How to Calculate Noise Criteria for Your Room

By Sarah Okonkwo · February 16, 2026

How to Calculate Noise Criteria for Your Room

1) What you’ll learn (and why it matters)

Noise Criteria (NC) is a standardized way to describe background noise in a room using octave-band sound pressure levels. Calculating an NC rating helps you answer a practical question with a single number: “Is this room quiet enough for the work I’m doing?”

If you record vocals, mix at low levels, edit dialogue, or run critical listening sessions, the room’s noise floor can mask detail, skew decisions, and force you to monitor louder than you should. NC gives you an objective target (and a repeatable method) so you can diagnose noise problems, compare “before/after” improvements, and communicate clearly with HVAC contractors, building managers, or clients.

2) Prerequisites / setup requirements

Measurement tool: Ideally a Type 1 or Type 2 sound level meter with octave-band capability. Acceptable alternative: a measurement mic + audio interface + software that can display 1/1-octave band SPL (e.g., REW, Smaart, SpectraFoo, etc.).
Calibration: Acoustic calibrator at 94 dB SPL @ 1 kHz is preferred. If you can’t calibrate, you can still compare relative changes, but the absolute NC number may be off.
Room state: Room set up the way you actually use it (doors closed, typical furniture, equipment powered as normal). You’ll likely run two measurements: HVAC on and HVAC off.
Band data you must capture: Octave bands from 63 Hz to 8 kHz (at minimum). If your tool includes 31.5 Hz, capture it too; low-frequency rumble often shows up there.
Time weighting: Use Slow (1 s averaging) if you’re using an SPL meter; in software, use stable averaging (e.g., 10–30 s average) and ensure levels settle.
Quiet window: Choose a time with minimal external noise (traffic peaks and trash pickup will corrupt results). Late evening or early morning often works.

3) Step-by-step: calculating NC in a practical way

Step 1: Define your target NC based on the room’s job

Action: Pick an NC target before measuring so you know what “good” looks like.

Why: Without a target, you can’t decide whether a 5 dB improvement matters. Targets also guide which noise sources are worth fighting (HVAC vs. street noise vs. computer fans).

Common NC targets (typical practice):
- Voiceover / dialogue edit rooms: NC 15–20
- Critical music control rooms: NC 20 (excellent), NC 25 (workable)
- Project studio / songwriting rooms: NC 25–30
- Podcast rooms with close mics: NC 25–30 (lower is always better)
Pitfalls: Picking a target that doesn’t match reality. If your room has a wall facing a busy road and no structural isolation, NC 15 may be unrealistic without construction changes. Start with what’s achievable: HVAC improvements and local noise sources can often get you into the NC 20–30 zone.
Step 2: Calibrate your measurement chain

Action: Calibrate the meter or measurement system so your SPL values are meaningful.

How:
- If using a sound level meter: run its internal calibration routine, or use an external calibrator if supported.
- If using a mic + interface: attach the calibrator to the mic, set the calibrator to 94 dB SPL at 1 kHz, and adjust the software calibration so the reading is 94.0 dB (or as close as possible).
Specific technique: Set your interface preamp so you have plenty of headroom; you’re not chasing loud signals here. Aim for a stable digital level (for example, around -20 dBFS average during calibration) so you’re not near the noise floor of the interface.

Pitfalls: Forgetting that a phone app is rarely accurate below about 100 Hz or at very low SPL. If you must use a phone app, treat results as approximate and focus on identifying dominant bands (e.g., a clear 125 Hz hump from HVAC).

Troubleshooting: If the reading jumps around by more than 1–2 dB during calibration, check the calibrator seal on the mic and disable any automatic gain control (AGC) in software.
Step 3: Choose mic position(s) that represent real listening

Action: Place the mic at typical ear height and location(s).

Why: NC is meant to characterize what occupants experience. A corner measurement exaggerates low-frequency buildup; a measurement 10 cm from a computer fan exaggerates that noise source.

Recommended positions:
- Mix position: Mic at the engineer’s head location, ear height (~1.2 m seated).
- Recording position: Where the vocalist/instrument typically is (especially for voiceover).
Technique: Keep the mic at least 1 m from large reflective surfaces when possible. If the room is tiny, do your best and take multiple measurements.

