An optical monitoring system has been developed by Sheffield University (Sheffield, England) that captures damage on the running surface of railroad tracks, continuously monitoring initiation and stabilization of dangerous wear without test interruption, a first for this type of system. The line-scan imaging system leverages the high power of the Chromasens Corona II LED illuminator to ensure needed brightness up to 500,000 lux.
Repeated cyclic interaction between railroad track and a train’s wheel often produces wear and what is known as “Rolling Contact Fatigue” (RCF) damage. Accumulation of damage can affect the integrity of the steel and the safe running of a railway network. For instance, RCF can change the rail profile and widen the track gauge, potentially leading to dynamic issues. If left unchecked, wear can also cause rail breakages and shelling. Expensive rail maintenance, such as the undertaking of grinding, milling, weld repair, or full replacement, is required to counter this damage.
Ductility exhaustion of a rail’s steel surface will cause material to be removed as “wear flakes,” whereas RCF damage is related to the length and depth of cracks, including ratcheting initiation and early growth, fluid-assisted crack growth, and rail bending and branching crack growth.
A variety of monitoring approaches have been used to investigate wear and RCF with the most promising being a Small-scale Twin Disc Machine. Its drawback is that data about rail steel wear and RCF is obtained only after a test has been interrupted or is finished, therefore providing a record intermittently instead of constantly. Moreover, it can be prohibitively expensive to conduct separate tests of different contact cycle durations, over and over. Interruptions also introduce transient effects into the experiments due to the delay in re-establishing contact load and traction after resumption.
With support from British Steel, the Sheffield University scientists sought a way to circumvent the regular interrupting of testing when using a small-scale twin disc machine. Their solution was an optical monitoring system engineered as two detachable modules — a camera module and a line light driver module — that can be mounted onto a SUROS2 twin disc machine. When in use, the two modules are connected by an intermediate slider on the machine guarding, while the line light driver module is mounted onto the rail roller driveshaft frame. This configuration allows the SUROS2 system to move with the rail sample it is testing, creating a stable image of the rail disc running track.
LINE SCAN IMAGING SYSTEM
A color line scan camera equipped with a manual focus Nikon f/2.8105 mm macro lens was used to capture images of the running track during testing. A sequence of line scans taken at high frequency while the disc is rotating was stitched together to create a complete 2D photograph of the entire disc circumference. The line scan camera and macro lens combination have a 10 μm sensor pixel size and an optical magnification of 0.5, with the spatial resolution being 20 μm/px. Horizontal field of view (FOV) was set at 800 pixels wide, allowing the captured images to include just the running track and exclude extraneous data. During testing it was found capturing 12000 line scans per image provided overlap, just in case a region of interest lay at the edge of the image, giving a total image size of 9.6 megapixels.
Once captured, the data was recorded using a frame grabber card connected using CameraLink cables. Post-processing of the data was carried out using XCAP image processing software developed by EPIX32.
THE IMPORTANCE OF LIGHT
Due to the extremely short line capture times, it proved necessary for the running surface of the rail twin-disc samples to be brightly illuminated. To provide sufficient lighting, a Chromasens Corona II LED line light was integrated with the optical monitoring system. The Chromasens light source delivers up to 500,000 lux of illuminance onto the sample running track at the required 95mm operating distance imposed by machine construction and guarding restrictions. It is controlled by a Chromasens XLC4 control unit that communicates with the optical monitoring computer via an ethernet cable and is programmed using Chromasens XLC4 Commander software.
The position of the Chromasens LED line light in the module is easily changed, permitting the light source to provide either brightfield or darkfield illumination. Brightfield illumination involves the LED line light positioned so that the main reflected light path from the sample runs directly into the line scan camera. For darkfield illumination, the line scan camera is instead positioned to receive light that is reflected from any defects on the surface.
PROMISING RESULTS
Results showed the system can clearly record the development of wear flakes from the point at which the flakes first visibly initiate and later when they stabilize in size to be observed. Analysis of the images obtained showed a correlation between the observed wear flakes and mass loss wear results, indicating the potential for this system to quantify wear behavior without the need to interrupt the test.
After an RCF crack was identified by the system it was shown possible to track the crack’s development throughout the test. Data acquired demonstrated the potential for identifying RCF crack initiation sites at a much earlier stage than previously possible with established techniques such as eddy current detection.
Application of the optical system is not confined to rail-wheel material investigation and has potential use in a range of tribological testing. In its immediate future the optical system will provide much greater insight than previously thought possible to the origin of rail surface damage, as well as the behavior of products such as flange lubricants or top of rail friction modifiers commonly used to improve wheel/rail interaction and fuel economy.
Learn more at chromasens.de/en/product/illumination
