Noise at Work Regulations: Protecting Your Hearing on Site

Why Noise on Site Is More Dangerous Than You Think

Noise-induced hearing loss doesn't announce itself. There's no cut, no bruise, nothing to point at on the day it happens. You finish a shift on the breaker, your ears are ringing, and by the next morning it's faded and you've forgotten about it. Do that most days for ten or fifteen years and the damage doesn't fade anymore. It's permanent, and there's no fix for it.

That's what makes noise different from most site hazards. A fall from height or a trapped hand gets your attention straight away. Hearing loss builds up quietly in the background while the job gets done. By the time someone notices they're asking people to repeat themselves, or they can't follow a conversation in a noisy pub, the damage was probably done years earlier.

Construction sits near the top of the list for noise-related ill health in the UK. Breakers, grinders, saws, and compressors are everyday kit on most jobs, and a lot of that work happens close to the ear, not at arm's length. Add in reflective surfaces like concrete and steel that bounce sound back at you, and it's easy to end up working in noise levels well above what's safe for a full shift without even realising it.

Tinnitus is the other side of this. A constant ringing or buzzing in the ears that some workers get after just one heavy exposure, like a nail gun going off close to the head, or that builds up gradually over years of grinder and saw work. Unlike hearing loss, tinnitus doesn't always show up on a hearing test, but it can be just as disruptive to someone's quality of life, making it hard to sleep or concentrate.

The law exists because this kind of damage is entirely preventable if it's managed properly. That's the point of the Control of Noise at Work Regulations 2005, and it's worth understanding what they actually require rather than treating "wear your ear defenders" as the whole story.

The Noise at Work Regulations 2005 — What the Law Actually Says

The Control of Noise at Work Regulations 2005 (SI 2005/1643) are the main piece of law covering noise exposure at work in the UK. They replaced older 1989 regulations and brought UK law in line with an EU noise directive, tightening the limits along the way. They apply to every employer, and if you're self-employed and running your own jobs, the duties fall to you too.

The regulations work around three set levels, measured in decibels using the A-weighting scale, written dB(A). This scale adjusts the raw sound level to reflect how the human ear actually perceives different frequencies, which is why you'll see dB(A) rather than just dB when noise regulations are being discussed.

There's also a separate limit for peak sound pressure, which matters for sudden, sharp noises rather than the steady grind of a running tool. A nail gun firing, a cartridge tool, or a hammer strike can produce a peak that's far higher than the average noise level around it, even if it only lasts a fraction of a second. The lower action level for peak sound pressure is 135 dB(C), and the upper action level is 140 dB(C). Note the C-weighting here rather than A-weighting, which is used because it better reflects how impulse noise is measured.

Each of these thresholds triggers a different set of duties. Nothing kicks in until you cross 80 dB(A), but once you're at or above that, the law starts asking things of you. Cross 85 dB(A) and the duties get firmer. Go above 87 dB(A) as an actual exposure once protection is accounted for, and you're in breach of the law, full stop.

Here's how the duties actually break down at each level:

One thing that trips people up: the exposure limit value of 87 dB(A) is measured with hearing protection factored in. It's the actual dose reaching the worker's ear that matters, not the raw noise level of the tool. That's the whole point of hearing protection, it brings the exposure the ear actually receives back under the limit. But it also means you can't use hearing protection as an excuse to ignore how loud the source noise is in the first place. If a worker needs protection that cuts out 30 dB(A) just to get under the limit, the honest answer is that the noise at source is too high and needs sorting, not just muffling.

How Loud Is Your Work? Common Construction Noise Levels

Decibels don't work on a straight line. The scale is logarithmic, which means a small increase in the number represents a much bigger increase in actual sound energy. A jump from 80 dB(A) to 83 dB(A) doubles the sound energy hitting your ears, even though the number only went up by 3. That's the basis of the 3dB exchange rate the regulations use, and it's covered in more detail below.

What that means in practice is that the loudest tools on site eat up your daily noise "budget" fast. A tool running at 110 dB(A) can push you over the upper action value in a matter of minutes, not hours. Here's how some of the most common trade tools stack up, and roughly how long it takes to reach the 85 dB(A) upper action value if you're standing right next to them running continuously.

