• WANTED: Happy members who like to discuss audio and other topics related to our interest. Desire to learn and share knowledge of science required. There are many reviews of audio hardware and expert members to help answer your questions. Click here to have your audio equipment measured for free!

Do Audio Speakers Break-in?

MarkWinston

Addicted to Fun and Learning
Joined
Jul 13, 2021
Messages
736
Likes
577
This is from (Amazon's preview of) Dr. Toole's book Sound Reproduction: The Acoustics and Psychoacoustics of Loudspeakers and Rooms.

He measured the frequency responses of the whole loudspeaker (not just driver) and conducted listening tests. There were measurable differences in the woofer compliance—no surprise here. Changes in the FR were minimal, and the listening tests showed no audible differences. You don't have to take his words for it, you are welcome to conduct your own tests.
Instead of coming up with other researches, why not debunk the one I posted the link to? I am very familiar with Toole's work and his views regarding this matter, thanks.
 

NTK

Major Contributor
Forum Donor
Joined
Aug 11, 2019
Messages
2,642
Likes
5,763
Location
US East
Instead of coming up with other researches, why not debunk the one I posted the link to? I am very familiar with Toole's work and his views regarding this matter, thanks.
Didn't Dr Toole say there were measurable changes in woofer compliance? They did not cause changes large enough to be audible.

What are Fs, Vas, and Qts? Where can I find their meaning/definition?
Wikipedia is our friend :)

 

Killingbeans

Major Contributor
Joined
Oct 23, 2018
Messages
4,077
Likes
7,511
Location
Bjerringbro, Denmark.
Do comment on this... manipulated/inaccurate/wrongly measured/false/forced data?


The same link is posted as #5 on the first page of this thread.

Danny Richie is known for being in the camp of "everything is audible!". He made an absolute ass of himself in a YouTube video where he tried to "prove" that audiophile cables make a difference, but did so in an involuntarily laughable manner.

But like they say, even a blind hen sometimes finds a grain of corn.

It's been mentioned many times in this thread that some bass drivers change T/S parameters a bit when being used for the first time.

The data is probably just fine. The big question is how audible the change is, and how much weight it carries when combined with the mental component of burn-in. (My guess: Not audible at all)
 
Last edited:

MarkWinston

Addicted to Fun and Learning
Joined
Jul 13, 2021
Messages
736
Likes
577
The same link is posted as #5 on the first page of this thread.

Danny Richie is known for being in the camp of "everything is audible!". He made an absolute ass of himself in a YouTube video where he tried to "prove" that audiophile cables make a difference, but did so in an involuntarily laughable manner.

But like they say, even a blind hen sometimes finds a grain of corn.

It's been mentioned many times in this thread that some bass drivers change T/S parameters a bit when being used for the first time.

The data is probably just fine. The big question is how audible the change is, and how much weight it carries when combined with the mental component of burn-in.
Danny's data show the T/S parameters keep changing overtime and its not a first few minutes thing which contradicts what AJ claimed.
 

blueone

Major Contributor
Forum Donor
Joined
May 11, 2019
Messages
1,164
Likes
1,478
Location
USA
Do comment on this... manipulated/inaccurate/wrongly measured/false/forced data?

I have more questions than answers. Like exactly what the set-up is for each measurement? I'm a big fan of Rod Elliott's white papers on various subjects, like this one:


And measuring individual drivers seems like no small endeavor.

Also, drivers are mechanical devices, so it seems logical that their performance will vary as environmental factors change and mechanical stress increases. The woofer tests are run at 40Hz at 50% of rated xmax. Given that criteria, I doubt any of my speakers I've ever owned have been broken in, including my 11 year-old Revels. So if the definition of break-in is 10-80 hours of stress none of us will likely ever do at home, is that a reasonable criteria for measuring change in these parameters?

Secondly, is measuring changes in individual drivers relevant, or are we only interested in the system-level effects? Your two links are driver-level, yet Amir's posting and the Toole reference are system-level. It might be that driver-level effects, while real, are mitigated by system-level considerations. I don't know.

So while I find the data from your links mildly interesting, the lack of real-world relevance (e.g. 50% of rated xmax testing) and only driver-level data not compelling enough, for me, to discount that speaker break-in is probably a system-level myth.
 

