What is Low Adhesion?

  1. Home
  2. /
  3. What is Low Adhesion

Low Adhesion Explained

Adhesion on the railway, is a measure of the Traction, or slipperiness, between the wheel and rail. We can measure adhesion levels and a value of adhesion is assigned normally expressed as ‘μ ‘ (a decimal fraction) or sometimes as a percentage.

This μ value is approximately equivalent to the maximum possible rate of deceleration of a given train, when expressed as the percentage of deceleration due to gravity (g). This approximate relationship makes understanding the effects of adhesion on train braking much easier. 

A modern disc-braked train has a nominal braking rate of about 9%g with a Full-Service brake application. Therefore, when all axles are braking their own weight, we need an adhesion level of at least 0.09 to support this braking rate without suffering wheel slide. For simplicity, in the remainder of this manual we shall refer to the adhesion level as a percentage.

For traction purposes, the adhesion level needs to be higher to start a train without the wheels spinning, ranging from 15% for a typical 4 car multiple unit, up to about 25% for a locomotive hauling a heavy freight train. This depends on a number of factors such as the number of motored axles, the axle loads and the trailing load, etc.

In dry weather with clean, uncontaminated rails, the adhesion level would commonly be found to be between 20% and 40%, in really wet conditions it may be between 10% and 20%. 

In both of these circumstances braking problems should not normally be encountered. However, adhesion levels lower than that required for Full Service braking are encountered from time to time, particularly in the autumn when moist crushed leaves on the rails can reduce levels to as low as 1%. 

For those familiar with the 3-Step brake system, this latter value is less than that required to sustain Step-1 braking! 

However, low adhesion can occur at any time of the year when moisture is present combined with contaminants.

Note that the spacing between signals is based on Full Service train braking (with a safety margin added), but braking instructions encourage Drivers to use lower braking rates, applying the brake earlier and possibly lighter at all times and especially when low adhesion is likely or is known to be present. It is convenient to classify low adhesion into a number of distinct bands as shown in the table below:

Adhesion LevelTypicallyDescription
High>15%Clean rails wet or dry
Medium10% – 15%Damp rails with some contamination
Low5-9%Typical autumn mornings due to dew / dampness often combined with light overnight rust
Exceptionally low<5%Typical autumn mornings due to dew / dampness often combined with light overnight rust

“Wet rail” is what happens when low levels of moisture are present at the train wheel/rail interface and can cause poor adhesion. These conditions are associated with very light rain or mist, and the transition between dry and wet rails at the onset of rain. 

Wet rail conditions are the likely cause of around 60% of station overruns and other performance problems.

There are a number of ways to measure the impact of low adhesion but train delays for paying customers is perhaps the greatest.

It can be seen that modern trains demanding higher rates of deceleration will exceed the available adhesion under certain conditions. This is not too much of a problem for ‘typical’ autumn mornings as adjustment to the normal driving technique, driving slower and braking earlier / lighter, should compensate for the reduced adhesion by demanding a lower braking rate. 

These conditions can be compared to driving a road car on a wet road where braking distances will be extended as the tyres cannot Traction as well as they can on a dry road.

On the rare occasions when the rails are severely contaminated, such as with dampened leaf contamination, the adhesion level can be extremely low (levels as low as 1% have been recorded) and other measures are necessary to compensate for this such as the application of sand. 

These conditions can be compared to driving a road car on ice when not only do you need to drive slower but the Council needs to grit the road.

The graph below, produced from actual surveys undertaken in the 1990s by the British Rail Tribometer train, shows how frequently the various adhesion levels occur on the railway. It can be seen that the exceptionally low levels of adhesion (below 5%) are rare as are the very high levels (above 35%). 

For most of the time the available adhesion levels are well within those required to sustain normal braking and traction power demands.

Low Adhesion Improver for Trainlines

Where Does it Come From?

Low adhesion arises from a number of causes, most notably from leaf contamination arising from lineside vegetation. Vegetation on the lineside has spread since the demise of track gangs employed to control it during the days of steam locomotive operation. 

However, it isn’t all the fault of leaves on the line, a number of buffer stops collisions have occurred in the past due to rail contamination caused by station well cleaning scattering oil droplets onto the railhead; and damp, light rust following periods of non-use of the track has led to similar incidents.

It can be seen that the causes can be split roughly into those that the railway has little or no direct control over (naturally occurring conditions such as leaves, moisture, rust, ice etc.) and those that are directly under our control (‘man made’ conditions such as fuel / oil spillages, defective flange lubricators etc.). To an extent, we can also reduce any safety risks to staff and passengers from contamination caused by our neighbours by seeking their co-operation in applying control measures.

The following list is by no means exhaustive but identifies some of the more typical causes of low adhesion.

The biggest single cause of low adhesion problems is as we all know, are the ‘wrong type of leaves’ crushed onto the railhead by the train’s wheels
(under a contact pressure of over 30 tonnes per square inch) to form a hard ‘Teflon’ type coating on the rails. It is known that leaves are drawn into the wheel / rail interface by the aerodynamic effects of passing trains and crushed under the wheels.

This hard coating can cause track circuit operating difficulties when it is dry (acting as an electrical insulator) and causes braking difficulties when damp (acting as a lubricator). With very wet conditions, the crushed leaves are softened and then the layer is broken up by the action of passing trains and washed off the rails by the rain. However, it has been shown through testing that the coating can reform after the passage of just a few trains, if the right drying weather conditions are present.

How is Low Adhesion ‘Activated’?

This is one of the least well-known subject areas. Setting aside fuel, oil and grease spillages which are obvious lubricators, moisture is the common denominator in low adhesion. Dry leaf contamination or dry rusty rails, will not lead to low adhesion conditions, but add a small amount of water then the contaminant becomes ‘activated’. The amount of water required is not generally understood, but more water is generally a good thing as it helps to soften the hard leaf-film layer and wash other contaminants off the railhead. Ice can lead to low adhesion because the melted ice is water. Certain atmospheric conditions will result in dew or light rain which will energise these low adhesion conditions.

In 2014, RSSB investigated the effects of moisture on rail adhesion (T1042). This research collated existing knowledge of the effects of moisture on rail adhesion to synthesise an evidence-based definition of ‘Wet Rail Phenomenon’ and review how mitigations could be improved. An extensive literature search revealed 283 pieces of relevant written evidence, which was then reviewed, and a knowledge map was created. Performance data was also analysed using a variety of sources, including Britain’s national autumn performance and weather data from 2010, 2011, and 2012.

Key Finding Were:

Two main scenarios leading to low adhesion events have been suggested although other factors
may also contribute: