Treat sound and vibration as one measurement discipline with two languages. Master logarithmic level arithmetic, weighting and band analysis on the sound side; parameter selection (displacement, velocity, acceleration) and path analysis on the vibration side; then bind both to documented field procedure and acceptance criteria.
Level arithmetic: why background noise must be subtracted logarithmically
Decibels combine and separate logarithmically, not linearly. Any measured level that includes background noise overstates the source, so the standard decision is: measure background separately, then apply logarithmic subtraction before judging the result.
Scenario: you measure 78 dB(A) at a duct-occupied office wall with a rooftop unit running, then 72 dB(A) with the unit off. The tempting move is to subtract linearly and report 6 dB(A) from the unit. That is wrong because decibels are powers, not pressures. The correct calculation is 10·log10(10^7.8 − 10^7.2), which works out to roughly 77 dB(A). The difference between 6 and 77 dB(A) is the difference between a trivial finding and a reportable source level.
Why it matters: acceptance criteria for occupied spaces are stated as source contribution, not as whatever the meter happened to read. If background is within a few decibels of the combined reading, the subtraction becomes numerically unstable and the honest decision is to flag the measurement as limited rather than force a number. Practicing this decision — subtract, but know when to stop trusting the subtraction — is a core NEBB S&V study target.
Drill it with a calculator and a table of 10^(L/10) values until the pattern is automatic: two equal sources add 3 dB, a source 10 dB above background is essentially unaffected, and a source less than 3 dB above background cannot be extracted reliably. These anchor points let you sanity-check any field result in seconds.
- Anchor 1: two identical levels add to 3 dB above one of them
- Anchor 2: a source 10 dB above background needs almost no correction
- Anchor 3: if the difference is under about 3 dB, report the limitation instead of the number
Sound pressure, sound power, and intensity: reporting the right quantity
Sound pressure is what a microphone reads at a point; sound power is a property of the source; intensity combines pressure and particle velocity. Reports and criteria name a specific one, so identifying which is requested precedes any measurement.
Pressure measurements depend on distance, room acoustics, and directivity, which is why a single pressure reading near a fan says little about the fan itself. Sound power is derived — through standardized methods — from pressure measurements at defined distances or in controlled environments, and it is the quantity equipment specifications usually reference. Intensity measurement adds directional information, letting you locate a dominant panel or casing leak rather than merely confirm noise exists.
The study task is mapping each quantity to its use: pressure for occupancy criteria such as NC or dBA limits; power for comparing equipment or predicting levels in other rooms; intensity for source location. A useful exercise is to read any acoustical clause and label every stated value with its quantity and unit. If you cannot label it, you cannot select the right microphone position, distance rule, or correction, and the rest of the procedure is guesswork.
Octave bands and weighting: reading the spectrum behind a single number
A-weighting approximates human hearing and collapses a spectrum into one number; octave and one-third-octave bands reveal where energy sits. Criteria in occupied spaces are often set per band, so a passing dBA can still fail a band limit.
Weighting de-emphasizes low frequencies that ears hear poorly, which is why a fan rumble can look mild in dB(A) while dominating a 63 Hz or 125 Hz octave band. NC-style criteria are evaluated band by band, so the exam-relevant skill is reading a spectrum shape: a broad broadband rise suggests aerodynamic noise, while a single dominant band points to a tone, rotation frequency, or resonance.
Practice by sketching plausible spectra: a well-balanced system falls smoothly across bands; a problematic one spikes in one or two. Then connect the shape to a hypothesis — low-frequency dominance suggests structure-borne transmission or slow rotating machinery; a high-frequency hiss suggests duct velocities or diffusers. This interpretive habit, from number to spectrum to cause, is what separates measurement from documentation. Work spectra by hand at first so the weighting curve and band centers become familiar rather than abstract.
Vibration parameters: choosing displacement, velocity, or acceleration
The three parameters emphasize different frequency ranges of the same motion. For rotating equipment in typical HVAC speed ranges, velocity (RMS) is the usual severity metric; displacement suits very low frequencies and acceleration suits high frequencies.
Scenario: a supply fan turns at about 900 rpm (15 Hz) and the reading at a bearing housing shows elevated acceleration while velocity is moderate. The plausible mistake is judging severity from acceleration alone and recommending an immediate shutdown. The better decision is to reason about the frequency content: acceleration amplifies high-frequency features, so on a slow fan the meaningful severity comparison is RMS velocity — checked across the bearing positions and trended — while displacement helps characterize the very low frequency motion itself.
Why it matters: reporting the wrong parameter can either exaggerate or hide a genuine problem, and the recommendation attached to the number changes accordingly. The decision habit is a two-step check — identify the dominant frequency range, then select the parameter that is sensitive there — before any comparison to severity references. Keep in mind that simplified classroom scenarios behave more cleanly than real machines, so the training goal is a defensible selection rationale, not a memorized verdict.
