dB vs dBA: Weighting Explained

• dBZ (Z-weighted). Flat: all frequencies count equally. Best for comparing spectra and for calibration math.
• dBA. A-weighting roughly follows human hearing at moderate levels — quiet low frequencies count less. Default on this meter for everyday environments and decibel chart comparisons.
• dBC. Flatter than A; closer to how loud peaks feel. Useful near loud machinery or music.

Which weighting should I pick?

Everyday room and street checks: A. Loud bass-heavy events: compare A and C. Technical comparison between two recordings: Z, with identical response. Whichever you pick, calibrate with the same weighting you measure with.

Fast vs Slow

Fast (125 ms) follows peaks; Slow (1 s) smooths them. Neither changes the underlying estimate — only how quickly the display follows it. Leq (energy average) accumulates independently of the display smoothing.

Three rooms where the letter matters

Picture three evening checks with the same phone on the same table. A bedroom with a low refrigerator drone might read near 33 dBA yet 52 dBC, because A-weighting discounts the bass your ears mostly ignore at night. A kitchen chat at normal voice might read 62 dBA and 63 dBC, nearly tied, since speech energy sits where both curves stay close. A living room playing bass-heavy music might read 78 dBA but 88 dBC, a ten-point split caused almost entirely by low strings and kick drum. The spectrum view explains each split: when the dominant band sits low, C and Z climb above A; when speech dominates the middle, the three converge.

Try an A/B flip in 20 seconds

1. Set the device down, pick A-weighting with Slow response, and note the Leq across 20 quiet seconds.
2. Without moving anything, switch to C-weighting and note the fresh Leq across the next 20 seconds.
3. Glance at the spectrum: strong left-side bars predict a higher C figure; strong mid bars predict near-equal figures.

Fast at 125 ms chases door slams while Slow at 1 s steadies the line, yet Leq keeps its own energy average regardless of that smoothing. Average 30 to 60 seconds for everyday rooms and ignore single decimals, since model spreads of 5 to 10 dB dwarf them.

Keep weighting locked for the whole comparison

Choose once per question, then freeze it. Bedroom-to-bedroom night comparisons belong on A. Bass troubleshooting belongs on C with an occasional Z glance for the unweighted picture. Calibrate with the same weighting you later measure with, because the offset belongs to that curve. When you want a worked room routine, follow the background-noise test; when the low end puzzles you, read the frequency analyzer guide.

Open the online decibel meter → · How the meter works →