detect them are shown in Table 1.

DEFECT TYPE	EMAT TEST METHOD	DEFECT TYPE	EMAT TEST METHOD
Transverse Fissure	0° and 90° SH	Head ,Web Separation	0° SH
Compound Fissure	0° and 90° SH	Split Web	0° SH
Detail Fracture- Shelling	0° and 90° SH	Piped Rail	0° SH
Detail Fracture- Head Check	0° and 90° SH	Square, Angular Break	0° and 90° SH
Engine Burn Fracture	0° and 90° SH	Defective Weld	35° and 70° SH
Welded Burn Fracture	0° and 90° SH	Bolt Hole Crack	35° SH
Horizontal Head Split	0° SH	Surface Cracks	Surface waves
Vertical Head Split	0° SH
Table 1: Defect detection methods

TECHNICAL BACKGROUND

Transducer technologies for rail inspection range from piezoelectrics to non-contact electromagnetics, including flux leakage, eddy currents and ultrasonics introduced by electromagnetic induction. The general capabilities and disadvantages of piezoelectric systems are well known to the railway companies using inspection services based on this technology. Thus, choice of a transducer technology involves detectability, complexity of the transducer array required to achieve target coverage, and immunity to interfering and masking phenomena.

EMATs Advantages and disadvantages:
EMATs offer an attractive alternative to piezoelectric transducers for rail inspection [1]. EMATs do not require a liquid couplant or physical contact with inspected rail. EMATs do not lose acoustic coupling (compared to a conventional transducer) when the rail surface is pitted or corroded, or covered with a non-conductor such as dust or grease. The non-contact feature also facilitates high-speed inspection. In addition, EMATS can generate uniquely useful shear horizontal (SH) waves, virtually impossible with piezoelectric transducers. The direction of propagation of their beams can be varied electronically. Graham and Martin [2] have shown that EMATs may improve inspection results in the case of transverse defects and vertical split head defects. This is quite significant, since these are two of the most dangerous flaws, which may develop during the service life of rail.

Despite the apparent advantages of EMATs, they are not free of disadvantages. A requirement of early EMATs for a large electromagnet has been overcome using permanent magnets in very compact probe designs. A general characteristic is that they are inefficient in the generation of ultrasonic energy in comparison with standard probes. The transmitting power of the probes can be increased to satisfactory levels with the availability of high-power pulsers. The efficiency problem can be addressed through the use of phased-array probes which focus the power of the ultrasonic beam in a single desired direction.

An additional problem with SH waves using permanent magnet transducers is that their inherent design leads to the introduction of Barkhausen noise when the probe is moved over the surface of a ferromagnetic material. This problem can be addressed through the use of a biasing (saturating) magnetic field to stabilize the magnetic domains or by digital averaging and filtering.

Excitation mechanism:
EMATs are acoustic probes that need no physical contact with the inspected specimen. They can generate and receive ultrasonic waves in conductive or magnetic materials. EMATs can excite several types of ultrasonic waves based on the coil-magnet configuration (Figure 1). The physical principles of EMAT operation are based on the principle of Lorentz force, which is caused by the interaction between the magnetic flux and the electric current, induced through the eddy current coil. The reverse process occurs in the receiving stage; the resulting eddy current is inductively picked up by the same coil in the pulse-echo mode or the receiver in the pitch-and-catch arrangement. Different coils and magnet configurations can be used to generate different types of elastic waves.

Fig 1: Typical EMAT configuration.

INTEGRATED RAIL INSPECTION SYSTEM

A rail test vehicle with an Electromagnetic Acoustic Transducer (EMAT) probe carriage designed for real-time testing of rails was developed. In addition, an inspection and analysis procedures coupled to the hardware and software was also developed. The developed system comprises the hardware and software modules as shown in Table 2.

System includes
Hardware	Software
Transducers	User Interface
Transducer Carriage and Deployment Mechanism	Transducer Carriage Deployment
Ultrasonic Data Acquisition Instrumentation	Ultrasonic Data Acquisition
System Host Computer	Data Interpretation
Positioning System	Display and Reporting
Communications	Positioning System
Host Vehicle	Communications
Table 2: Inspection system hardware and software modules

Figure 2 shows the mobile inspection system with the necessary equipment for automated rail inspection. The challenge was to design and develop a mechanism to carry the EMAT probes along the rail surface when the vehicle is operated on rails but which could be retracted to allow normal vehicle operation off rails. To transport the EMATs along the rail surface for defect inspection, a transducer carriage system was designed and installed under the inspection vehicle. This carriage consists of an enclosure, the transducer assembly and of two levels of pneumatically activated articulations (Figure 3). The transducer assembly is the part that is brought in contact with the rail surface. It also contains a winding mechanism with a polymer tape used as a protective layer between the transducers and the rail surface. Flanged wheels are placed at both end of the assembly to guide it along the gauge side of the rail and maintain the transducers over the centre of the head.