Use the table below as a selection aid, then test yourself by assigning parameters to descriptions of different machines and fault types until the choice feels automatic.
| Parameter | Sensitive range | Typical use | Common misapplication |
|---|---|---|---|
| Displacement | Low frequencies | Characterizing slow, large-amplitude motion | Using it as the sole severity metric on mid-speed equipment |
| Velocity (RMS) | Broad mid-range typical of rotating machinery | Severity comparison and trending across bearing points | Ignoring that very low or very high frequency content distorts the picture |
| Acceleration | High frequencies | Detecting impacts, bearing distress, gearmesh effects | Reading it on a slow fan and calling the result 'severe' without context |
| dB levels / weightings (sound side) | Audio band per weighting curve | Occupied-space criteria and spectra | Treating a single dBA value as a complete diagnosis |
Airborne versus structure-borne paths and isolator transmissibility
The same source reaches a receiver through air and through the building structure. Distinguishing the path changes the remedy, and transmissibility — the ratio of transmitted to input vibration — is the concept that links isolator behavior to the measurement.
A trace: occupants below a fan room report rumble. Floor vibration measurements show low-frequency dominance; the airborne level in the occupied space is unremarkable. The evidence points to structure-borne transmission through mounts, curbs, or piping connections rather than duct-borne noise. Had the spectrum shown strong airborne low-frequency energy instead, the investigation would move to duct attenuation and breakout. Same source, different path, different fix.
Transmissibility explains why an isolator can worsen matters: near its natural frequency the ratio exceeds one and vibration is amplified rather than isolated. The practical lesson is that a measurement near a machine is not enough — readings on both sides of the isolation point show whether the isolator is doing its job. In study scenarios, always ask which path the data supports, and resist naming a remedy before the path is established. This habit of evidence-first path diagnosis is the applied core of the vibration domain.
Field procedure and documentation: making a measurement defensible
A defensible measurement chain covers instrument condition and calibration, defined measurement points, background readings, stated weighting and bandwidth, and a report that lets a reader reproduce the measurement. Documentation is part of the method, not an afterthought.
NEBB disciplines are built on published procedures and specifications, and the sound and vibration work follows that pattern: the procedure defines where and how you measure, and the report shows you followed it. Concretely, that means noting the calibration state of the microphone or accelerometer before and after the session, recording sensor orientation and mounting, capturing background conditions, and labeling every value with its quantity, weighting, and band.
The study application is to practice writing, not just measuring: take any classroom exercise and produce the report line for it — quantity, unit, weighting, bandwidth, location, conditions. A reader who can reconstruct your measurement from the report is the standard to aim for. In preparation, review NEBB's published specifications and discipline descriptions on its own site to understand the framework your procedures sit inside; administrative details such as eligibility and exam logistics belong to the issuer, so treat nebb.org as the single reference for those rather than memorizing secondhand versions.
- Label every number: quantity, unit, weighting, band, location
- Record calibration and sensor mounting for each session
- Capture background readings before judging any source
- Write the report line for every measurement, not a bare figure
A scenario drill, self-check rubric, and preparation sequence
Combine the domains in one exercise: measure a steady household source with background on and off, correct the level logarithmically, sketch the octave-band shape, and write the report line. Score yourself against the rubric below as a learning milestone.
Exercise: pick a steady source such as a vacuum cleaner or bathroom fan. Measure combined level and background level at a fixed position, compute the source contribution by logarithmic subtraction, and predict from the anchor points (about 3 dB for equal levels, negligible correction at 10 dB difference) before calculating. If you have access to an analyzer app, sketch the octave bands; otherwise draw plausible band shapes by hand and annotate which band would dominate. Finish by writing a two-line report entry: one for the sound measurement, one for a hypothetical velocity reading at the machine.
Rubric — award yourself a point for each: (1) the corrected source level matches a dB calculator within about 1 dB; (2) background conditions and measurement position are documented; (3) every value carries quantity, unit, weighting, and band labels; (4) the spectrum shape is annotated with a plausible cause hypothesis; (5) a vibration reading line names its parameter with a one-sentence selection rationale. A realistic sequence: weeks one to two on level arithmetic and weighting; weeks three to four on spectra and parameter selection with the table; weeks five to six on path analysis and isolation; final weeks on timed written scenarios and full report lines. Strong self-check scores indicate learning progress only — they are not predictions of any exam outcome.
- Readiness check 1: you can subtract background levels correctly without a reference sheet
- Readiness check 2: given a spectrum, you can state a plausible cause and the next measurement
- Readiness check 3: given any machine description, you can justify a vibration parameter choice in one sentence
- Readiness check 4: every practice measurement you produce has a complete, reproducible report line
References and further reading
Use these references to explore the concepts and check the latest information from the relevant organizations.