Do Audio Speakers Break-in?

[voodooless]

Here is impedance measurements of same SB Acoustics MW16P-4 Woofer before and after break-in as recommended before TS are measured - 10min of 30Hz sine with xmax amplitude.

Nobody denied that new drivers have a very short period of break-in. 10 min or less is very typical. Most of it will probably be done within the first 2 minutes or so. This is a far cry from the hours and hours that are generally mandated.

If you measure speaker before/after it is almost unnoticable on the graphs, but you will definetly hear it. Simple experiment how it can affect - try how ported speaker sounds as is and then partially block port - put some stick inside what takes pot's square. There will be small changes on the FR graphs, but almost unnoticable, but you definetly hear the difference.

There will most certainly not be a small change. You change the port diameter, which will lower the tuning significantly. That can never be a small change. You will also have less surface area, thus increasing port noises significantly. This is very audible.

Another moment is crossover components, wires, connectors - they change the sound after warm-up, but do not want to run the battle, as there are many internet threads about all cable sound same, all amplifiers sounds same, all capacitors same.

Well, you bring it up... Unless you bring the cable into superconducting temperatures, it will have no effect on sound. I also highly doubt that a cable will warm up in a domestic hi-fi environment. Let's assume a 2.5mm2 cable, and a 50Hz 200W constant sine wave into a 4R speaker, which is probably 10 to 15 dB worse in power than real music. This would cause maybe a 1 to 2 C heatup. That is nothing! Most crossover stuff also doesn't really heat up. Resistors will, but so what?

What heats up is your voice coil. The wire is much thinner and wound up. And heat will influence the magnet, and therefore the magnetic field. This can have a significant impact on the TS parameters and can lead to audible things like power compression. Again, this has nothing to do with the general break-in nonsense that people talk about.

An example of gated measurements for a 3-way speaker with 2 ports. Red is one port blocked, green is both ports open. The difference is small - 0.2db for midrange (in a different areas!!), but it is measurable and hearable, as it change the balance. About same we have with broken-in and out of box speakers.

And now we're suddenly talking about a midrange? A midrange does not react to the port. The primary reason you see a change would be that the port resonances change if both bass and midrange are in the same enclosure, but since this is only a sweep of the midrange, not much of that would show up anyway. What I see here just looks like typical run-by-run differences. Again, has nothing to do with break-in.

we definitely hear it

I very much doubt that. What were the conditions?

 

[Maxim-ICOM]

And now we're suddenly talking about a midrange? A midrange does not react to the port. The primary reason you see a change would be that the port resonances change if both bass and midrange are in the same enclosure, but since this is only a sweep of the midrange, not much of that would show up anyway. What I see here just looks like typical run-by-run differences. Again, has nothing to do with break-in.

3-way speaker's FR is a sum of 3 drivers. If one got changes, overall result changed.

I very much doubt that. What were the conditions?

Pretty standatrad conditions, from my archive - April 2019: Audiomatica Clio, mic at 1m tweeter axis, gated window. Speakers Audes Orpheus

 

[olds1959special]

I haven't read this whole thread. I always thought burn-in was just a psychological phenomenon that was mistaken for changes in the gear.

 

[Kal Rubinson]

I haven't read this whole thread. I always thought burn-in was just a psychological phenomenon that was mistaken for changes in the gear.

Sure. Unconscious bias and neural adaptation.

 

[RexrothPigeon]

So I've been lighting my speakers on fire for nothing this whole time?

 

[Kal Rubinson]

So I've been lighting my speakers on fire for nothing this whole time?

Are you keeping warm?

 

[voodooless]

3-way speaker's FR is a sum of 3 drivers. If one got changes, overall result changed.

Then that is a gated measurement given that there is early drop off in bass. In that case the measurements might show the effect of the port resonance being altered. In which case you also didn’t capture the much bigger difference in bass response due to the lower tuning, which will be quite obviously audible.

 

[Beave]

Here is impedance measurements of same SB Acoustics MW16P-4 Woofer before and after break-in as recommended before TS are measured - 10min of 30Hz sine with xmax amplitude.

The driver change TS after break-in and it affects how it works with existing cabinet volume and port. If you break-in ported speaker it can easily go to xmax and break-in takes some 10min. If driver is mounted in sealed cabinet it may take longer.

If you measure speaker before/after it is almost unnoticable on the graphs, but you will definetly hear it. Simple experiment how it can affect - try how ported speaker sounds as is and then partially block port - put some stick inside what takes pot's square. There will be small changes on the FR graphs, but almost unnoticable, but you definetly hear the difference.


Another moment is crossover components, wires, connectors - they change the sound after warm-up, but do not want to run the battle, as there are many internet threads about all cable sound same, all amplifiers sounds same, all capacitors same.

