Best picks
Best Sound Level Meters: When a Phone App Is Enough
Choose a sound level meter for spot checks or logging, see when the NIOSH iOS app is enough, and understand weighting, calibration and Class 1 versus Class 2.
By thisMETER EditorialUpdated November 2, 20267 min read
For occasional noise checks, start with NIOSH's free iOS Sound Level Meter app; our dedicated pick for spot checks is the REED R8050. Choose the REED R8080 when you need timestamped readings stored inside the instrument and USB downloads. These recommendations come from official specifications and manuals; thisMETER has no instrument readings to publish here.
thisMETER is reader-supported: when you buy through our links, we may earn a commission at no extra cost to you. As an Amazon Associate, thisMETER earns from qualifying purchases.
Quick picks
| Choice | Our recommendation | Published range and weighting | Recording | Standard claim |
|---|---|---|---|---|
| NIOSH SLM app | Try first for occasional screening on iOS | A, C and Z; microphone conditions matter | Saved reports, LAeq and TWA | Type 2 under IEC 61672 Part 3 with a calibrated external microphone |
| REED R8050 | Best dedicated meter for spot checks and simple audio checks | 30–130 dB; A and C | Data hold and MAX; no time-series logging documented | IEC 61672-1 Class 2, per REED |
| REED R8080 | Best for storing a history | 30–130 dB; A and C | 64,000 timestamped datapoints; USB to Windows software | IEC 61672-1 Class 2, per REED |
Sources: NIOSH app specifications, R8050 specifications and R8080 specifications. Awards are our judgments about these documented functions.
When a phone app is enough
Start with the app if your question is exploratory: which location or operating condition deserves a closer look? NIOSH's app is free and available only on iOS. It offers A, C and Z weighting, LAeq, TWA, maximum and peak levels, and report sharing, according to its official documentation.
NIOSH states accuracy within ±2 dBA, evaluated in a reverberant chamber at its acoustics laboratory. That is a published laboratory result, not a promise about every recording situation. The same page says its Type 2 claim under IEC 61672 Part 3 requires a calibrated external microphone.
There is a further distinction: Part 3 concerns periodic tests. The NIOSH help guide describes evaluation of the phone and external microphone together and says work toward Parts 1 and 2 compliance continues. We would use the app for screening and deciding whether to commission a professional survey, rather than treat the download alone as proof that a measurement meets a specified standard.
NIOSH says the variety of Android hardware prevents it from currently verifying an Android version's accuracy. That does not establish that every Android sound app is inaccurate; it means the NIOSH results cannot simply be transferred to one. For an Android owner who wants an instrument with a documented class specification, our recommendation is a dedicated meter.
Best dedicated meter for spot checks: REED R8050
The R8050 is our straightforward choice when you want to read the display yourself. REED specifies 30–130 dB, ±1.4 dB accuracy, 0.1 dB resolution, A/C frequency weighting and Fast/Slow time weighting. Its specification table lists IEC 61672-1 Class 2.
It has a backlit display, data hold, MAX hold and a tripod mount. REED includes a windscreen, battery and carrying case. For simple audio checks requiring C weighting, it already offers that setting; buy the logging model only if you need its additional recording functions.
The limitation is what happens after the reading: the documented functions hold a displayed value or a maximum, without a stored, timestamped series. Choose the R8080 if a history matters. The R8050's manual also describes calibration with an external acoustic calibrator; the included-items list does not include one.
Documentation conflict: REED's specification table starts the high range at 60 dB, while the manual's operating instructions say 65. Both give 130 dB as its upper endpoint. Confirm that transition with REED if your procedure depends on it, and follow the instrument's range indication.
Best for logging: REED R8080
The R8080 adds internal memory for 64,000 datapoints, time and date stamps, and selectable recording intervals from 1 to 60 seconds. Its 30–130 dB coverage, A/C weighting and stated ±1.4 dB accuracy make stored history the main reason to upgrade from the R8050.
REED includes a USB cable, PC software, batteries, windscreen and hard carrying case. The software supports live measurements, downloading recorded data, graphing and export to Excel. REED lists Windows compatibility, including Windows 11, on its software support page; it does not list macOS or Linux support for this package. The software guide lists versions only through Windows 10, so the documents differ. Check access to a supported computer before buying it for exports.
Know what is being logged. REED describes a CAL Leq feature in the PC software, while its instrument manual documents sampled sound levels. We would not assume that this establishes standalone integrating LAeq measurement. If your task specifies continuous integrating measurements, get confirmation of that capability before choosing a logger.
The recording interval and Fast/Slow weighting are separate settings in the manual. Set the clock and interval before recording so the resulting history answers your question. For a quick display reading with no files to keep, our preference remains the simpler R8050.
Buying guide: the specifications that change the choice
Class 1 versus Class 2
IEC 61672-1 defines both performance classes; Class 2 acceptance limits are greater than or equal to Class 1 limits. The standard also distinguishes time-weighting, integrating-averaging and integrating instruments. Class and measurement function therefore deserve separate checks.
