To investigate how to block noise emanating from a cooling system at the Uskmouth B power station, a simulation of the duct system was built
To investigate how to block noise emanating from a cooling system at the Uskmouth B power station, a simulation of the duct system was built

When CMS Danskin Acoustics was brought in to reduce dangerous noise levels at a UK power station it had to create a tailor-made efficient and cost-effective solution.

Paul Absolon, CMS Danskin Acoustics, UK

The fact that power stations produce high levels of noise will not come as much of a surprise to anyone who has spent more than a few minutes inside one. However, just how dangerous these levels of noise can be might.

To give you an idea of the sensitivity of the human ear, the average person can hear sounds down to about 0 decibels (dB), the equivalent of a whisper or rustling leaves, and people with exceptional hearing can detect sounds as low as -15 dB.

It is at the other end of the scale, however, that serious problems can occur. At around 85 dB, you are likely to experience discomfort. As little as eight hours of continuous exposure to this level of noise can result in permanent damage to the inner ear, which is why 85 dB is the maximum allowed under the Noise at Work regulations.

At 100 dB, just 15 minutes of exposure can cause permanent damage. And at 110 dB the exposure time drops to around a minute before harm is inflicted. Pain is experienced at 125 dB and hearing loss can be permanent. At 140 dB or above not only is the damage permanent, it is also immediate.

The Noise at Work regulations stipulate that an “employer shall ensure that risk from the exposure to noise is either eliminated at source or, where this is not reasonably practicable, reduced to as low a level as is reasonably practicable”. And that “if any employee is likely to be exposed to noise at or above an upper exposure action value, the employer shall reduce exposure to as low a level as is reasonably practicable by establishing and implementing a programme of organisational and technical measures, excluding the provision of personal hearing protectors, which is appropriate to the activity”.

In other words, an employer is expected to do everything they can, within reason, to protect their employees from the harmful effects of noise. And, no, it isn it enough to simply supply a pair of ear-defenders.

The damage caused by these dangerous levels of noise is referred to as Noise-Induced Hearing Loss, and can be caused by a single exposure to a very loud sound or by repeated exposure to even relatively low levels of noise over a long time span. In fact, according to research carried out by Xiaoming Zhou from the East China Normal University in Shanghai, even seemingly innocuous sounds, such as the whirr of a desk fan, can cause damage if exposure is consistent and long term.

The human ear does not hear all frequencies with the same intensity. It is most sensitive to sounds in the 500 Hz to 8 kHz range. Above and below this range the ear becomes progressively less sensitive. To compensate for this, sound level meters incorporate electronic filtering to correspond to the varying sensitivities of the ear. This filtering is called A-weighting and readings obtained with this weighting are referred to as A-weighted and signified as dB(A).

Tailor-made solution

Uskmouth B power station is a combined-cycle gas turbine plant near Newport in Wales, built by Siemens and operated by Severn Power, a subsidiary of Dong Energy.

Acoustics and soundproofing specialist CMS Danskin Acoustics was brought in by Siemens and SPX Cooling Technologies after the recorded noise coming from the dry cooling system was between 130 dB(A) and 135 dB(A), a full 50 dB above Noise at Work regulations’ acceptable levels.

Although it was identified that these dangerous and unacceptable levels of noise came from the dry cooling system, the cooling system was not creating the noise. In fact, the turbines proved to be the source of the noise.

The steam roaring from the turbines at incredibly high speeds enters the main 5.5 metre steam ducts, passes up five risers and is channelled into the steam distribution manifolds. Not only does the steam enter the dry cooling system, the accompanying noise does, too. You might think that the 8 mm thick steel from which the ducts are constructed would go some way to containing the noise. Unfortunately, steel is extremely adept at transmitting noise and is, in many respects, the acoustician’s worst enemy.

To make matters worse, we discovered not only high levels of noise, but also that the noise generated had a very low-frequency bias.

Low-frequency noise is the most difficult to treat from a soundproofing perspective due to the excessive length of the wave cycle. This is one of the reasons people in apartments, terraced houses and semi-detached homes will often complain of the problems of bass noises intruding from neighbouring properties, as the walls and floors filter out the higher frequencies while the lower frequencies manage to penetrate. This can seem a little counter-intuitive, as we imagine higher frequency noise to be more piercing. Their short wave cycle, however, means they can be blocked out with relatively thin soundproofing materials.

The low-frequency nature of the noise also meant that this was not just a Noise at Work regulations problem. Low-frequency noise can be particularly problematic to the population in the vicinity of the source of that noise.

A new sound solution

Solutions for low-frequency noise issues typically involve wrapping the problem in significant quantities of acoustic insulation, with many standard solutions being as deep as 500 mm to 700 mm. The sheer volume of lagging required for an insulation-based approach to a project like Uskmouth, with a daunting 8000 m2 of ducting to be covered, would be expensive, time-consuming to install and prohibitively disruptive. What is more, there were areas around the ducting at Uskmouth which simply would not have been able to accommodate such an excessive construction height of soundproofing material.

CMS HT1B elastomeric isolation pads form the first layer of acoustic insulation for Uskmouth B's dry cooling system
CMS HT1B elastomeric isolation pads form the first layer of acoustic insulation for Uskmouth B’s dry cooling system

We had to create a ‘thinner’ soundproofing system that would meet the necessary Noise at Work regulations requirements but would be cost-effective and efficient to install.

