Teknis: Wheel Condition Monitor

Wheel Condition Monitor (WCM)

OVERVIEW

Research was undertaken in the early 1990's in Australia to quantify the effects of wheel defects on heavy haul track structures using concrete sleepers. The primary goal of the research was to understand the spectrum (frequencies) that typical wheel impacts imparted on the structure.

Accelerometers were installed adjacent to strain gauges during this research and it was evident from the results that they were a superior sensor because they provided better coverage and linearity than strain gauges.

It was also discovered that certain signal processing of acceleration data yielded a measure of wheel impacts that was independent of sprung mass, thereby negating the need for traditional normalisation techniques such as "Impact Factors" In addition, because of the continuous 100% wheel circumference sensing it was now possible to identify multiple defects on a wheel.

The Teknis system was developed based on this research and refined in installations over a 5 years period and 18 sites throughout Australia, Europe and the USA.

The Teknis wheel condition monitoring system is a high level integrated wayside inspection station that is based on four important principles:

To operate on "typical" track structure without any modification to that structure. To measure actual effect as opposed to taking a measurement of a symptom.
To measure actual wheelset behavior and not rely on single measures. To analyze actual full dynamics of the train passage and merge data from multiple sources.
A perfect wheel reads zero and so there is no need for separate normalization artifacts such as Impact Factors. Output of wheel analysis is normalized to the fully loaded condition.
To merge data from multiple sensors and sources to enable real-time exceptions based on composite factors like:

Warm bearing & medium grade rough wheel & marginal bogie alignment
Empty wagon with > 2000 tonne trailing
One increasing bad impacting wheel with asymmetric wear and marginal alignment of all other wheels on the bogie.

Installed Teknis WCM systems have provided full return on investment in less than three months by identifying wheel defects known to have caused derailments through broken rails, screwed journals (bearing failure) and broken wheels. In September 1998 one system identified two significantly out-of-round wheels which passed visual inspection on first alarm but were removed from service on the second alarm and found to have been machined incorrectly. An out-of-round wheel, identified by WCM, was found to be a long period defect caused by progressive cracking and collapse of the wheel tread. In these instances it is important to note that other systems may not have isolated these defects because the singular impact forces were not excessive. In another instance, a wheel was identified with multiple sharp defects. The individual impacts were near the medium alarm level. The wheel was allowed to run for 3000km. On its return trip the bearing failed and the train derailed. The operator has since changed their alarm criteria for these specific wagons (bearing types). Identification of multiple defects is not possible with other systems and is one of the strongest areas of the Teknis WCM system - identifying and evaluating multiple defects over an entire wheel surface plus accommodating alarm criteria by wheel defect type as well as by rolling-stock type.

The Teknis WCM system also provides unique measures such as total damage potential, impacts as a function of total train mass and alarm forecasting to assist rail operations in the following risk areas:

TRACK AND STRUCTURE

Rail head damage - especially in highly stressed applications with high axle loads and aggressive wheel profiles.
Rail breakage due to extreme impacts - most often occurring at welds.
Fatigue of structures (ie: bridges) due to long term exposure to high shock forces.

ROLLING STOCK

Premature axle box (bearing) failure
Cracked wheels
Out of round wheels which compromise vehicle tracking
Suspension failure and other collateral damage
Scheduling maintenance and forecasting wheel requirements.
Lowering wheel costs by identifying wheels which require maintenance early, thereby reducing the instances where wheel defects cause irrecoverable damage (eg: deep machining and subsurface damage) The picture on the right is an example of trend analysis and alarm forecasting at the vehicle level.