Wheel-off accidents prevented with data

Published: 10.3.2023

A wheel coming off a bus or truck is a serious hazard. This needs no further explaining. You don’t want to see a loose 100 kg+ wheel bouncing towards you at 90 km/h. 

There are best practices and industry guidelines to prevent wheel-off accidents but still they happen in surprisingly big numbers. There is only so much you can do with conventional methods. The elements of human errors, technical problems and simply bad luck will remain no matter how much effort is put into preventing the incidents. 

Why is data not used to predict and avoid wheel-off accidents? Because such data is not available, said an experienced fleet manager when I asked him. That is a common understanding, but it is not true. The data is available but until now it has not been understood. 

We at E3 Innovations recently analyzed wheel-off accidents hoping to understand if and how they could be predicted using the data before the wheel actually gets loose. Yes they can! With our 3D dynamics calculation. 

Here are some examples of how a wheel-off accident shows in data and more importantly how it can be predicted hours or days beforehand. In all cases the wheel has said “I’m about to get loose, do something!” Unfortunately in these cases we did not know what to look for and the wheels got loose. Now we do know.

Example 1.

Rear left starts resonating (vertical axle) several hours before the wheel goes off.

Wheel is mounted back in place and everything is ok. Problem fixed.

Example 2.

Rear right tire starts to resonate two days before getting loose. When the tire is mounted back the core problem has clearly not been fixed (yellow diagram acting wild).

If nothing further is done, something bad will happen again soon. It can be another wheel-off, bearing damage, parts getting loose due to vibration (steering, chassis, body)

Example 3.

Rear right starts resonating hours before getting loose. Front right needs immediate attention.

Rear right is back in place and does not resonate abnormally anymore. The problem causing the wheel-off has been fixed.

Why has the front right not taken off yet? That we don’t know, but what we do know is that something is wrong with the front right and it needs immediate attention.

Having seen the signs that predict wheel-off we can set our system to automatically look for them and warn the customer when there is still time to react.

How do we get this information?

Simply said, We use our E3 Blue vehicle computer to read the data from the vehicle’s ABS/EBS system and J1939/FMS data buses. This data is analyzed by our E3 Sense software which runs in the E3 Blue computer.

The outcome of this data analysis is a dynamic model of the vehicle which compares the actual behavior of  the vehicle to the desired behavior. Deviations in those two are problems needing to get fixed. The amount of data provided by the vehicles is much bigger than what can be used in a way that would make sense, but it enables highly detailed analysis of the things we want to monitor.

Let’s take tires as an example. Knowing the turning speed of the tire enables calculating a good enough estimate of the tire diameter. And that’s about it.

Understanding the tire behavior in more detail and adding other vehicle data into the equation enables a whole different level of monitoring and analysis. Air pressure, tread depth, brake balance, bearings, suspension etc. And now predicting wheel-off incidents.

When a wheel makes a full turn it travels about 3,3 meters. Speed is not constant through that distance. If we look close enough we see that the wheel speed is constantly changing. On top of that wheels are always sliding. On top of that, wheels are not only flexible in the longitudinal direction but also latitudinal.  On top of that, wheels constantly take hits.

Adding up different wheel behavior related data and integrating it with other vehicle data, such as engine torque, enables providing highly detailed and reliable information on not only the condition of the tires but also the condition of the chassis. Wheel-off detection is the latest example.

 

Cost savings

The safety aspect of avoiding wheel-off accidents is clear but there are also material cost savings available. 

When a wheel gets loose while driving, the bearing is almost always damaged. In addition other damage is often done to the chassis or body. These alone add easily up to 7000 euros or more. And of course, anything that a loose tire hits or injures will be charged from the operator. 

 

If you are interested in increasing safety and decreasing cost, please contact:

Pekka Möttö, Head of Growth, E3 Innovations Oy, +358 44 974 2477, [email protected]

Jarmo Mäki, Sales Director, E3 Innovations Oy, +358 40 127 9686, [email protected]

E3 Innovations Oy

Helminraitti 50
Lohja, Finland
[email protected]