These figures are typical values, not guarantees for every make and model. The actual noise a tool produces on your job depends on the tool's condition, the material being worked, and the surroundings. A cut-off saw through steel on an open site behaves differently to the same saw through block inside a half-built room with concrete walls bouncing the sound straight back at the operator. Reflective, enclosed spaces routinely push real-world readings higher than the manufacturer's stated figure.

It's also worth remembering these are the levels reaching the operator using the tool. Anyone working nearby, a labourer holding materials steady, someone on the next bench, is exposed too, just usually at a lower level depending on distance. That's part of why noise controls need to think about the whole work area, not just the person holding the tool.

When Do You Need a Noise Assessment?

The trigger for a formal noise assessment is simple: if you think noise on the job might reach the lower action value of 80 dB(A), you need to assess it. In practice, if your tool list includes anything from the table above, that trigger has almost certainly been crossed, and you should treat a noise assessment as a normal part of your paperwork, not an optional extra.

A noise assessment doesn't need to be a specialist exercise with calibrated sound meters for every single job, though for ongoing or high-risk work that level of measurement is the right call. HSE publishes noise exposure data for common tools and tasks that can be used as a starting point, alongside manufacturer's noise data that should come with new equipment. What matters is that the assessment is a genuine attempt to work out realistic exposure, not a guess written down after the fact to fill a gap in the file.

A proper assessment looks at which tools are used, for how long, by whom, and in what environment. It should identify who's exposed above 80 dB(A) and who's exposed above 85 dB(A), because those two groups get different legal duties applied to them. It should also flag any peak noise sources, cartridge tools, nail guns, hammering, that could push past the peak sound pressure action levels even if the average noise over the day looks manageable.

This isn't a standalone document sitting apart from the rest of your paperwork. Noise should show up as a specific hazard in your risk assessment, with the controls, hearing protection requirements, and health surveillance needs recorded against the task, the same way you'd record any other hazard. If you're already working through a risk assessment for a job involving breakers, grinders, or saws, noise needs its own line, not a generic mention.

One thing that catches people out is combining exposures across a working day. A worker who spends two hours on a grinder and three hours on a breaker isn't just assessed against each tool separately, their total daily dose needs adding together. Someone doing short bursts on several loud tools across a shift can end up with a daily exposure just as high as someone on one continuously loud tool, even though no single task looked that bad in isolation.

Practical Noise Controls — From Quieter Tools to Better Planning

Hearing protection is often the first thing people reach for, but it sits at the bottom of the control hierarchy, not the top. The regulations expect you to look at reducing noise at source and limiting exposure before relying on PPE to plug the gap. That's not just a legal box-ticking order, it's genuinely the more effective way to protect people, because it reduces risk for everyone in the area, not just the person wearing ear defenders.

Buying quieter tools when you're due to replace kit anyway is one of the easiest wins going. Manufacturers publish noise data, and there's often a meaningful gap between models doing the same job. A quieter breaker or grinder on the next equipment order costs the same as a loud one and cuts exposure for as long as that tool's in the fleet.

Maintenance matters more than most people assume. A worn bearing, a loose guard, or a blunt blade forcing the motor to work harder all push noise levels up. A tool that's been well looked after is usually the quieter one on site, on top of being the safer one to use.

Barriers and enclosures can cut noise for people working near a loud task without changing how the task itself is done. Screening off a cutting station, or running a compressor somewhere it isn't bouncing sound directly at a work area, reduces exposure for everyone nearby without slowing the job down.

Job planning is an underused control. Rotating who's on the breaker through the day spreads exposure across the team instead of loading it onto one person. Scheduling the noisiest tasks so they don't overlap with other trades working close by keeps more people under the action values without extra cost. And where it's practical, keeping noisy tasks physically apart from the rest of the team, running the cut-off saw at the far end of the plot rather than in the middle of everyone else's work, reduces the number of people who need to be considered in the assessment at all.

