Fast and Slow are time weightings. Both average the sound continuously, but Fast uses a time constant of 125 milliseconds and Slow uses 1 second. The sound is identical either way; only the display behavior changes. Fast follows every bang and gap, Slow smooths them into a steadier number.
Pick Slow for steady noise and for anything you’ll compare against a published limit, and Fast when you need to see how much a level swings. OSHA’s permissible exposure table is written in dBA measured on Slow response.
What time weighting actually does #
A meter doesn’t show the sound pressure at one instant. It keeps a running average of the sound energy, weighted so that recent sound counts more than older sound, decaying exponentially. The time constant sets how quickly older sound fades out of the calculation.
- Fast (125 ms) reacts in roughly an eighth of a second, so short events show up as visible peaks.
- Slow (1 s) takes about a second to respond, so brief peaks are flattened into the average around them.
That’s why the same room can read 68 on Fast and 66 on Slow at the same moment. Neither is wrong. They are answers to slightly different questions.
Fast vs Slow at a glance #
| Fast (125 ms) | Slow (1 s) | |
|---|---|---|
| Reacts in | About 1/8 second | About 1 second |
| Display behavior | Jumps with every event | Steady, easy to read |
| Best for | Impulsive or changing noise, finding peaks | Steady noise, readings you’ll write down |
| Maximum reading | Higher, catches short peaks | Lower, peaks get averaged away |
| Used by | General measurement, event spotting | OSHA Table G-16, most published limits |
| Typical use | A workshop where tools start and stop | An HVAC hum, traffic, a plant room |
Which response should you use? #
Use Slow when:
- The noise is fairly constant: machinery, traffic, ventilation, a busy room.
- You’re comparing against OSHA’s permissible noise exposures, which specify sound level in dBA on slow response.
- You want a number you can read off the screen and record without averaging by eye.
- You’re checking a level against a local noise ordinance that says “slow.”
Use Fast when:
- The noise comes in bursts: hammering, a nail gun, dogs, door slams, a cafeteria.
- You want to know how high the peaks go rather than what the typical level is.
- You’re diagnosing a specific event inside a longer recording, such as the moment the neighbor’s bass drops.
Use whatever the rule says when: a standard, ordinance or employer procedure names one. Noise ordinances vary from city to city, and many specify the weighting, the response, the measurement location and the duration. Check the wording of the ordinance that applies to you before measuring for a complaint.
Fast, Slow, Impulse and Peak are four different things #
Two more terms turn up on meters and cause confusion.
Impulse is a legacy time weighting with a very short rise time of about 35 milliseconds and a slow decay, designed for sharp sounds. Modern sound level meter standards treat it as obsolete and it’s rarely required, though some meters still offer it.
Peak isn’t a time weighting at all. It’s the highest instantaneous sound pressure in the measurement, with no averaging whatsoever, usually reported C-weighted. Peak values are always higher than any Fast or Slow reading of the same event, sometimes by 20 dB or more. This is the number impulse-noise rules use: OSHA states that exposure to impulsive or impact noise should not exceed a 140 dB peak sound pressure level.
So a single hammer blow might produce a 138 dB peak, a 105 dB Fast maximum and a 95 dB Slow maximum. All three describe the same hammer.
Does Fast or Slow read higher? #
For maximum values, Fast almost always reads higher, because it catches short events that Slow averages down. For the current level on a steady source, the two agree closely.
For Leq, the energy average over a measurement period, time weighting makes no meaningful difference at all. Leq integrates the actual sound energy over the whole period, so the smoothing applied to the display doesn’t change the result. If you measure a machine for 60 seconds, the Leq is the same on Fast or Slow, even though the number on screen behaved differently while you watched. What is Leq? explains what that average represents.
This is worth remembering when you record a measurement. Maximum values need their time weighting attached to mean anything, which is why professional reports write LAFmax (A-weighted, Fast, maximum) or LASmax (Slow). An Leq needs only the weighting and the duration.
Setting the response in a phone app #
Sound Meter dB puts both options in Settings, labeled Fast (125 ms) and Slow (1 s), with Fast as the default. The same screen has the frequency weighting choice, A-weighted dBA or flat, and a calibration offset. Whichever combination you use is saved with each measurement session, alongside the Leq, the minimum, the maximum and the time spent over 85 dB, so a session you look at next month still says how it was measured. Sessions of ten seconds or more are saved automatically on the device and can be exported as CSV.
The meter updates ten times a second with a 30-second history graph, which makes the difference between the two settings easy to see: switch to Fast and the trace grows spikes, switch to Slow and it settles. As with any phone meter, the readings are estimates rather than certified measurements. Our guide to how accurate phone sound meters are covers the gap, and calibrating a sound meter app covers closing part of it.
Frequently asked questions #
Which response should I use for a noise complaint? #
Check the ordinance first, since many specify the setting, the measurement position and the averaging period. When the rule is silent, Slow with A-weighting and a stated measurement duration is the conventional choice, and it’s harder to argue with than a peak caught on Fast. See how to measure neighbor noise.
Does Fast or Slow change the Leq? #
No. Leq is the energy average over the measurement period, so it’s independent of display smoothing. Only the instantaneous reading and the maximum change.
What does LAFmax mean? #
It’s the highest level recorded during a measurement, A-weighted (the A), on Fast response (the F). LASmax is the same thing on Slow. Because Fast catches short peaks, LAFmax is normally higher than LASmax for the same event.
Why does my reading jump around so much? #
Either the noise genuinely varies or you’re on Fast response. Switch to Slow for a steadier number, and read the Leq rather than trying to average a moving display by eye. Handling noise and wind across the microphone also cause false jumps.
Which response do professional standards require? #
It depends on the standard. OSHA’s Table G-16 is written for slow response, while many environmental measurements use Fast for maximum values and Leq for the average. When you report a level, always state the weighting, the response and the duration.