BDWoody

Chief Cat Herder
Moderator
Forum Donor
Joined
Jan 9, 2019
Messages
6,929
Likes
22,538
Location
Mid-Atlantic, USA. (Maryland)
Danny's data show the T/S parameters keep changing overtime and its not a first few minutes thing which contradicts what AJ claimed.

Audibility tests would be great.
 

BDWoody

Chief Cat Herder
Moderator
Forum Donor
Joined
Jan 9, 2019
Messages
6,929
Likes
22,538
Location
Mid-Atlantic, USA. (Maryland)
Debunk those data with facts or sit down and let others do it.

Easy killer...we try to have a little fun here and there.

You really do need to take the edge off.
 

Raindog123

Major Contributor
Joined
Oct 23, 2020
Messages
1,599
Likes
3,554
Location
Melbourne, FL, USA

Thanks.

@MarkWinston So, ok. Assuming the thermal hysteresis has been accounted for, this data shows the resonance frequency does go down by a couple of Hz (after a few tens of hours), and the resonance widens by a few percent… How does it affect the sound of the speaker, a forced oscillator (ie driven by external periodic force, largely above this resonance frequency)!?
 
Last edited:

Killingbeans

Major Contributor
Joined
Oct 23, 2018
Messages
4,077
Likes
7,511
Location
Bjerringbro, Denmark.
Danny's data show the T/S parameters keep changing overtime and its not a first few minutes thing which contradicts what AJ claimed.

Yeah, but they can't keep changing forever. Otherwise all bass drivers in well used speakers would have an Fs of 'DC' :D

Looking at the data, I'd say the worst case changes plateaus around the 30 hour mark. That's where the differences get completely obscured by measurement tolerances.

But again; is it audible? Probably not. But I'm open to tests showing the contrary.
 
Last edited:

MarkWinston

Addicted to Fun and Learning
Joined
Jul 13, 2021
Messages
736
Likes
577
I have more questions than answers. Like exactly what the set-up is for each measurement? I'm a big fan of Rod Elliott's white papers on various subjects, like this one:


And measuring individual drivers seems like no small endeavor.

Also, drivers are mechanical devices, so it seems logical that their performance will vary as environmental factors change and mechanical stress increases. The woofer tests are run at 40Hz at 50% of rated xmax. Given that criteria, I doubt any of my speakers I've ever owned have been broken in, including my 11 year-old Revels. So if the definition of break-in is 10-80 hours of stress none of us will likely ever do at home, is that a reasonable criteria for measuring change in these parameters?

Secondly, is measuring changes in individual drivers relevant, or are we only interested in the system-level effects? Your two links are driver-level, yet Amir's posting and the Toole reference are system-level. It might be that driver-level effects, while real, are mitigated by system-level considerations. I don't know.

So while I find the data from your links mildly interesting, the lack of real-world relevance
I have more questions than answers. Like exactly what the set-up is for each
Easy killer...we try to have a little fun here and there.

You really do need to take the edge off.

I see nothing to be made fun of, especially when its directed at me. I would like to see these data get debunked so this topic can actually move along while waiting for the next research. Till now the best answer to the data linked is that the whole research was made on a driver level and not system level making it most probably inaudible. Ill wait for better answers to these data.
 

MarkWinston

Addicted to Fun and Learning
Joined
Jul 13, 2021
Messages
736
Likes
577
Yeah, but they can't keep changing forever. Otherwise all bass drivers in well used speakers would have an fs of 'DC' :D

Looking at the data, I'd say the worst case changes plateaus around the 30 hour mark. That's where the differences get completely obscured by measurement tolerances.

But again; is it audible? Probably not. But I'm open to test showing the contrary.
Here is what Danny says. You might find additional 'claims' from Danny here.

 

ctrl

Major Contributor
Forum Donor
Joined
Jan 24, 2020
Messages
1,611
Likes
6,049
Location
.de, DE, DEU
Do comment on this... manipulated/inaccurate/wrongly measured/false/forced data?
@NTK has already provided a lot of information with his post, but I know that some here on the forum consider any information in which the word "Toole" appears to be baised - therefore, a few more details.

First of all, how big are the manufacturing fluctuations in the production of drivers?
These can sometimes be high, as the measurements from audioexpress.com show with this 250€ woofer from Seas:
1634512424781.png

As can be seen, the deviations of the resonance frequency in the TSP data for individual driver samples are 10-15%. This is consistent with the data from other manufacturers.

