Noise Criterion Rating Explained: The Complete Guide to NC Ratings for HVAC and Mechanical Noise

A noise criterion rating, usually written as an NC rating, is a single number that describes how much background noise from HVAC and mechanical equipment is allowed in a room. It is based on measurements across eight separate frequency bands, not one blended reading. A room's final noise criterion rating equals the highest band that touches or crosses a given NC curve. It is never an average.

That is the short answer. Everything below explains where the noise criterion system came from, how to read the chart, how it differs from RC and dBA ratings, what target to aim for in different room types, and how professionals actually bring a noisy room back under control.

Quick Facts About Noise Criterion Ratings

Before going deeper, here is the noise criterion system in five points:

1. It was created in 1957 by acoustician Leo Beranek to address a specific problem: HVAC noise in office buildings was interfering with speech, and nobody had a reliable way to measure why.

2. It covers eight octave bands, from 63 Hz up to 8000 Hz.

3. It is rated by worst case, not average. One loud band sets the entire room's rating.

4. Lower numbers mean quieter rooms. A recording studio might target NC-15. A factory floor can run as high as NC-60.

5. It is used across almost every building type, from concert halls and hospitals to classrooms, offices, and apartments.

Why Noise Criterion Ratings Actually Matter (Beyond the Spec Sheet)

Most articles on this topic treat the noise criterion rating as a paperwork requirement: hit the number, pass the inspection, move on. In practice, it affects real outcomes for the people using the building.

Here is what the research actually shows:

1. Noise is now the single biggest reported cause of workplace inefficiency. A global survey of more than 11,000 office workers found that employees ranked noise as the top factor behind reduced productivity, ahead of lack of privacy and poor temperature control.

2. Noisy offices drive turnover. A separate study of executives and employees found that people working in the noisiest environments were more likely to say they would leave their job within six months, and only a small fraction of executives had actually equipped their offices with any noise-mitigating features.

3. Sleep and health guidelines are built around these same numbers. The World Health Organization's community noise guidance recommends indoor bedroom noise stay under roughly 30 dBA at night, a threshold that maps closely to the NC-25 to NC-30 range used in bedroom and hotel-room design.

4. Speech privacy in offices, classrooms, and healthcare settings depends on it. A noise criterion rating that is too high does not just feel unpleasant. It actively degrades speech intelligibility, which matters in a boardroom, a classroom, or a patient consultation room.

In short: a noise criterion rating is not a bureaucratic checkbox. It is a proxy for occupant comfort, retention, health, and legal compliance, all rolled into one chart.

A Short History of the Noise Criterion System

Understanding where NC ratings came from helps explain why the system works the way it does, and why newer systems like RC and RNC exist alongside it.

1. 1957 - Noise Criterion (NC) curves. Leo Beranek publishes the original NC curves, built from survey data on what office workers found acceptable, tied closely to speech interference.

2. 1971 - Preferred Noise Criterion (PNC) curves. Beranek, along with Blazier and Figwer, tightens the original curves, lowering allowable levels at both the low and high end of the frequency range.

3. 1981 - Room Criterion (RC) rating. Warren Blazier develops RC for ASHRAE specifically to fix a blind spot in the NC system: low-frequency rumble from newer, more efficient HVAC equipment.

4. 1988 - Balanced Noise Criterion (NCB) curves. Beranek revises the system again, extending sensitivity down to very low frequencies.

5. 2008 - RNC (Room Noise Criterion) added to ANSI/ASA S12.2. This version is built specifically for HVAC noise that surges or fluctuates rather than staying perfectly steady.

Each new system exists because the previous one had a specific, documented weak spot. That matters, because it means the "right" rating system to use depends on what kind of noise problem you are actually dealing with, not just habit or convenience.

The Noise Criterion Octave Band Chart

This is the core reference chart. It shows the maximum allowable sound pressure level, in decibels, for each of the eight standard octave bands, at each noise criterion rating from NC-15 up to NC-60.

