Measurements show burn-in makes no difference you can hear. If you want to run your headphones in anyway, this plays a safe, measured program and doubles as a soak test.
Probably not in any way you can hear. In the most thorough test, RTINGS played noise and sweeps through four new headphones for 120 hours. The measured changes were smaller than taking the headphones off and putting them back on, and distortion mostly moved by less than 0.1%, with no pattern over time. We could not find a single blind test in which listeners told a burned-in pair from a new one.
So why offer a tool? Some manufacturers still recommend burning in, and a few hours of sweeps and noise at a safe level do no harm. The same run is also a useful soak test: it is a good way to catch a rattle, a buzz or a channel that cuts out while the headphones can still be returned.
Five sources that measured something, from the most to the least direct. None found a change you could hear.
Tested. AKG Q701, Audeze LCD 2 Classic, Beyerdynamic DT 1770 Pro and Etymotic HF5 on a dummy head in a sealed box. Each hour: a 20 s sweep from 10 Hz to 22 kHz, 50 min of pink noise at 90 dB SPL, 10 min of silence. Measured every 20 hours.
Found. No evidence of burn-in. Frequency and phase changes were smaller than re-seating the headphones, and distortion moved mostly under 0.1% with no pattern. The Q701's level shift was traced to its pads softening. Music recorded at 0 and 120 hours sounded nearly identical.
Read the studyTested. An AKG Q701, measured new and at intervals up to 90 hours.
Found. Very small changes, clearly above measurement noise, mostly around the driver's 80 Hz resonance. His own conclusion: the data did not show that burn-in exists, and some of the change could come from the headphones shifting on the rig.
Read the studyTested. The RTINGS measurements.
Found. Ear pads compress over time, which moves the driver relative to the ear. That changes the measured response: a real change, but from the pads, not the driver.
Read the studyTested. Earphones from its first in-ear line, launched in 1997.
Found. Used pairs measure the same as new. Small balanced-armature drivers have very little that can change, though Shure allows that large over-ear drivers might.
Read the studyTested. Loudspeaker drivers.
Found. Only an extremely slight measured effect of break-in, and no audible effect.
Read the studyResearch reviewed in October 2026.
Plenty of people are sure their headphones changed. They are usually right that the sound changed, and wrong about why.
Hearing gets used to a new tonal balance quickly. Headphones that sounded harsh at first start to sound balanced, and it feels like the headphones changed.
Foam and leather compress with use, which moves the driver relative to your ear and changes the seal. RTINGS traced the one visible shift in its test to exactly this.
Headbands relax and you learn where the cups sit best. A better seal adds bass, which is easy to mistake for a driver loosening up.
Once you have been told a pair will open up after 50 hours, you listen for it, and you tend to hear what you expect. Only blind comparisons remove that, and none has shown burn-in.
Hours of sweeps and noise are a good stress test, whether or not burn-in exists. Listen during the first sweep: a rattle or buzz at one frequency, a channel that cuts out, or distortion that comes and goes is a fault worth returning the headphones for.
Take the headphones off and keep them off. Set the system volume to your usual listening level and leave it there. The run stops by itself at the length you chose.
Each hour starts with a 20-second sweep from 20 Hz to 20 kHz, then 50 minutes of pink noise, then 10 minutes of silence. This is the cycle RTINGS used in its 120-hour test, so the run matches the only published schedule that was measured.
A laptop or desktop on mains power is the most reliable: sound keeps playing in a background tab, and the run counts only time when sound actually played. On a phone, keep the screen on and the charger plugged in.