When Does OSHA Require a New Noise Survey? 1910.95(d)(3)
November 17, 2026 · 6 min read
By Jeff Schroeder, Foster Special Instruments
Most plants did a noise survey once, put the results in a binder, and built the hearing conservation roster from it. Then the line got a new press, a second shift started running the grinder, and a sound enclosure came off for maintenance and never went back on. The binder still says the same thing. This post covers what OSHA actually requires for repeat noise monitoring, what counts as a trigger, and the signals that tell you a survey is overdue.
What does 1910.95(d)(3) say about repeat noise monitoring?
The rule is performance-based, not calendar-based. Under 29 CFR 1910.95(d)(3), monitoring shall be repeated whenever a change in production, process, equipment, or controls increases noise exposures to the extent that either of two things may happen. First, additional employees may be exposed at or above the action level, which is an 8-hour time-weighted average of 85 dBA. Second, the attenuation of the hearing protectors employees are using may become inadequate to meet paragraph (j) of the standard. Both conditions turn on the change raising exposure. A change that lowers noise, such as a new enclosure or a quieter compressor, does not create a duty to remonitor, although you may want data to justify removing people from the program later. The standard does not set an annual survey requirement, and it does not give a fixed number of days to complete the new survey once a trigger occurs. In practice, an inspector will expect the survey to follow the change closely.
Why doesn't OSHA require a survey every year?
An earlier version of the hearing conservation amendment did set a fixed cycle, and OSHA replaced it with the change-driven language when it revised the amendment in 1983. The reasoning was that periodic resurveys cost money without telling an employer anything new when nothing on the floor had changed. OSHA's own Appendix G on monitoring noise levels still notes that many companies choose to remonitor periodically, once every year or two, to make sure every exposed employee is in the program. That is a practice, not a rule. The risk of skipping it is that small changes add up. One extra machine does not move the needle, but three years of added equipment, longer shifts, and worn mufflers can push a department across 85 dBA without any single event that someone would flag as a trigger.
Which changes count as triggers?
The standard lists four categories: production, process, equipment, and controls. The table below maps common floor changes to those categories.
| Change on the floor | 1910.95(d)(3) category | Why it can raise exposure |
|---|---|---|
| New or relocated machine | Equipment | Adds a source or moves one closer to workstations |
| Higher line speed or added shift hours | Production | More cycles per shift, longer daily exposure |
| Switch to a new material or tooling | Process | Harder materials and impact tooling often run louder |
| Enclosure, silencer, or barrier removed or damaged | Controls | Engineering control no longer reducing noise |
| Job rotation or new task assignments | Production | Employees spend time in louder areas than before |
| Change of hearing protector model | Controls (indirect) | Lower attenuation can fail paragraph (j) at the same noise level |
Shift length matters more than people expect. OSHA uses a 5 dB exchange rate, so exposure time and level trade off. Moving a crew from 8 to 10 or 12 hours at the same sound level raises the time-weighted average, which can put people over the action level without the noise getting louder. Our post on the OSHA 5 dB exchange rate walks through the math.
What does "hearing protectors rendered inadequate" mean in practice?
Paragraph (j) sets the target. Hearing protectors must reduce an employee's exposure to an 8-hour TWA of 90 dBA or below, and for an employee who has had a standard threshold shift, to 85 dBA or below. Protector adequacy depends on the exposure level, so when noise goes up the same earplug can fall short. Appendix B of 1910.95 gives methods for estimating protector attenuation from the Noise Reduction Rating, including subtracting 7 dB from the NRR when the exposure was measured in dBA. If your last survey showed a department at 95 dBA and a new process puts it at 103 dBA, the protector selection made for 95 may no longer meet (j). That is a (d)(3) trigger even if no new employees cross 85 dBA, because everyone in that area was already in the program. The fix is new monitoring data, then a fresh look at protector selection for that area.
What signals suggest a survey is overdue?
Not every trigger shows up on a change request. These are the signs we see most often that exposure has drifted since the last survey:
- A cluster of standard threshold shifts in one department or job title in the same testing year. See what OSHA requires after an STS.
- Employees outside the program asking for earplugs or reporting ringing after shift.
- Equipment added under a capital project with no EHS sign-off on noise.
- A survey date older than the most recent layout change.
- Roster gaps: people working in a posted hearing protection area who are not on the annual audiogram list.
Noise exposure is still widespread. Occupational Health & Safety reported in April 2026 that CDC puts the number of workers exposed to hazardous noise at over 22 million each year, and that most occupational hearing loss is preventable with noise controls and properly fitted protection.
What has to happen after the new survey?
Repeat monitoring is the start of a chain. Employees newly at or above 85 dBA must be notified of the results under 1910.95(e), and affected employees or their representatives get the chance to observe monitoring under 1910.95(f). New program members need a baseline audiogram within 6 months of first exposure, or within 1 year if a mobile test van is used, with protectors worn in the interim beyond 6 months. They also need the annual training in 1910.95(k) and access to protectors. Monitoring instruments must be calibrated, and the survey must integrate sound levels from 80 to 130 dB under 1910.95(d)(2). The rules on who moves in and out of the program are covered in our post on enrollment and removal criteria.
How Foster handles this
Foster runs the audiometric side of the program: on-site hearing testing with a mobile unit, and audiometer calibration for employers who test in-house. When we test a site each year, we look at the roster against the departments that show threshold shifts. If one area is producing several STS results while the rest of the plant is stable, we tell the safety contact that the exposure data for that area may be stale and a new survey is worth scheduling. We also ask about new hires and transfers into loud areas before the visit, because a person added to the roster after a remonitoring needs a baseline inside the 6 month or 1 year window, and that deadline is easier to hit when the testing day is already on the calendar. If your noise data is older than your floor layout, request a quote for testing and we will build the visit around the updated roster.
Frequently asked questions
Is there a fixed interval for repeating noise monitoring under OSHA?
No. 1910.95(d)(3) ties repeat monitoring to changes that raise exposure. OSHA's Appendix G notes many employers remonitor every year or two by choice.
Does a quieter machine require a new survey?
Not under (d)(3), which only covers changes that increase exposure. New data is still useful if you want to document that employees have dropped below the action level.
Does changing earplug brands trigger remonitoring?
The trigger is a noise increase, but a lower-rated protector can fail paragraph (j) at existing levels. Check the new protector's attenuation against your current exposure data.
Who can perform the noise survey?
The standard does not name a credential for noise monitoring. It requires calibrated instruments, an appropriate sampling strategy, and personal sampling where workers move or noise varies widely.
Sources
Get a price for your program
Foster runs the entire 1910.95 program on-site: audiograms, STS handling, audiologist review, notifications, and records. Headcount, shifts, and ZIP get you a price in two minutes.
Related guides
No-Show Policy for Mobile Hearing Testing: Keeping Every Employee on the Annual Audiogram
How to write a no-show policy for mobile hearing testing days so missed appointments get rescheduled, documented, and closed before the OSHA annual deadline.
6 min read Read →Hearing ConservationWhen an Audiogram Looks Wrong: Validity Checks Before You Call It a Shift
How to tell a bad audiogram from a real threshold shift: transducer fit, ambient noise, test-retest at 1000 Hz, and OSHA's 30-day retest rule.
5 min read Read →Hearing ConservationCombining Hearing and Breath Alcohol Testing on One Mobile Visit
Can a mobile testing visit cover annual audiograms and DOT breath alcohol testing in the same stop? What to schedule, staff, and document to keep both compliant.
6 min read Read →