An example of gated measurements for a 3-way speaker with 2 ports. Red is one port blocked, green is both ports open. The difference is small - 0.2db for midrange (in a different areas!!), but it is measurable and hearable, as it change the balance. About same we have with broken-in and out of box speakers.
If I scale to 50dB, there will not be any visible difference, but we definitely hear it

Do your measurements exhibit the effects of break-in, or is it that the measurements exhibit the effects of heating up the voice coil temperature? Running the woofer 10 minutes at 30Hz with xmax amplitude would significantly heat up the voice coil and thus alter the woofer's behavior.

 

[bmc0]

The difference is small - 0.2db for midrange

OK. Maybe you should include frequencies around the Helmholtz resonance in your measurement as well.

Running the woofer 10 minutes at 30Hz with xmax amplitude would significantly heat up the voice coil and thus alter the woofer's behavior.

Power dissipation is quite low for a sine wave near the bass resonance in free air. Increased temperature will increase the electrical Q (due to Re increasing), but won't change the resonance frequency. Lowered Fs means the suspension became more compliant (i.e. Kms increased).

 

[JeffGB]

I built many different speakers over the years. The break-in of woofers and possibly some midranges is real, as many have measured. There is literally no point in designing a speaker without breaking in the drivers as a matter of course. Many of you have pointed out, correctly, that it doesn't take too long to break in the drivers but that assumes you have broken them in at relatively high volume.
I live in an apartment where we listen quite quietly so as not to disturb our neighbours. The levels never reach the point of large cone movement and take much longer to loosen up. When I have a new speaker or driver, I have learned to hook them out out of phase with each other and face them directly at each other from a close range. I can turn up test tones or music far past what would normally be acceptable in the setting without any noise, and quickly break them in.
No offense to those that disagree, just a slightly different take on it.

 

[RexrothPigeon]

The break-in of woofers and possibly some midranges is real, as many have measured.

link to these studies?

 

[Keith_W]

I built many different speakers over the years. The break-in of woofers and possibly some midranges is real, as many have measured. There is literally no point in designing a speaker without breaking in the drivers as a matter of course.

I would wager that sample to sample variability is going to be larger than any effect of "break-in".

Here are quasi-anechoic measurements of my woofer.

And my mids. Compare these graphs to the break-in measurement posted by @Maxim-ICOM in post 1760. I don't have any measurements of the effect of the voice coils warming up. But i'll wager that they would be substantially greater than the effect of break-in as well.

So yes, "break-in" is real. There are tiny measurable differences. But as usual, its importance is vastly overstated and wildly exaggerated. Don't you care that your left and right speakers may not be the same? Of course you don't, because the thought probably hasn't crossed your mind. Manufacturers would rather keep that fact hidden than to publicize it. Instead, they promote myths that benefit them. Don't drink their Kool-Aid.

 

[MAB]

Here is impedance measurements of same SB Acoustics MW16P-4 Woofer before and after break-in as recommended before TS are measured - 10min of 30Hz sine with xmax amplitude.

The driver change TS after break-in and it affects how it works with existing cabinet volume and port. If you break-in ported speaker it can easily go to xmax and break-in takes some 10min. If driver is mounted in sealed cabinet it may take longer.

If you measure speaker before/after it is almost unnoticable on the graphs, but you will definetly hear it. Simple experiment how it can affect - try how ported speaker sounds as is and then partially block port - put some stick inside what takes pot's square. There will be small changes on the FR graphs, but almost unnoticable, but you definetly hear the difference.


View attachment 498348

Looks like you didn't allow the driver to cool after stress.
Without complete measurements, hard to know what happened here. Would need full measurements including R(e) to know if the driver cooled, which takes hours to re-equilibrate with room T.

Another moment is crossover components, wires, connectors - they change the sound after warm-up, but do not want to run the battle, as there are many internet threads about all cable sound same, all amplifiers sounds same, all capacitors same.

This would be so easy to prove, if true. But it isn't and really takes your previous partial measurements and casts doubt. You could clear all of this up by demonstrating.

 

[MAB]

I don't have any measurements of the effect of the voice coils warming up.

I do.

I always put drivers in the fridge, and cook them too.

