Speaker Break-In and Environmental Measurements

[Dumu]

Interesting thread. I've purchased quite alot of speakers along my life and have only had few cases where I've noticed any change. Two of the cases were extremely obvious.

1. A 12" car subwoofer I installed on a self-made cabinet that initially had very low output. When I took it to 15min test drive to town I had to stop twice to adjust the amplifier gain cause it started to be too loud. Going from full gain to around 1/4th of full. No other explanation than that the element had very stiff suspension in the beginning.

2. Pair of Wharfedale diamond 12.1's. I had them in my bedroom and was able to put them right against the wall where no ported speaker should be able to live. They had amazing thight bass until I one day tested them with 40hz sine wave. After that they became so boomy they had to be moved away and high passed, like you would expect with that kind of placement. I had never played them loud before, probably not even over 75dB. I also bought a pair of 12.3's that initially had few db high tweeter level (measured), but settled to be exactly the same as the 12.1's.

So in my experience the break-in is rare, but it exists in some speakers. I also believe it will happen in few minutes or less when enough movement is applied to the woofers.

 

[Salt]

Woofers get a gym for free over 24 hours at 20 Hz with sub max excursion just for quality control.
With some of them the amplification has to be attenuated within the first minutes to avoid damage as excursions rise, with some no changes over 24 hours.
It all depends...

 

[Dumu]

I would be very surprised to see budget speakers go through so thorough quality control. Wouldn't even surprise me if some chinese company would skip it all together. But I was also surprised to see the car subwoofer output improving so much, as no other sub had ever behaved like that.
I have to say it would be nice to know how rare it is to have some kind of break-in. For now I will feed some sine waves to every speaker I will audit, just to make sure

 

[Head_Unit]

Woofers get a gym for free over 24 hours at 20 Hz with sub max excursion just for quality control.

@Dumu is correct, no they don't, presuming you mean at the factory. Nobody has time for that except maybe some tiny brands. While some of the various factories I visited in Japan and Philippines and China were doing 100% QC at customer demand, it was a frequency and/or impedance and/or distortion sweep.

As for the engineer types quoted by @WillBrink well sure yes there absolutely is physical break-in however I'd love to see their DATA showing any significant change in the frequency response. I did once see data a woofer where the resonance dropped a lot. I consider that defective, that just should not happen. And if they tried to tell me it was "designed that way on purpose" GONG! sorry nope don't believe you.

 

[WillBrink]

Andrew Jones covering speaker break in:

Part #1, covering mostly base drivers and general details:

Part #2, tweeters, etc:

 

[Soniclife]

Andrew Jones covering speaker break in:

Part #1, covering mostly base drivers and general details:

Part #2, tweeters, etc:

What's the summary of what he says that's relevant?

 

[Salt]

Wavecor Datasheets contain before burn-in and after burn-in data.
Just saying...

 

[Head_Unit]

What's the summary of what he says that's relevant?

First is that you do get some quick initial break in of the spider (I'd add that maybe a tiny bit of surround, depending on the type). Now, he mentions a driver going from 50 Hz down to 40 Hz, that seems crazy extreme to me and I never measured that much but hey maybe he made a design like that. He mentions that in a sealed box the box usually dominates which then dilutes any change-that was my experience, the measured response changing very little as he says. If he mentions changes in a ported box I missed it.

He says "there's not much evidence that if you work it for 1000 hours that it will have permanently shifted it to an even lower value [than initial working from brand new]." Now he does go on to qualify that as I'll phrase it that if you are beating the living hell out of something day after day then things will continue to change which agrees with testing I had done.

Also mentioned is temperature: if you get speakers delivered in winter and play them right away well they are physically cold and that can sound different. (I also believe humidity could be a factor).

In short I would say just play your speakers and don't worry about it. If they don't sound good before the end of the return window then bye-bye send them on back, accept no bullsh!te from the seller.

 

[MAB]

Wavecor Datasheets contain before burn-in and after burn-in data.
Just saying...