In order to minimise disruption at Uskmouth, an off-site simulation was created near Burton-on-Trent, using a large section of identical ducting with a ‘door’ sealing up either end. Within the duct were several very powerful speakers. For testing, highly sensitive microphones were placed in strategic positions along the outside of the duct to measure any ‘leakage’. Acoustic insulation solutions were conceived, implemented and assessed in this controlled environment with the assistance of acoustic consultants Muller-BBM and the installation company Western Thermal Insulations.

Exploring a wide range of acoustic materials from CMS Danskin Acoustics’ industrial acoustics range, it became clear that a single product was not going to be able to solve the problem on its own, so we opted for a combination of products working in concert, layer upon layer.

The first layer consisted of CMS HT1B elastomeric isolation pads, constructed from a polyurethane-bound rubber granulate specifically formulated to dampen and/or isolate noise and vibrations at source and independently tested by the Institute of Structural Dynamics at the Technical University of Dresden, Germany. The 50 mm thick pads were bonded to the surface of the duct at a rate of nine per m2, creating 300 mm spacings; so, as well as the dampening effects of the material itself, the construction benefited from large, evenly distributed airspaces in its foundations. Sound waves move less effectively through dead air.

The second layer consisted of 50 mm QuietSlab SVX3, a high-performance, mineral-fibre acoustic lagging.

A combination of materials in layers was the chosen soundproofing solution at Uskmouth B
A combination of materials in layers was the chosen soundproofing solution at Uskmouth B

The third layer comprised CMS WBBKT acoustic barrier, a high-density, barium-sulphate-loaded thermoplastic polymer, which is thin, flexible and easy to work with. Whereas the QuietSlab SVX3 layer is designed to absorb and dissipate noise, this dense acoustic barrier is designed to resist the passage of noise and is particularly adept at preventing the passage of low-frequency noise.

The forth layer duplicated the second, the fifth layer duplicated the third and the sixth and final layer consisted of a corrosion-resistant Aluzinc casing.

By alternating between thick noise-absorbent layers and thin but dense noise-resistant layers, we were able to create a soundproofing solution with a depth of just 170 mm – between 66 per cent and 76 per cent thinner than a 500–700 mm standard solution. However, the successful reduction of the construction height would mean nothing at all if it failed to deliver the necessary levels of noise reduction. The proof would be in the testing.

Personnel from Siemens attended the test. They were standing in relatively close proximity to the simulated duct while technicians from Muller-BBM set up their equipment. As always with these situations, there were delays, so the Siemens team were standing around for quite some time. Naturally, they were a little impatient and asked when the test was going to commence. They were told the test had been running for the last ten minutes. The speakers within the ducts had been generating noise levels of 130–140 dB and no-one had noticed. Only when the lagging protecting the ‘door’ to the duct was removed could the true extent of the racket within be appreciated.

The testing revealed that CMS Danskin Acoustics’ solution cut the noise generated by 39 per cent to just 82–83 dB(A), well under the 85 dB required by the Noise at Work regulations.

Peter Ullrich, project director at Siemens Energy, says: “Effectively controlling noise and reducing sound emissions was a top priority for us in the Uskmouth project. Not only was it essential that the dry cooling system satisfied all the legal acoustic obligations and regulations but just as important was that neighbouring properties were not disturbed by additional noise levels.”

Paul Absolon is techncial director of CMS Danskin Acoustics, a specialist in acoustic insulation, sound absorption and reverberation. For more information, visit


To understand how Noise-Induced Hearing Loss (NIHL) occurs, it is necessary to understand how hearing works.

The generally accepted view is that sound waves strike the eardrum and these vibrations are translated into coherent information by the brain. It is more complex than that.

Sound waves do, indeed, strike the eardrum, causing the eardrum to vibrate. These vibrations are then transmitted through the ossicles (the small bones of the middle ear) to the cochlea, a spiral-shaped chamber filled with fluid and lined with tiny hair cells called stereocilia.

The vibrations cause the fluid to move which, in turn, causes the stereocilia to move. The stereocilia’s movements generate neural signals with are picked up by the auditory nerve which ‘forwards’ these signals onto the brain where they are interpreted as intelligible sounds such as human speech, music, the beep of a car horn.

Exposure to harmful levels of noise can damage the stereocilia, breaking them or flattening them so they no longer vibrate as effectively or so they no longer vibrate at all. The result: impaired hearing or, in extreme cases, total hearing loss.

Low frequency noise is often not even ‘heard’ in the traditional sense. Complainants often will not even realise that noise is the problem at all; instead they will describe ‘pressure sensations’ and ‘physical discomfort’, experiencing the incursion as vibrations.

Areas of the human body can resonate when exposed to low frequencies. The chest, for example, can resonate at frequencies between 50 Hz and 100 Hz, and the head at frequencies between 20 Hz and 30 Hz. It is not unusual, therefore, for sufferers of low frequency noise to complain of anxiety, nausea and headaches. Often, they will not even be aware of the root cause of their symptoms, instead attributing them to a virus or some mystery illness.