Pitfalls: Holding the mic in your hand (body reflections and handling noise). Use a stand and step away.
Step 4: Measure octave-band SPL with HVAC ON

Action: Capture a stable, averaged set of octave-band levels from 63 Hz–8 kHz.

Why: HVAC is often the dominant contributor in studios. It also has a recognizable spectral “shape” (rumble + mid-band air noise) that NC curves are designed to evaluate.

Settings to use:
- Frequency bands: 63, 125, 250, 500, 1k, 2k, 4k, 8k (add 31.5 if available)
- Weighting: Z (flat) if available. If not, use C rather than A. (NC uses band levels, not A-weighted overall SPL.)
- Averaging: 30 seconds average is a solid starting point. If the HVAC cycles, capture at least 2–3 minutes or measure during the “steady” part of the cycle.
Technique: Do two takes: one during a typical steady state, and one when the system first kicks on. Some blowers surge and create short bursts that are noticeable in real work.

Pitfalls: Measuring while you’re hearing intermittent external noise (a truck passing). If it happens, note the time and repeat the measurement.

Troubleshooting: If the lowest bands (31.5/63 Hz) are unstable, check for vibration coupling: mic stand touching a desk, floor vibrations from a mini-fridge, or mechanical rumble from a rack.
Step 5: Measure octave-band SPL with HVAC OFF (baseline)

Action: Repeat the same measurement with HVAC off, same mic position.

Why: This separates building/environment noise (traffic, neighbors, electrical hum) from HVAC noise. It also tells you whether improving HVAC will meaningfully lower NC, or whether the room is dominated by exterior leakage.

Specific technique: Wait 60–120 seconds after turning HVAC off so airflow and duct noise fully decay. Then capture a 30–60 second average.

Pitfalls: Forgetting that turning HVAC off may change other noises (some systems power down relays, some rooms get quieter but also reveal faint electrical tones).

Troubleshooting: If HVAC-off looks almost identical to HVAC-on across all bands, the noise may not be HVAC. Check computers, hard drives, powered speakers, dimmers, refrigerators, or outside noise.
Step 6: Determine the NC rating by comparing your band levels to NC curves

Action: Find the lowest NC curve that is not exceeded by your measured octave-band levels (or in common practice, the curve that your spectrum just touches at its highest “over” point). This resulting number is your room’s NC rating.

Why: NC is based on a family of curves. Your measured spectrum will sit above some curves and below others. The controlling factor is the band (or bands) that sit highest relative to the curves.

How to do it (practical methods):
- Method A (recommended): Use software or an online NC calculator that accepts octave-band SPL values and returns an NC number. Enter your bands for HVAC-on and HVAC-off and store both results.
- Method B (manual): Use an NC curve chart. For each octave band, look up the curve value at that band. The NC rating is approximately the highest curve number that your measured levels meet or exceed in any band (the “worst” band sets NC).
Specific example: Suppose your HVAC-on measurement at the mix position is:
- 63 Hz: 38 dB SPL
- 125 Hz: 33 dB SPL
- 250 Hz: 28 dB SPL
- 500 Hz: 24 dB SPL
- 1 kHz: 22 dB SPL
- 2 kHz: 21 dB SPL
- 4 kHz: 19 dB SPL
- 8 kHz: 17 dB SPL
On an NC chart, you’d typically find the low-frequency bands (63/125 Hz) are controlling. If the 63 Hz point sits near the NC 30-ish region while the mid/high bands sit near NC 20–25, your room might still rate around NC 30 because the worst band defines the rating.

Pitfalls: Using A-weighted overall SPL (e.g., “my room is 28 dBA”) as a substitute for NC. Two rooms can measure 28 dBA and have very different low-frequency rumble, which affects recording and comfort differently.

Troubleshooting: If your calculated NC seems “too high” compared to how the room feels, check that you used octave bands (not 1/3-octave) and that you entered SPL values (not dBFS). Also confirm you didn’t apply A-weighting to the band data.
Step 7: Identify the controlling noise band and link it to a real cause

Action: Look at the band(s) that push your NC rating upward and diagnose the likely source.