Both REED picks list Class 2 in their specifications. If a brief requires Class 1 or an integrating instrument, choose equipment explicitly documented for that brief. A maker's headline accuracy number is only part of the information: NIOSH's help guide describes standards tests covering frequency response, directionality, environmental conditions and the complete microphone/instrument system.
Weighting, response and the displayed metric
REED recommends A weighting for general noise levels. Its manuals say a much higher C-weighted reading indicates substantial low-frequency noise. Both picks offer A and C, making either a candidate for audio procedures that call for those settings. For more detail on reporting units, see How we measure.
Fast and Slow are response settings: REED specifies 125ms and 1 second, respectively. These are separate from an average over the whole session. NIOSH distinguishes LAeq, maximum, C-weighted peak and TWA in its help guide, while the R8050's MAX mode retains the highest detected reading. Do not substitute MAX for an average or assume it provides a dedicated peak measurement.
Range floor and calibration
Both picks' published ranges start at 30 dB. If your question requires quantifying sound below that, choose an instrument documented for a lower range. A generic decibel level chart cannot extend a meter's specified range; use the range and metric required for your actual task.
REED's manuals say the instruments are calibrated during manufacture and that a calibration certificate can be requested separately. Both describe adjustment using an external calibrator. Factory calibration, a documented certificate and a field check are separate buying considerations; establish which your measurement brief needs.
How to measure decibels consistently
REED's manuals advise keeping the microphone away from your body, pointing it toward the source and using a tripod for longer monitoring. They also warn that blowing air can cause errors and say the windscreen reduces that effect. The R8080 manual instructs users to select a suitable range and resolve over/under indications; NIOSH recommends calibration before and after measurements.
Choose the metric
Record the frequency weighting and response setting; decide whether you need a current level, maximum or average.
Check calibration
Follow the maker's calibration procedure; NIOSH recommends checks before and after measurements.
Position the microphone
Keep it away from your body, point it toward the source and use the windscreen where air movement affects it.
Check the range
Resolve over/under indications before relying on the displayed value.
For comparisons, keep the source condition, microphone position and settings consistent, and write those details down with each result. For logging, check the clock and recording interval too.
- Partly: NIOSH app: start with iOS screening; its Type 2 claim requires a calibrated external microphone.
- Yes: R8050: a dedicated A/C-weighted display with data hold and MAX.
- Yes: R8080: internal timestamped logging and USB downloads using Windows software.
Workplace figures are references, not a spot-check verdict
OSHA's noise overview gives 85 dBA as an 8-hour TWA as the hearing-conservation trigger for general industry and requires engineering or administrative controls above 90 dBA as an 8-hour TWA. Its standards page gives NIOSH's recommended exposure limit as 85 dBA-TWA with a 3 dBA exchange rate.
NIOSH distinguishes a sound level at a particular time from exposure averaged over time. A brief reading cannot establish an entire shift's exposure. These workplace references do not provide a household pass/fail judgment; have a qualified occupational specialist handle a workplace assessment.
How we choose
We compared official product pages, manuals, software documentation, NIOSH guidance and the IEC standard abstract, checked November 2, 2026. Range, accuracy and class claims remain the makers' specifications; our awards reflect fit to a task, without independent instrument testing.
We also considered the planned UNI-T UT353BT Bluetooth option. Its official page describes iENV app transfers but currently says the product is out of stock and unavailable. We have left it off the buying shortlist pending confirmed availability.
Our wider method is on How we choose. For other practical measuring tools, see our plug-in watt meter guide and gifts for electricians and engineers.
Verdict
Try the NIOSH iOS app first for occasional screening, keeping its microphone conditions in mind. Buy the REED R8050 for a dedicated A/C-weighted spot-check meter, or the R8080 for internal logging and Windows exports. Choose by the measurement and record you need, then check class, calibration and range against that task.
Frequently asked questions
Can a phone app measure decibels accurately?
NIOSH publishes accuracy within ±2 dBA for its iOS Sound Level Meter app, based on laboratory evaluation. Its Type 2 claim under IEC 61672 Part 3 requires a calibrated external microphone. Start there for occasional screening, but do not assume the same performance for another app or microphone. NIOSH says it cannot currently verify Android app accuracy across the different devices.
Do I need a Class 1 or Class 2 sound level meter?
IEC 61672-1 defines both classes, with acceptance limits for Class 2 greater than or equal to those for Class 1. For informal spot checks, our dedicated picks carry REED's Class 2 specification. If a measurement brief specifies Class 1, choose an instrument documented for that requirement. A single advertised accuracy figure does not describe the whole standard.
What is the difference between dBA and dBC?
The letters identify the frequency weighting used for the reading. REED recommends A weighting for general noise measurements and says a substantially higher C-weighted reading can indicate considerable low-frequency noise. Both dedicated picks offer A and C. Record which weighting you used, and compare readings with matching settings rather than treating the numbers as interchangeable.
Does a sound meter's MAX reading give the average noise level?
No. The R8050's MAX function retains the highest detected reading. NIOSH separately identifies equivalent continuous level, LAeq, maximum level, C-weighted peak level and time-weighted average, TWA. Decide which metric you need before buying. For stored readings, the R8080 adds internal logging, but its PC software's CAL Leq feature should not be assumed to establish standalone integrating-meter performance.