Dev.to WebDev πŸ›  Dev πŸ‘ 0 πŸ“– 7 min read

A Midrange Beep Misses the Rattle: Left-Right Tones, Frequency Sweeps, and Held Tones

A midrange beep can tell you that a speaker is not completely dead. It cannot tell you that the left channel is missing, that bass has collapsed, that a grille rattles at 140 Hz, or that a room cancellation punches a hol

A midrange beep can tell you that a speaker is not completely dead. It cannot tell you that the left channel is missing, that bass has collapsed, that a grille rattles at 140 Hz, or that a room cancellation punches a hole in the sweep. Music hides those problems because it plays many frequencies at once. A useful speaker check is a set of tones with a known frequency, a known channel, and a way to mark the moment something goes wrong β€” then a way to hold that frequency still.

FixNHandy publishes those checks as in-browser tools in its speaker set. The sounds are generated locally with the Web Audio API; the site states that nothing is streamed or recorded and that what you play is not uploaded. The site caps generator gain so the digital tone does not clip in the browser; that cap does not guarantee the speaker itself will be free of distortion. The three pages below are most useful when you treat them as a sequence: prove both channels, glide through the range, then hold the frequency that failed.

Left and right first, then bass and treble β€” not a single beep

FixNHandy’s speaker test starts with hard-panned clips. Left plays only on the left; Right plays only on the right. Both clips are 440 Hz sine tones, a midrange A4 that most small speakers can reproduce, which is exactly why they are not enough on their own. After the channel check, the page adds Bass at 80 Hz, Treble at 6 kHz, a sweep, and a Both or Stereo clip. The point of the extra steps is to catch a speaker that still beeps in the midrange and fails everywhere else.

You mark each clip Heard, Unclear, or Not heard. The PASS, WARNING, or FAIL card is built from those marks. The site cannot hear your room; it can only score what you report. That is an honest limit, and it is also why the setup matters. Unplug headphones and dongles, turn Bluetooth off if you intend to test the built-in speakers, and start around 50–70 percent. The tool already caps peak gain of the generated signal, which is not the same as preventing speaker distortion. If Safari is silent while ringtones still work, check Silent mode: the Ring/Silent switch, or, on models that have an Action button, the Action button or Control Center. Browser and iOS behavior can vary, so treat that as a troubleshooting check, not a universal rule. On a laptop, the browser follows the operating system’s selected output; a monitor or a headset can steal the test while the lid speakers look selected. Center the system balance before you decide one side is dead.

What the test can show is a silent, swapped, or dull channel, or a dropout at the bass or treble clips. What it cannot do is measure distortion in a laboratory sense, or see water inside a sealed module. A PASS on the loudspeaker also does not clear the earpiece. On many phones those are different drivers; muffled calls after a clean left/right result still need a call-side check. If both channels play but sound distant, the next question is blockage or moisture in the grille, not an automatic replacement. If one channel stays silent after the output and balance have been ruled out, and the same pattern does not follow a pair of headphones, the speaker or its connector is the suspect.

A sweep finds the frequency a single tone will skip

The frequency sweep tool glides a sine from one frequency to another so a problem that lives at one point can stand out. The full-range preset runs 20 Hz to 20 kHz in 20 seconds. A rattle hunt runs 40–400 Hz, the range where a cone moves most and most buzzes live. A hearing-range sweep runs 8–20 kHz over 30 seconds, intended for headphones in a quiet room, because most phone and laptop speakers roll off before many ears do. A custom mode lets you set the start, end, length, curve, and left, right, or both.

The default curve is logarithmic: 20 to 40 Hz takes as long as 10 to 20 kHz. That matches how pitch is heard. A linear sweep moves a constant number of hertz per second, so it races through the bass and spends most of its time in the treble; the page reserves that option for cases that need a constant hertz-per-second rate. When you hear a rattle, a buzz, or the tone disappearing, you tap Mark. The page subtracts a typical reaction time of about a third of a second so the mark lands closer to the frequency that actually caused the sound.

Volume has to be set in the right part of the sweep. The tone is played at one level throughout, but it sounds loudest around 2–5 kHz because hearing is most sensitive there, and quieter at the extremes because ears and speakers both roll off. Set the volume in that sensitive band, not on the bass, and keep it moderate. Stop if anything sounds harsh. Small bumps and dips are normal; no speaker or room is perfectly flat. Loud and quiet patches in the bass that move when you move are usually the room, not the driver.

The hearing-range mode is a rough check of an upper limit, not an audiology exam. The page notes that many adults stop hearing somewhere around 15–17 kHz, that the limit drifts with age and noise exposure, and that a large difference between ears is more meaningful than the number itself. Headphones vary, and some cut off before your ears do. If the limit is far below 12 kHz or the ears differ a lot, the page points to an audiologist. Never turn the volume up to chase a tone you cannot hear.

A rattle at a marked frequency is often mechanical. The follow-up is to hold that frequency, then press lightly on the grille, the cabinet, the case, and nearby objects until the buzz stops. If nothing you touch changes it, and it also appears at low volume, the speaker itself may be damaged β€” a different test from a loose screw.

Hold the frequency still once you have found it

FixNHandy’s tone generator is the page that keeps a single frequency playing. You can type any value from 20 Hz to 20 kHz, drag a logarithmic slider, step by 1 Hz, or jump by octaves. Changes glide while the tone plays, so you can sweep by hand without clicks. A sine wave is a single frequency with no harmonics, which is what you want for a speaker check or a rattle hunt. Square, sawtooth, and triangle waves add harmonics. At the same peak amplitude, square and sawtooth waves contain more harmonic energy than a sine, which is why the tool lowers their level automatically; how much of that energy a given device actually reproduces still depends on its bandwidth. Some of those harmonics fall in the 2–5 kHz range where ears fatigue quickly. Output can be both channels, left only, or right only. A share link stores the frequency and waveform in the page address; a 10-second WAV can be downloaded if you need the tone away from the browser.

The presets are practical bookmarks rather than a specification. 50 Hz and 60 Hz are the common mains-hum frequencies; 440 Hz is A4; 1 kHz is a familiar reference; 10 kHz, 15 kHz, and 17.4 kHz sit in the high-frequency band some adults no longer hear. Silence at the top is common, and phone and laptop speakers often stop well before 20 kHz. The page also says the tool will not repel pests; that is a statement about this generator’s 20 Hz to 20 kHz range, not evidence about pest-repeller devices. Whether a tone above 20 kHz can be produced at all depends on the audio sample rate and the hardware, not on a fixed browser limit. Do not raise the volume to hunt a tone that is simply outside the device, the headphones, or your hearing.

A pure tone can sound quieter than it is, which is why the page tells you to start low and raise the level slowly. High frequencies and complex waveforms are tiring even when they do not sound loud. Use these tools at a moderate level, keep a phone or an earpiece away from your ear, and stop if the sound distorts or becomes painful. A held tone is for finding a loose panel or confirming a dropout, not for long listening sessions.

Taken together, the three pages replace a single β€œtest tone” habit with a small protocol. Prove that each channel plays. Glide through the range and mark what rattles or disappears. Hold that frequency and see whether a hand on the cabinet, a change of position, or a lower level makes the problem move. The protocol will not give you a laboratory frequency-response curve, a distortion number, or a medical hearing result. It will keep you from declaring a speaker healthy because it can still play a midrange beep.

πŸ“° Read the original article on Dev.to WebDev

Originally published by Dev.to WebDev. Aggregated on AIWithGhost for educational purposes β€” full credit and traffic to the original publisher.