Choosing and Using Hearing Protection

Once noise at source has been reduced as far as reasonably practicable, hearing protection covers the remaining risk. But "any ear defenders will do" is a mistake that undoes a lot of the good work elsewhere. The right protection depends on the noise, the task, and the person wearing it.

The HML method (High, Medium, Low) is the standard way of matching hearing protection to the actual noise you're dealing with, rather than just picking whatever has the biggest number on the packet. Different tools produce noise weighted toward different frequencies, some skew toward low-frequency rumble, others toward high-frequency screech, and the HML figures on hearing protection packaging tell you how much attenuation that product gives across high, medium, and low frequencies specifically. Matching protection to the frequency profile of your actual noise source, rather than just chasing the highest single number (SNR), gives a more accurate picture of how much real protection you're getting.

This connects to a risk that surprises a lot of people: over-attenuation. Protection that cuts out far more noise than needed isn't automatically the safer choice. If a worker can't hear a reversing alarm, a shout of warning, or a colleague trying to get their attention because their ear defenders are blocking everything, that's a new hazard being introduced to solve the noise problem. The goal is protection that brings exposure to a safe level, not protection that isolates someone from their surroundings entirely.

Fit matters as much as the rating on the box. Disposable foam earplugs that aren't rolled down properly and inserted deep enough into the ear canal can lose most of their stated protection, sometimes down to a fraction of what the packaging claims. Training people to fit earplugs correctly, and checking that fit occasionally rather than assuming everyone's doing it right, is a cheap step that makes a real difference. Earmuffs have their own failure points too. A worn or cracked cushion seal, or a pair squashed at the bottom of a toolbox for months, won't seal against the head properly and lets noise leak straight through.

Reuse is another common slip-up. Disposable earplugs are designed for limited use and should be replaced daily, not kept in a pocket and reused for a fortnight once they've gone grubby and lost their shape. Reused foam plugs compress less effectively and don't expand back to size the way a fresh one does.

Under the Personal Protective Equipment at Work Regulations 1992 (SI 1992/2966, as amended by SI 2022/8), hearing protection has to be provided by the employer free of charge once it's required. That cost doesn't get passed down to the worker in any form, deductions, charges, nothing. For more on where PPE costs sit and who's responsible for what, see our guide to PPE requirements under UK law.

Health Surveillance — Protecting Your Team's Hearing Long Term

Health surveillance for noise means regular audiometric testing, a hearing test carried out by someone qualified to do it, usually against a baseline test taken when the worker first starts in a noisy role. The point isn't just ticking a compliance box. It's catching hearing damage early, while it's still small enough that changing how someone works, tightening up their hearing protection use, moving them off the loudest tasks, can actually make a difference before the loss becomes permanent and irreversible.

Health surveillance is required for anyone regularly exposed above the upper exposure action value of 85 dB(A), and it should also be considered for workers exposed between 80 and 85 dB(A) if the risk assessment flags a particular risk, for example someone already showing signs of hearing loss or known to be more susceptible to noise damage.

A baseline test when someone starts in a noisy job gives something to compare against later. Follow-up tests then get repeated periodically, and the results should be reviewed by someone competent to interpret audiometric data, not just filed away. If a test shows a change from the baseline, that's the trigger to look again at how that worker's noise exposure is being managed, and whether their hearing protection or working pattern needs to change.

Keeping records matters here too. Health surveillance results need to be kept, and workers are entitled to see their own results. Employers also need a system for acting on results that show a problem developing, rather than just recording it and moving on. A hearing test that flags a decline and gets no follow-up isn't much better than not testing at all.

There's a reporting duty that sits alongside all this. Occupational hearing loss becomes RIDDOR-reportable once it's been diagnosed and a doctor has confirmed it was likely caused by work. When that happens, it needs reporting using form F2508A within 10 days. This isn't something to report on a hunch or a worker mentioning their hearing "isn't what it used to be", it's specifically triggered by a medical diagnosis linking the condition to work exposure. Our guide on RIDDOR reporting requirements covers the wider reporting duties if you want the full picture of what else needs reporting and when.

Frequently Asked Questions