So then let's compare how big the difference is from "freshly unpacked" to "ran the woofer hard for 20h" and give only a few hours of rest (if Danny had rested the woofer for 48h, the TSP data might have changed again).
Here is the TSP data determined by GR-research:
1634510979580.png
1634509890158.png

The deviations in the resonance frequency are about 10-15% for the two extreme states (no break-in, full break-in).

What immediately stands out about the GR-research article? Exactly, no frequency response, distortion or waterfall measurements of the driver to show the effects of the different TSP data.

Then let's catch up and simulate the effects of these changes for the driver in a sealed enclosure. First, a sealed enclosure is simulated with the freshly unwrapped driver, and then the same enclosure is simulated with the full break-in driver.
Both frequency response and impedance response simulations are superimposed to make the differences as easy to see as possible:
1634511651431.png


Nice to see, the TSP data changes have no effect on the frequency response at all and thus do not cause any tonal change.

The article provides zero information on possibly sound-relevant parameters - no measurement on distortion, no measurement on decay, no frequency response measurement.
And the information the article provides shows literally zero tonal impact as the simulation shows.
 

Killingbeans

Major Contributor
Joined
Oct 23, 2018
Messages
4,077
Likes
7,511
Location
Bjerringbro, Denmark.
Here is what Danny says. You might find additional 'claims' from Danny here.

I'd like to see the graphs showing profound changes in the spectral decay. And I'd like to see some listening tests with proper controls showing the clear audibility of all the things he makes claims about. The fact that he brings "electrical burn-in" into the mix doesn't bode well for either of those things ever happening.
 

Raindog123

Major Contributor
Joined
Oct 23, 2020
Messages
1,599
Likes
3,554
Location
Melbourne, FL, USA
Till now the best answer to the data linked is that the whole research was made on a driver level and not system level making it most probably inaudible. Ill wait for better answers to these data.

Do we agree that speaker drivers are not "free" oscillators? That they do not resonate at their resonance frequencies and produce tones? Instead, they are "forced" to oscillate by the magnetic-force interaction (between the electromagnetic coil of the speaker and its constant magnet)? With the frequency of the AC current (ie the sound waveform) flowing through the coil!?

Yes, in theory the internal resonance frequency(ies) of the speaker membrane can affect - eg amplify or reduce - the driving signal... However, (a) this might happen only very close to the resonance frequency and (b) even there would be negligible at the Q levels we're looking at.

So, to me the data you've pointed at does indicate that speakers' internal stiffness thus resonance frequency can/does change a bit over time. However, as this resonance frequency does not define/affect the output signal (aka the "frequency response" :) ) - that is driven by the efficiency/strength of the electromagnetic force - it does not have either (a) substantial audible or (b) positive ("sound improvement") effect on the music reproduction.
 
Last edited:

YSC

Major Contributor
Joined
Dec 31, 2019
Messages
3,192
Likes
2,567
I'd like to see the graphs showing profound changes in the spectral decay. And I'd like to see some listening tests with proper controls showing the clear audibility of all the things he makes claims about. The fact that he brings "electrical burn-in" into the mix doesn't bode well for either of those things ever happening.
Well maybe we kinda let him go wining and thinks he wins everybody… you can’t convince a flat earth er that the world is…

When he only exaggerating his own experience and ignores our experience I am done with this, discussion only goes as far as logical discussion with proofs
 

YSC

Major Contributor
Joined
Dec 31, 2019
Messages
3,192
Likes
2,567
And one final comment from me:
If running in takes so long and causing night and day difference, by no way it will just gets stable and be happily ever after.

A material will experience significant change with a few hundred hours of use won’t just magically stabilise itself and never change again. Much like a car engine after the initial wearing out of imperfections in metal processing you might get a bit smoother but after that the engine is constantly degrading. So if a speaker change drastically after a few hundred hours. You at max have another fee hundred hours to enjoy and it will goes very off sounding again. Does that make sense?
 

audio2design

Major Contributor
Joined
Nov 29, 2020
Messages
1,769
Likes
1,829
OK, so you don't have a clue what GR "Research" is?

His data still appears completely accurate.

The parameters that change are complementary so the resultant in enclosure frequency response change ends up small. You would likely need test tones to audibly detect a difference and even then I doubt it.
 
Last edited:
Top Bottom