NC Rating	63 Hz	125 Hz	250 Hz	500 Hz	1000 Hz	2000 Hz	4000 Hz	8000 Hz
NC-15	47	36	29	22	17	14	12	11
NC-20	51	40	33	26	22	19	17	16
NC-25	54	44	37	31	27	24	22	21
NC-30	57	48	41	35	31	29	28	27
NC-35	60	52	45	40	36	34	33	32
NC-40	64	56	50	45	41	39	38	37
NC-45	67	60	54	49	46	44	43	42
NC-50	71	64	58	54	51	49	48	47
NC-55	74	67	62	58	56	54	53	52
NC-60	77	71	67	63	61	59	58	57

How to read this chart in three steps:

1. Measure the sound pressure level in each of the eight octave bands using a calibrated sound level meter.

2. Compare each measured value against the chart above, band by band.

3. The room's noise criterion rating equals the highest curve touched or exceeded by any single band. It is not the average of all eight readings.

This last point trips up more people than anything else on this page. A ductwork run can look fine on paper if someone averages the readings, then fail badly once a single band, usually a low-frequency one, gets checked against its own curve.

A useful detail buried in this chart: the low-frequency bands at 63 Hz and 125 Hz allow much higher decibel readings than the high-frequency bands at every single rating level. That is intentional. Human hearing is naturally less sensitive to bass frequencies at moderate volumes, so a reading of 45 dB at 63 Hz is barely noticeable, while the same 45 dB reading at 4000 Hz would sound like a persistent, grating whine.

Noise Criterion vs. dBA, RC, RNC, and NR Ratings: What's the Real Difference

This is the section most competing articles skip entirely, and it is often the reason a "quiet" HVAC system still generates complaints.

dB (decibel) is a raw unit of sound pressure with no frequency weighting at all.

dBA is a decibel reading passed through a filter that mimics human hearing sensitivity, discounting very low and very high frequencies.

NC (Noise Criterion) is derived from a full eight-band spectrum rather than one number, and was purpose-built to track speech interference.

RC (Room Criterion) extends the measured range down to 16 Hz and adds a diagnostic layer that NC simply does not have.

RNC (Room Noise Criterion) is a refinement of RC built specifically for HVAC noise that fluctuates or surges, rather than staying perfectly steady.

NR (Noise Rating) is the ISO-based system used more commonly outside North America, particularly in the UK and Europe, and is not directly interchangeable with NC even when the numbers look similar.

Here is the comparison in table form:

Metric	Frequency range	Best used for	Main limitation
dBA	Single weighted number	Quick checks, some code compliance	No spectral detail
Noise Criterion (NC)	63 Hz - 8000 Hz	General HVAC specification, speech interference	Misses sub-63 Hz rumble, no diagnostic descriptor
Room Criterion (RC)	16 Hz - 4000 Hz	Diagnosing rumble or hiss in modern systems	More involved to calculate
RNC	16 Hz - 4000 Hz, surge-adjusted	Fluctuating or surging HVAC noise	Less commonly used outside consulting work
Noise Rating (NR)	31.5 Hz - 8000 Hz	UK and European building services	Not a direct substitute for NC

Why this gap matters in the real world:

The original NC system was never designed to evaluate frequencies below 63 Hz, which happens to be exactly where most rumble from modern, energy-efficient HVAC equipment shows up. A room can technically pass its noise criterion target and still have an audible, complaint-worthy rumble that the NC chart is structurally unable to detect.

This is exactly why Warren Blazier built the RC system. An RC rating comes with a letter attached to the number, describing the character of the noise, not just its volume:

• Neutral (N): No particular frequency stands out. This is the target for a well-designed system.

• Rumble (R): Flagged when octave bands below 500 Hz run more than 5 dB above the RC curve.

• Hiss (H): Flagged when bands above 500 Hz run more than 3 dB above the RC curve.

Practical takeaway: if you are writing a basic mechanical spec for an office, school, or apartment building, NC is still the standard and it lines up with how most equipment sound-power data is published. If you are troubleshooting a specific complaint, such as "there's a low hum in the boardroom" or "the vents hiss whenever the system kicks on," RC or RNC will actually tell you something NC cannot.

A quick mental shortcut: as a rough rule of thumb, the equivalent dBA reading for a typical, well-balanced HVAC spectrum runs about 7 to 10 dB higher than its NC number. It is not an exact conversion, but it is close enough to sanity-check a spec sheet at a glance.

Recommended Noise Criterion Levels by Space Type

Different rooms need different noise criterion targets, and the right number depends on what the space is actually used for. Here is a breakdown, organized from quietest to loudest, with the standards behind each range where they exist.

1. Concert halls and recording studios: NC 15-20. Critical listening environments where even a faint hiss competes directly with the material being performed or recorded.

