Vibrations induced by football fans on stadia grandstands

Sports stadia play host to large, lively crowds of people and, in order to accommodate the maximum capacity, frequently incorporate long, flexible, cantilevered grandstands. These are designed to be column-free in order to provide spectators with an uninterrupted view, which has led to designers using structures more susceptible to dynamic crowd loading.

As a result of the Hillsborough disaster of 1989, Premier League football stadia must only accommodate seated spectators. In recent years, football supporters and some professional football clubs have solicited the reintroduction of standing areas.

A series of laboratory tests investigate how the introduction of safe standing areas to increase capacity in football stadia will impact the dynamic loading. A series of actions were identified that represent common actions performed by seated and standing spectators causing the largest dynamic loading. These actions were reproduced in dynamic laboratory over the floor structure.

Sports stadia test photo 3    Sports stadia test photo 4

In order to represent the proposed increase in capacity that is facilitated by the installation of safe standing areas, the standing tests used a larger number of subjects than the seated ones (5 instead of 3), placed in the same area. To best recreate the feeling of being in an actual stadium, the actions were performed in response to visual and aural stimuli from video footage of goals being scored in football matches.

Data on the motion of each person was collected using wireless sensors and the response of the floor was measured using wired accelerometers. The wireless sensors indirectly estimated the force exerted by each person. Preliminary results show that, for lively crowd actions, the introduction of safe standing areas could cause an increase in the dynamic response, but for less lively crowds, a likely decrease.

Sports stadia plot 1

Dynamic forces (blue) measured by APDM Opal™ placed on sternum for seated and standing tests, and total weight of people (red)

Sports stadia plot 2

Maximum structural responses as a function of dynamic force peaks (Sternum model) for seated (left) and standing (right) tests

Got a question about MSc Structural Engineering?

Talk to our recruitment team

+44 (0)1392 725017

Crowd Loading Experiment


The Vibration Engineering Section collaborated with consultant WSP-Parsons Brinckerhoff in a “crowd loading experiment” held in the Guildhall in Exeter, which is undergoing massive refurbishment. The experiment was undertaken with the permission and supervision of ISG, the principle contractors for the site.


Participants were asked to congregate around the column heads of the ground floor and move from one position to another; in the meantime, strains in the columns supporting the 1st floor were measured, using vibrating wire strain gauges.


This type of experiment is designed to ascertain how much load a certain structure can support in a real-life situation. The results will be compared with the designed values largely dictated by codified assumptions such as the Euro Codes. The end game is to propose alternative design values for procedures and practices.

20150731_165655 (00000003)

Find out more about MSc Structural Engineering

Running tests at Exeter Arena

Running tests at Exeter Arena

Crowd loading of bridges can become a source of undesirable vibrations. A canonical example of this is the widely publicised wobbly behaviour of the London Millennium Footbridge on its opening day. Much effort has been made in recent years to understand the cause of this problem in the case of walking pedestrians. However, little attention has been given to the effects of running crowds.

With the proclaimed “renaissance in running”, running crowds are becoming a common sight in urban environments. An increasing number of marathons are organised from year to year in major cities around the globe. These events are often celebrated by large crowds of spectators, accompanied by music (scroll to 2:10 to see a brass band crossing Bosporus Bridge during Istanbul marathon in 2014) across major bridges. While it is well known that spectators can react to the music, often synchronising their actions to its beat, how the music affects the behaviour of runners is uncertain. Uncertainty also exists in interactions between runners that might prompt synchronisations of their footsteps. However, synchronised groups of runners could input large amplitude resonant force to the bridge, causing a build-up of undesirable vibration, similar to those evidenced on the London Millennium Footbridge.

In order to shed light on this problem, a series of tests have been conducted at Exeter Arena. Wireless headphones were used to provide auditory signals, prompting some runners in a group to synchronise their steps, while some other runners were unaware of this stimulus, hence allowed to act freely. Data on the motion of each runner was collected using wireless sensors. This data was used to analyse synchronisation between runners, as can be seen on the plots below.

Plot - running tests at Exeter Arena 2 Plots - running tests at Exeter Arena

The top image: patches with alternating colours indicate two runners stepping at different frequencies (taking different number of steps to cover the same distance in a given time). The bottom image: red patches indicate periods at which two runners were stepping at the same frequency.

The presented preliminary results indicate that synchronisation between runners can indeed occur spontaneously or prompted by the beat of music, hence the problem of excitation of structures under running crowds requires further investigation.

Find out more about MSc Structural Engineering

Mid-span deflection monitoring of Humber Bridge using cameras

Humber Figure 1 Humber Figure 2 Humber Figure 3 Humber Figure 4   Humber Figure 5Humber Figure 6     
These photographs show an experiment carried out by the Vibration Engineering Section earlier this month to investigate the feasibility of monitoring the mid-span deflection of a suspension bridge using a camera-based monitoring system. Accurate displacement values are useful for bridge health monitoring; however, historically, displacement has been difficult to measure on bridges.

The bridge monitored was the Humber Suspension Bridge near Hull in the UK. Figure 1 shows a photo of the bridge. To monitor the displacement, we mounted a custom made target to the bridge parapet at mid-span, then we tracked the target using a camera, which was placed at the base of the tower as shown in Figure 2. Figure 3 shows a close in view of the equipment at the base of the tower. The camera was mounted on a tripod so, as the bridge deck moves, the ‘target’ moves in the image frame. Software was used to track this movement and quantify the displacement.

Figures 4, 5 and 6 show the movement of the bridge deck at mid-span as a truck passed.

Human Perception of Vibration Testing

HPOV 3 HPOV 2 DSC_4233 (00000006) biomechanics_visualisation

These photographs show a current experiment being undertaken in our Structures Lab. This work investigates human perception of vibration in a structure. The vibration serviceability of engineering structures can be rationalised into three parts of scientific problems: The vibration source, the transmission path and the receiver system.

This experiment was about the pedestrian’s perception of underfoot vertical vibrations. It aimed to construct a relationship between the walking student’s subjective sensations and the physical quantities of vertical footbridge vibrations.

This was studied by using cutting edge wearable sensors, a dedicated lab structure to simulate a footbridge, along with electricity driven shakers to excite the vibration of the structure and a treadmill for the students to run on.

The findings of this investigation can be used in engineering design of footbridges and walkways to permit a balance between the construction cost of the structures and the vibration serviceability.

Find out more about MSc Structural Engineering

Ambient modal testing of a seven-story building

Figure 1 Figure 2 Figure 3 Figure 4 Figures 5 and 6

These photographs show a recent ambient modal test of the seven-story building shown in Figure 1. The purpose of test was to check the feasibility of using wireless sensors to monitor lateral building vibrations. The test was carried out by the members of the Vibration Engineering Section shown in Figure 2.

Acceleration measurements were taken, using both wired and wireless accelerometers. The accelerometers used are shown in Figure 3, the wired accelerometers are connected to a cable reel as shown in Figure 3 and then the cable reel carries the signal to the data acquisition system shown in Figure 4. Figure 5 shows a 50 second portion of the signals recorded. Because the accelerometers are oriented horizontally, the signals have a mean value of zero or very close to zero. Figure 6 shows a zoomed in version of the signal between 18 and 22 seconds. Despite the signal from the wireless sensor containing slightly more noise than the wired sensor, broadly speaking there is good agreement between both sensing systems.

Find out more about MSc Structural Engineering