Fig 2: Integrated mobile inspection system.

Fig 3: Designed transducer carriage and holder.

To operate the inspection system, state-of-the-art windows based software was designed and developed to support high-speed data acquisition, data analysis (Intelligent interpretation and recognition) and system management. The developed system architecture is composed of nine computer stations distributed at three levels and communicating via IPX/Ethernet connections.

INSPECTION RESULTS

The prelimenary detection capabilities of the system were assessed over 150 meter evaluation track at Canadian National Taschereau Yard in Montreal. These tracks are specially designated for system test calibration and have been prepared with defects of various types. During these tests, different EMAT configurations were used to evaluate their detection performance at 10 to 15 Km/h. Table 3 describes the 12 defects known to be in the track and the type of probes used to detect them with the rail inspection system.

Defect	Defect Name	Defect Acronym	Main Detection Transducer	Secondary Transducer Detection	Position	Dimensions
1	Horizontal Split Head	HSH	0° SR	90° SH	Rail Head 1/2" from top of rail	1/16" milled slot, 2" long
2	Bolt Hole Crack	BHC	35° SV	0° SR	Rail Web 45° from top of Bolt Hole	1/16" machined slot,1/2" long
3	Vertical Split Head	VSH	80° SH	0° SR	Rail Head Gauge side	Cut and reweld 6" Long
4	Defective Weld (in flash butt weld)	DW	80° SH	0° SR 35° SV	Rail Head 5/8" from top - 20° from web head separation	Flat Bottom Hole Drill 1/4" x 2-3/16"
5	Split Web	SW	0° SR	35° SV	Rail Web 1/2 thickness of web	Milled Slot 2" long
6	Vertical Split Head	VSH	80° SH	0° SR	Rail Head Field side	Cut and reweld 6" Long
7	Bolt Hole Crack	BHC	35° SV	0° SR	Rail Web 225° from top of Bolt Hole	1/16" machined slot, 1/2" long
8	Defective Weld (in thermit weld)	DW	80° SH	0° SR	Rail Head 5/8" from top - 20° from web head separation	Flat Bottom Hole Drill 1/4" x 2-3/16"
9	Head and Web Separation	HW	0° SR	35° SV	Head and web interface	2" milled slot 1/2 thickness of web
10	Transverse Defect	TD	90° SH	0°< SR	Rail head	Machined slot 1/16" x 1/2"
11	Bolt Break Crack	BHC	35° SV	0° SR	Rail Web 315° from top of Bolt Hole	1/16" machined slot, 1/2" long
12	Bolt Hole Crack	BHC	35° SV	0° SR	Rail Web 135° from top of Bolt Hole	1/16" machined slot, 1/2" long
Table 3: Test calibration defects

The following results show the detectability of four type of defects displayed as B-scan images, and observed in the inspection cabin. The horizontal axis on the B-scans corresponds to the displacement along the rail tracks and the vertical axis corresponds to the time-of-flight (or distance) across the head rail section.

Fig 4: HSH #1 B-scan	Fig 5: BHC #7 B-scan
Fig 6: SW #5 B-scan	Fig 7: HW #9 B-scan


Most of the defects in Table 3 were detected on the first testing round. Tests were repeatable and accurate at a small range of different inspection speeds. During our tests, only conventional and simple defect interpretation was used to make the calls of the detected defects. However, the system has advanced capabilities for defect characterisation and visualisation included in the software to assist the operator to distinguish between indications (defects) and anomalies (candidate defects).

REFRENCES

1. Maiseri, H., D. MacLauchlan, and G. Alers, "Application of EMAT's to In-Place Inspection of Railroad Rails." DARPA/AFWAL Review of Progress in Quantitative NDE, Materials Laboratory, Air Force Wright Aeronautical Laboratories, OH, 1981 544-547.
2. Graham, L. J. and Martin, J. F., Ultrasonic Inspection of Railroad Rails by Electromagnetic Acoustic Transducers (EMATs) U.S. Department of Transportation, Washington D.C., May 1986.

© AIPnD , created by NDT.net

|Home|    |Top|