2. Theaters: NC 20-25. Speech intelligibility is the entire purpose of the room, and a low noise floor protects even the quietest dialogue.

3. Bedrooms and luxury hotel rooms: NC 25-30. This range tracks closely with sleep science. The World Health Organization recommends indoor bedroom noise stay under roughly 30 dBA at night for good sleep quality. Go quieter than NC-25 in a hotel and guests rarely notice the extra spend. Go above NC-30 and light sleepers notice immediately.

4. Executive offices and lecture halls: NC 25-30. Confidential conversations and unamplified speech both need a genuinely low noise floor to function properly.

5. Classrooms and private offices: NC 30-35. This lines up directly with school acoustics standards. ANSI/ASA S12.60 caps unoccupied classroom background noise at 35 dBA and recommends the teacher's voice run at least 15 dB above that background level at every seat for reliable comprehension. HVAC noise eats directly into that margin.

6. Hospitals and patient care areas: NC 25-35. Healthcare facility guidelines generally call for this range in patient rooms and exam spaces, where both rest and confidential conversation matter.

7. Open-plan offices and multifamily living rooms: NC 35-40. Active, transitional spaces where some ambient noise is tolerable, and can even help with speech privacy between desks or units.

8. Lobbies and restaurants: NC 40-45. Higher ambient noise is expected here, and total silence in a restaurant or hotel lobby often feels wrong rather than luxurious.

9. Retail and mechanical or equipment rooms: NC 45-55. Acceptable for spaces built around equipment rather than conversation or focus.

10. Factories: NC 40-65. Industrial limits driven more by hearing-conservation regulation than comfort.

A general pattern worth remembering: spaces built around sleep, focus, or confidential speech sit at the bottom of this scale. Spaces built around activity, equipment, or ambient energy sit at the top.

Common Sources of a High Noise Criterion Rating

Before jumping to fixes, it helps to know what typically drives the rating up in the first place:

1. Undersized or overworked fans and blowers, running at or near full capacity to compensate for restrictive ductwork.

2. Excessive duct velocity, especially at elbows, branch takeoffs, and diffusers.

3. Missing or degraded vibration isolation, allowing structure-borne rumble to travel through the slab or ductwork.

4. No duct silencers, or silencers that were value-engineered out during construction.

5. Rigid duct connections directly to equipment, transmitting mechanical vibration straight into the room.

6. Poor duct layout, with too many turns or restrictions close to the diffuser.

7. Aging equipment, where worn bearings or fan blades introduce noise the original design never accounted for.

How to Control and Lower a Noise Criterion Rating

Professionals generally work through this in three distinct phases. Skipping the third one is the single most common reason a "quiet" design ends up generating noise complaints after occupancy.

Phase 1: Design

1. Oversize equipment rather than sizing it to the wire. A fan running below its top capacity moves the same volume of air more slowly, and slower typically means quieter.

2. Specify duct silencers at the air-handler discharge. Packed or pack-less silencers can reduce noise by 8 to 15 points, depending on the model and placement.

3. Control duct velocity. As a general guideline, keeping main trunk ducts under roughly 1,500 fpm and diffusers under 500 fpm helps limit velocity-driven hiss in NC-30 target spaces. Duct velocities around 800 fpm or lower generally produce no audible noise on their own.

4. Install vibration isolation on all rotating equipment. Spring or neoprene isolators stop structure-borne rumble before it ever reaches the ductwork or the building slab.

5. Plan duct routing to avoid tight turns close to occupied spaces, since turbulence generated near a diffuser has little distance to dissipate before reaching the room.

Phase 2: Installation

This is where a well-designed system quietly goes wrong. Common failure points include:

1. Isolators skipped or improperly torqued because the install crew is behind schedule.

2. Duct runs resized in the field without recalculating velocity against the new dimensions.

3. Flex duct kinked or compressed during installation, adding unplanned turbulence.

4. Silencers installed backward or in the wrong sequence relative to the fan.

None of these show up on a drawing. They only show up once the system is running.

Phase 3: Verification (the step almost everyone skips)

1. Field-test the room after commissioning, with the HVAC system running under normal load.

2. Measure actual octave-band levels using a calibrated sound level meter, not just an overall dBA reading.

3. Compare the measured spectrum against the original design target, band by band, not as an average.

4. If a band exceeds target, identify the specific fix, usually one of three things: an added or upsized silencer, a replaced isolator, or a resized duct run.