Most alleged break-in claims don't let the driver cool after stress - it takes hours. The effects of heat are larger than break-in, hence the title of the thread:

Speaker Break-In and Environmental Measurements

The subject of speaker break-in comes up often. It's also called run-in. It's different than aging, or wear-out. I see people posting YouTube videos of woofers being deformed into the shape of a potato-chip, while the 'tuber yells at me to make sure and break in my drivers before ever...

www.audiosciencereview.com

How come nobody talks about the optimal temperature to listen to your speaker at? It's way more important than the break-in characteristics, and changes constantly with audio power and room conditions. Fortunately changes due to Temperature and Break-In tend to keep F(s)/Q(ts) constant. Vance Dickason's comments on the tiny changes in actual sound due to Break-In:

"Changes in box performance are trivial. The reason is that the F(s)/Q(ts) ratios remain constant before and after Break-In. This is true for all woofers."
The Loudspeaker Design Cookbook, 6th Edition, Chapter 8.20 Break-In, pages 155-156

Unlike Break-In, Temperature also changes R(e) and L(e). So changes in Temperature don't cancel out as completely as Break-In effects. The becomes more important at high-power where many non-linearities occur that audibly change the sound, which is why there is an entire field of study on high power sound reproduction.

TBH, I don't worry too much about T. I don't play my mains at 100+ dB and listen critically for changes in tone as they heat up, nor could I discern at that volume.

 

[gnarly]

How come nobody talks about the optimal temperature to listen to your speaker at? It's way more important than the break-in characteristics, and changes constantly with audio power and room conditions. Fortunately changes due to Temperature and Break-In tend to keep F(s)/Q(ts) constant. Vance Dickason's comments on the tiny changes in actual sound due to Break-In:
Unlike Break-In, Temperature also changes R(e) and L(e). So changes in Temperature don't cancel out as completely as Break-In effects. The becomes more important at high-power where many non-linearities occur that audibly change the sound, which is why there is an entire field of study on high power sound reproduction.

Not with you this on this one...but pls feel free to convince me otherwise.
I don't think ambient temperatures have much to do with anything, other than how long does it take for motor heat saturation....which is pretty long.

Voice coil temps are so much higher, and rise so much quicker, they matter I think, ........and a little bit of difference in ambient temp is squat to how fast they heat.

great vid on heat..

 

[MAB]

Hi ya MAB, not with you this on this one...but pls feel free to convince me otherwise.

I actually think we are saying the same thing.

I don't think ambient temperatures have much to do with anything, other than how long does it take for motor heat saturation....which is pretty long.

Voice coil temps are so much higher, and rise so much quicker, they matter I think, ........and a little bit of difference in ambient temp is squat to how fast they heat.

Yeah, this is my whole point, the ambient changes are squat compared to the huge swings in VC temp. And also squat compared to long-term temperature change at high power. The refrigerator and space-heater experiments are just to show that even ambient changes are as large or larger than the break-in change. So that makes break-in changes diddly-squat. Which is why I sarcastically asked why people aren't worrying about the optimal operating temperature for their amps and speakers and crossovers.

 

[NTK]

An example of of the effects of measuring a speaker at "non-standard" temperature (Amir's garage KH-80 DSP NFS measurement during winter at 57 °F [13.9 °C] vs Neumann's test at 21 °C):

Neumann KH80 DSP Monitor Measurements #3

Seems that if you want to create a riot in the streets of ASR, all you have to do is show a non-ruler flat response of Neumann KH80 DSP professional monitor (speaker). Not knowing that, that is precisely what I did in my first review of KH80 DSP: This was one of the earliest speakers I...

www.audiosciencereview.com

 

[gnarly]

I actually think we are saying the same thing.

Yeah, this is my whole point, the ambient changes are squat compared to the huge swings in VC temp. And also squat compared to long-term temperature change at high power. The refrigerator and space-heater experiments are just to show that even ambient changes are as large or larger than the break-in change. So that makes break-in changes diddly-squat. Which is why I sarcastically asked why people aren't worrying about the optimal operating temperature for their amps and speakers and crossovers.

Oh, I see, thx. Sorry i didn't catch where you were coming from.

 

[Maxim-ICOM]

Looks like you didn't allow the driver to cool after stress.
Without complete measurements, hard to know what happened here. Would need full measurements including R(e) to know if the driver cooled, which takes hours to re-equilibrate with room T.

For "warm" driver after 10min of breaking-in impedance have offset due to higher temperature.
10-15min is ok to get back ho room temperature - it does not heat up that much.
"Cooled" for 10min vs "warm" immediately after break-in: MW16TX-4

Previous image of MW16P-4 impedance was measured free-air on same driver right out of box and then broken-in for 10min at x-max and then cooled for 10-15min . You may see there is no offset in Re starting from 100Hz and up. We have only shift of impedance peak, and it affects TS, and how driver interact with cabinet volume and port tune.

 

[Maxim-ICOM]

I would wager that sample to sample variability is going to be larger than any effect of "break-in".

that's another story, and yes, drivers vary from batch to batch and it affects almost all moment - TS, frequency response and impedance what is quite important while speaker designed. While it is within tolerance what manufacturers have, we still need to re-measure drivers to get matching pairs and then match it to tweeter and sometimes re-adjust crossover for optimal result. Big job and long process to get everything done right.