Yes they do, but only for a few of their drivers. The changes they do publish are greater than any driver I have measured. So we are clear, I measure real changes due to Break-in.

Most of the changes listed in their spec sheets are 10% in f(s), with an equivalent change in Q. The offsetting changes in f(s) and Q cancel each other out, consistent with Vance Dickason's comments. Here is a Wavecor WF182BD12 modeled with their published before and after break-in data.

Wow, that's really large change, even larger than the HiVi B3n I measured.

There are indeed differences in output. Even with nearly 15-20% change in f(s) and Qms, still barely moves the needle when put in a box.

I've used two different Wavecor drivers, but they were units without break-In data in their spec sheet. I did measure before and after break-in, they only changed a few percent, which will result in an even smaller difference in response. Sorry, I didn't save the data since it was before I realized people think big changes happen due to break-in. I did notice the spec sheet was over 5% different than measurements, a much larger difference than the changes I measured due to break-in. I think it is highly unlikely their driver actually changes 15% after break-in. Until I see measurements I doubt. Even if they really change that much I can't hear sub 1dB changes in bass.

 

[Salt]

It depends on the construction/materials of surrounds and suspensions, if a break/burn in has any effect over time... or not.
But that's easy to control today.

 

[MAB]

It depends on the construction/materials of surrounds and suspensions, if a break/burn in has any effect over time... or not.
But that's easy to control today.

Yeah I tested that in post#1, multiple drivers made from different materials. Do you have any examples of drivers that have large changes after break-in or over time?

 

[Salt]

If so, I will post it here (when I've the time after retirement). Will take some time ...

 

[MAB]

If so, I will post it here (when I've the time after retirement). Will take some time ...

OK. Until that possibility, if a seller or manufacturer tells you that a pair of speakers you don't like "just need a few hundred hours to break-in until they sound good", you know the only thing that is getting broken into is your wallet. The new speakers will not dramatically change sound as salespeople often claim. The claim is a scam to keep people from returning bad-sounding speakers.

If drivers did audibly break in, wouldn't it show up routinely in testing and in the spins Amir and Erin produce? Amir showed the affect of testing a speaker in a cold room (the change was really tiny), but was picked over at ASR with the usual flair. As I show in this thread, temperature is a larger factor than break-in for driver performance. It does leave open the possibility that a new driver could be barely audible as it breaks in. But that would go away in a few moments since break-in doesn't take very long, and after that the speaker changes negligibly until it wears out.

To date, I have never seen data to support the claim of large and potentially audible changes in a driver after Break-In. My experience goes back to the '80s on this, and I admit perhaps older drivers than I know were made from materials or manufacturing that had serious changes during infancy. Or maybe there is a corner-case driver, in which case it would be interesting to see. Aside from that corner case, I am not anticipating much progress in the "speakers really do have audible break-in" side of audio.

edit: typo

 

[Hayabusa]

If so, I will post it here (when I've the time after retirement). Will take some time ...

I take that with grain of salt

 

[stoneeh]

If it hasn't been linked yet, here's a study of the effect over days, not hours: https://www.der-akustische-untergrund.de/app/download/13853327834/Lautsprecherchassis_einspielen.pdf

Any effect of break-in in the hour timespan disappears over the long term.

 

[MAB]

If it hasn't been linked yet, here's a study of the effect over days, not hours: https://www.der-akustische-untergrund.de/app/download/13853327834/Lautsprecherchassis_einspielen.pdf

Any effect of break-in in the hour timespan disappears over the long term.

Thanks for the link. This looks to be a careful test.
The changes they find are really quite tiny. This is consistent with other tests that are carefully done, specifically avoiding measuring a hot woofer just after stress.

Extremely interesting that the small break-in effect nearly completely disappears after 10 days of rest. Andrew Jones has commented on suspension hysteresis (I think he called it 'creep'), this appears to be a measurement of the effect on several drivers.