5. Document the results so future renovations or complaints have a baseline to compare against.

A design-stage noise criterion target is a prediction. Field verification is proof. Skipping this step is how buildings end up with post-occupancy noise complaints that cost far more to fix after the fact than they would have during construction.

Room-side treatment options

Beyond the mechanical system itself, the room can also be tuned:

• Adding sound-absorbing surfaces to the occupied space can lower perceived noise by roughly 3 to 5 points.

• Sound masking, calibrated to around 45-48 dBA, is a reasonable fallback when equipment noise fluctuates in ways that are difficult to fully eliminate at the source.

If a building is already occupied and a full mechanical redesign is not realistic, this kind of room-side acoustic treatment is often the fastest, least disruptive way to bring a space back within its target range. This is also generally where an experienced acoustic and soundproofing contractor earns their fee: diagnosing whether the real fix belongs in the mechanical system or in the room itself, since guessing wrong wastes both time and budget.

A Worked Example: Reading a Real Spectrum

Example 1 - Small conference room. A single overhead diffuser feeds off a branch duct. Measured octave-band readings come back at 52, 47, 40, 33, 28, 24, 20, and 18 dB across the 63 Hz to 8000 Hz bands.

Checking each band against the chart:

1. The 63 Hz reading of 52 dB sits close to the NC-25 curve.

2. The 500 Hz reading of 33 dB clears NC-30 comfortably.

3. Every band from 250 Hz upward clears NC-25 with room to spare.

4. The 63 Hz band is the one driving the final rating, landing the room at roughly NC-25 to NC-30, even though the room "sounds" mid-range quiet to a casual listener.

That is the exact mechanism that catches people off guard. One low-frequency band, not the loudest number on the page, sets the room's actual rating.

Example 2 - Open-plan office. Measured readings come back at 58, 50, 44, 38, 34, 30, 27, and 24 dB across the same eight bands. Checked against the chart, this spectrum lands around NC-35 to NC-40, which is exactly the target range for this type of space. No corrective action needed here. This is what a correctly designed system looks like on paper.

Signs Your Building Has a Noise Criterion Problem

Not every noise complaint requires a full acoustic survey to diagnose. Watch for these warning signs:

1. Recurring complaints about "the hum" or "the buzz" in a specific room, especially one near a rooftop unit or mechanical closet.

2. A consistent low rumble that seems to intensify when the HVAC system cycles on.

3. A high-pitched hiss near diffusers or return grilles, particularly after a duct resize or renovation.

4. Conference calls where people on the other end regularly ask you to repeat yourself.

5. Staff in a specific area consistently choosing to work elsewhere in the building.

6. Noise that was not present when the building was newly commissioned but has grown worse over time, often pointing to worn bearings, degraded isolators, or a failing damper.

If more than one of these sounds familiar, it is usually worth getting actual octave-band measurements taken rather than guessing at a fix.

What It Typically Costs to Fix a High Noise Criterion Rating

Costs vary widely by building type and the size of the problem, but the general pattern holds across most commercial projects:

1. Room-side acoustic treatment (added absorption, panel installation, sound masking) is usually the least expensive fix and is often enough for moderate overages.

2. Retrofitting a duct silencer costs more than room treatment but less than replacing equipment, and is common when the mechanical system itself is the source.

3. Vibration isolation replacement is a relatively low-cost, high-impact fix when structure-borne rumble is the culprit rather than airborne noise.

4. Full duct resizing or equipment replacement is the most expensive path and is generally reserved for systems that were undersized from the start.

An experienced acoustic contractor will typically walk a space, take field measurements, and recommend the least invasive fix first, since the cheapest and fastest solution is often room-side treatment rather than a mechanical overhaul.

Common Noise Criterion Mistakes to Avoid

1. Averaging the octave bands instead of applying the worst-case rule. This single error accounts for a large share of noise criterion specs that fail once tested in the field.

2. Trusting a single dBA reading as proof a room is quiet. A room can post a perfectly reasonable dBA number and still have an obvious rumble or hiss that dBA, by design, smooths over.

3. Confusing NC with STC. Noise criterion describes noise generated inside a room by mechanical systems. Sound Transmission Class describes how well a wall or floor blocks sound from passing through it. A room can have an excellent STC-rated wall and still fail its NC target because the real noise source is the diffuser in the ceiling, not the neighbor next door.