Why: NC is a number, but fixing noise requires knowing whether you’re dealing with duct rumble, fan hiss, structural vibration, or outside leakage.

Quick mapping (common in studios):
- 31.5–63 Hz elevated: mechanical rumble, compressor vibration, building vibration, subways/traffic. Often structure-borne.
- 125–250 Hz elevated: duct resonance, large fans, HVAC blower tones, thin-wall transmission.
- 500 Hz–2 kHz elevated: air velocity noise at vents, whistling grilles, turbulent airflow, computer fans near mic position.
- 4–8 kHz elevated: hiss from air leaks, small fans, electronic noise, some lighting drivers.
Pitfalls: Trying to “EQ the room” acoustically with absorption when the problem is structure-borne rumble. Thin foam does nothing for 63 Hz.
Step 8: Verify improvements with a repeatable before/after measurement

Action: After making one change (only one, if possible), repeat the same measurement in the same mic position and compare.

Why: Noise control is iterative. If you change three things at once, you won’t know what worked. Repeatability is how you learn quickly and avoid wasted effort.

Specific techniques that commonly change NC:
- Reduce HVAC fan speed or set the system to a “low” continuous mode instead of cycling high.
- Add lined ducting or a duct silencer; reduce air velocity at the grille.
- Move noisy computers out of the room; replace small high-RPM fans with larger low-RPM fans.
- Isolate vibrating devices (mini-fridge, rack UPS) with dense rubber pads.
Pitfalls: Measuring too soon after changes. If you adjusted HVAC, let it run 10 minutes to stabilize.

4) Before/after comparison (expected results)

In a typical project studio, an initial HVAC-on measurement might land around NC 30–35 due to 63–125 Hz rumble and 500 Hz–2 kHz air noise. After addressing the dominant issues, it’s realistic to see:

HVAC airflow adjustments: a 3–6 dB reduction in 500 Hz–2 kHz bands, often dropping NC by 2–5 points.
Computer noise mitigation: a 5–10 dB reduction in 1–8 kHz bands near the workstation, sometimes dropping NC by 3–8 points if those bands were controlling.
Rumble control (harder): low-frequency improvements are often smaller (2–5 dB) unless structural changes are made, but even 3 dB at 63 Hz can change perceived comfort and can reduce the NC rating if 63 Hz was the controlling band.

For real-world context: if you’re recording a quiet vocal with a condenser mic at 20–30 cm, dropping room noise from NC 35 to NC 25 can reduce audible HVAC between phrases dramatically and can allow gentler noise reduction in post (or none at all).

5) Pro tips for taking it further

Measure multiple positions and average thoughtfully: For control rooms, measure at the mix position and 0.5 m left/right. If one position is much worse, that’s often a local noise source (PC exhaust, vent aimed at the chair).
Capture “NC over time”: If your HVAC cycles, log measurements every 30 seconds for 10 minutes. A room that’s NC 20 half the time and NC 35 when the system kicks can still be a problem for recording.
Look for tones, not just broadband noise: NC is curve-based and doesn’t explicitly flag tonal components. If you hear a whine at 250 Hz or 1 kHz, investigate even if NC seems acceptable. Tonal noise is more annoying and more audible in recordings than the same SPL of broadband noise.
Use your band data to guide fixes: High 1–4 kHz bands often respond to airflow changes and quieter fans. High 63–125 Hz often requires vibration isolation, mass, and mechanical decoupling—plan accordingly.
Document everything: Note date/time, HVAC mode, mic position, and any equipment on/off. This turns one measurement into a repeatable test method you can use for every room you work in.

6) Wrap-up (practice to build speed and confidence)

Calculating Noise Criteria isn’t about chasing a perfect number—it’s about building a reliable picture of your room’s noise floor, identifying what’s actually controlling it, and verifying changes with repeatable measurements. Do the HVAC-on and HVAC-off measurements, calculate NC for both, and then pick one fix aimed at the controlling octave band. Repeat the measurement and watch how your NC number (and your recordings) respond. After a few rounds, you’ll be able to walk into a room, predict the likely NC problem spots, and confirm them with data instead of guesswork.