Burn-in testing - real or imaginary?

[GRBoomer]

Has there been any testing here to confirm if it takes 100s of hours of burn-in before an audio component "sounds" its best?
I'm a bit skeptical, especially for solid-state devices, but prove me wrong with data.

 

[RayDunzl]

https://www.audiosciencereview.com/...185288/&q=burn+in&c[title_only]=1&o=relevance

 

[AdamG247]

Has there been any testing here to confirm if it takes 100s of hours of burn-in before an audio component "sounds" its best?
I'm a bit skeptical, especially for solid-state devices, but prove me wrong with data.

Welcome Aboard @GRBoomer.

 

[MakeMineVinyl]

As a manufacturer, If our electronics 'changed' during 'burn-in', I would be extremely worried. The point is to make stable circuits which are not subject to the whims of how long they are burned-in to sound 'better' or 'different'.

 

[RayDunzl]

The point is to make stable circuits which are subject to the whims of how long they are burned-in to sound 'better' or 'different'.

aren't

 

[MakeMineVinyl]

aren't

I think you caught me mid-edit and quoted the 'defective' version......I probably should have let the post 'burn-in' before clicking the OK button.

 

[Katji]

Has there been any testing here to confirm if it takes 100s of hours of burn-in before an audio component "sounds" its best?

There is no need to test/confirm every audiophile consumer psychosis idea, it would be like checking that the planet is not "flat earth." Even the suggestion of testing is part of the psychoses. Otherwise we would just accept the authority of bona fide experts. - In this case, electronic engineers who are familiar with this aspect of the field.

 

[Purité Audio]

Yes hundreds of hours and equipment always sounds better.
Keith

 

[jtwrace]

Turn the shit on, listen. Done.

 

[muslhead]

Burn in for electronics is to remove those that normally fail early out of the entire pool thereby increasing the average lifespan of those left. It doesn't change their function or character (performance) but rather their increases the probability of a longer life through reduction/elimination of the typical bell curve, early fail tail.

 

[MakeMineVinyl]

Burn in for electronics is to remove those that normally fail early out of the entire pool thereby increasing the average lifespan of those left. It doesn't change their function or character (performance) but rather their increases the probability of a longer life through reduction/elimination of the typical bell curve, early fail tail.

Burning in for infant mortality is a different thing, and most manufacturers including us do this. Expecting the end customer to do this is insanity.

 

[abdo123]

I can swear when I replaced the crossover capacitors of some vintage speakers I have it sounded really really weird the first 10-20 minutes or so.

they were film/foil capacitors.

 

[Dialectic]

No one has ever produced any evidence other than sighted listening tests that burn-in alters the sound of any component or cable. The benefits of burn-in thus seem to be a result of the placebo effect.

We welcome the dissenting views of @analogplanet and the other contrarians here.

 

[Pdxwayne]

No one has ever produced any evidence other than sighted listening tests that burn-in alters the sound of any component or cable. The benefits of burn-in thus seem to be a result of the placebo effect.

We welcome the dissenting views of @analogplanet and the other contrarians here.

A while back, I asked about what measurements I could take to check for DAC burn in. The advice I got from experts here is to do a blind test.

The problem for such advice is this: if a blind test is taking long enough, the device would have burn in (if indeed there is such thing as burn in and how ever brief it is) and I would have failed the blind test.

If you don't mind to offer advice, what kind of measurements can one take to properly check for burn in?

 

[audio2design]

If I am selling a non-UL approved, $2,500, 1 amp square wave generator, or something that will take >30 days for someone to accept they have been swindled, then yes, burn in matters. Otherwise almost never. I have, back in the day, with laser vibrometry, measured a reduction in cone breakup out to 5-6 hours.

I think a poorly designed product without consideration of production tolerance could take a long time to warm up to optimum operating. I am thinking single ended amps that are highly conditional on their operating point.

 

[Dialectic]

A while back, I asked about what measurements I could take to check for DAC burn in. The advice I got from experts here is to do a blind test.

The problem for such advice is this: if a blind test is taking long enough, the device would have burn in (if indeed there is such thing as burn in and how ever brief it is) and I would have failed the blind test.

If you don't mind to offer advice, what kind of measurements can one take to properly check for burn in?

Either instrument-based measurement or double-blind comparison of a non-burned-in device X against a device X with 1,000 hours on it.

 

[Pdxwayne]

Either instrument-based measurement or double-blind comparison of a non-burned-in device X against a device X with 1,000 hours on it.

Good. Do you know of any thread here or anywhere else, that specifically did measurements or blind tests for burn in?

 

[VintageFlanker]

Has there been any testing here to confirm if it takes 100s of hours of burn-in before an audio component "sounds" its best?
I'm a bit skeptical, especially for solid-state devices, but prove me wrong with data.

I'm guessing your question was about electronics in particular, right?
Anyway, regarding headphones: https://www.rtings.com/headphones/learn/break-in

 

[DonH56]

I did, decades ago, but on speakers with a series of measurements and DBTs. No proof, no idea what happened to my box of notes from ca. 1984, lost or misplaced in some move long since then (they made it to CO from MO but disappeared in our last move, about 25 years ago now, along with several other boxes -- we finally decided the moving company must have misplaced a load but far too late to file a claim, especially as we don't know exactly what all is missing).

 

[MakeMineVinyl]

If you don't mind to offer advice, what kind of measurements can one take to properly check for burn in?

I personally haven't performed this, but I suspect one way would be to take component "A" (new, not burned in) and component "B" (well burned in) and sum their outputs out of polarity so that there is zero output with signal input. Let them play for as long as desired - weeks if need be - and see if they are still perfectly nulled to zero output. If you hear sound at the end of this test, the non-burned-in component has 'changed' its sound. This wouldn't necessarily 'prove' burn-in, but it would show that a change has taken place.