4. Treating the design number as the finished product. A specification is a prediction. Only a field test after commissioning confirms it actually holds.

5. Ignoring low-frequency bands because the room "sounds fine." Rumble in the 63 Hz and 125 Hz range is easy to miss casually and is exactly what drives most real-world NC failures.

6. Value-engineering out silencers or isolators late in construction without recalculating the resulting noise criterion impact.

Noise Criterion Glossary

Octave band: A range of frequencies used to break down a noise spectrum into measurable segments, running from 63 Hz to 8000 Hz in the standard NC system.

Tangency method: The standard procedure for determining an NC rating, where the room's rating equals the highest curve touched or exceeded by any single octave band.

Speech Interference Level (SIL): The average sound pressure level across the 500, 1000, and 2000 Hz bands, used as an early basis for the original NC curves.

Rumble: Excess low-frequency noise, typically below 500 Hz, that produces a felt or heard hum even when overall dBA looks acceptable.

Hiss: Excess high-frequency noise, typically above 500 Hz, often caused by velocity-driven turbulence at diffusers or duct fittings.

Vibration isolation: Spring or neoprene mounts placed under rotating equipment to prevent structure-borne noise from transmitting into the building structure.

Sound masking: A deliberately introduced, steady background sound used to reduce the perceived intrusiveness of fluctuating noise sources.

Working With a Local Acoustic and Soundproofing Contractor

Reading an octave-band chart is one thing. Actually, measuring a real room, correctly diagnosing whether the problem is mechanical or architectural, and specifying the right fix is a different skill set entirely, and it is usually not something worth learning on the fly for a one-time project.

For commercial buildings across North Texas, the team at De-Walls Acoustic Specialties handles exactly this kind of diagnostic and treatment work, from tracking down a rumble in a mechanical room to full acoustic panel installation in offices, restaurants, and multifamily buildings. If the space in question is a restaurant fighting dining-room echo on top of kitchen noise, a restaurant noise control approach usually differs from what a warehouse or a home theater need, which is why a proper site walk matters more than a generic spec sheet.

A few situations were bringing in a soundproofing and acoustics specialist tends to pay for itself quickly:

1. Warehouse and industrial spaces where equipment noise carries into adjacent offices or loading areas, often addressed through targeted warehouse noise control treatment rather than a full mechanical redesign.

2. Open-plan and executive offices across the Dallas-Fort Worth area struggling to hit comfortable speech-privacy levels, where office acoustic treatment can often solve the problem without touching the HVAC system at all.

3. Home theaters and media rooms where isolation from the rest of the house matters as much as the sound quality inside the room, which is its own specialty distinct from commercial noise criterion work.

Frequently Asked Questions

What is a good NC rating for an office?

Most general offices target NC 30 to 40, with executive offices and boardrooms aiming lower, around NC 25 to 30, for better speech privacy.

What does NC 35 mean?

NC 35 means no single octave band in the room exceeds the NC-35 curve, a level generally considered comfortable for classrooms and standard private offices.

Is a lower NC rating always better?

Not necessarily. Lower ratings cost more to achieve and are only worth it where the activity demands it, such as sleeping, recording, or confidential conversation.

What's the difference between NC and RC rating?

NC covers 63 Hz to 8000 Hz and gives a single number. RC extends down to 16 Hz and adds a letter descriptor that flags rumble or hiss, giving a diagnosis alongside the number.

How is a noise criterion rating calculated?

Sound pressure levels are measured in each of the eight standard octave bands and plotted against the NC chart. The rating equals the highest curve touched or exceeded by any single band.

What NC rating is required for hospitals?

Patient care areas generally target NC 25 to 35, depending on the specific space and applicable facility guidelines.

Can soundproofing lower an NC rating?

Yes. Adding absorption to a room typically lowers perceived noise by a few points, and is often enough to resolve moderate overages without touching the mechanical system.

What causes a high NC rating in an existing building?

The most common causes are oversized duct velocity, missing or worn vibration isolation, and equipment that has degraded since original commissioning.

If you're dealing with a specific complaint right now, rather than a design spec, the direction usually splits based on what you're hearing. A low hum or rumble almost always points back to fan speed, duct velocity, or missing vibration isolation. A hiss points to diffuser selection or a duct silencer that's undersized for the airflow it's handling.