I recently received an email from someone asking the following question:
“I’ve been reading your blog for a while and a question popped up. What do you think of the practice of ”breaking in” speakers? Is there any truth to it or is it simply just another one of the million myths believed by audiophiles?”
To answer this question, I’ll tell a story.
At work, we have a small collection of loudspeakers – not only current B&O models, but older ones as well. In addition, of course, we have a number of loudspeakers made by our competitors. Many loudspeakers in this collection don’t get used very often, but occasionally, we’ll bring out a pair to have a listen as a refresher or reminder. Usually, the way this works is that one of us from the acoustics department will sneak into the listening room with a pair of loudspeakers, and set them up behind an acoustically transparent, but visually opaque curtain. The rest of us then get together and listen to the same collection of recordings at the same listening level, each of sitting in the same chair. We talk about how things sound, and then we open the curtain to see what we’ve been complaining about.
One day, about three years ago, it was my turn to bring in the loudspeakers, so I set up a pair of passive loudspeakers (not B&O) that have a reasonably good reputation. We had a listen and everyone agreed that the sound was less than optimal (to be polite…). No bass, harsh upper midrange, everything sounded like it was weirdly compressed. Not many of us had anything nice to say. I opened the curtain, and everyone in the room was surprised when they saw what we had listened to – since we would have all expected things to sound much better.
Later that day, I spoke with one of our colleagues who was not in the room, and I told him the story – no one liked the sound, but those speakers should sound better. His advice was to wait until next week, and play the same loudspeakers again – but the next time, play pink noise through them at a reasonably high level for a couple of hours before we listened. So, the next week, the day before we were scheduled to have our listening session, I set up the same speakers in the same locations in the room, and played pink noise at about 70 dB SPL through them overnight. The next morning, we had our blind listening session, and everyone in the room agreed that the sound was quite good – much better than what we heard last week. I opened the curtains and everyone was surprised again to see that nothing had changed. Or had it? I was as surprised as anyone, since my religious belief precludes this story from being true. But I was there… it actually happened.
So, what’s the explanation? Simple! Go to the store and buy two identical pairs of sneakers (or “running shoes” or “trainers”, depending on where you’re from). When you get home, take one pair out of the box, and wear them daily. After three or four months, take the pair that you left in the box and try them on. They will NOT feel the same as the pair you’ve been wearing. This is not a surprise – the leather and plastic and rubber in the sneakers you’ve been wearing has been stretched and flexed and now fits your foot better than the ones you have not been wearing. In addition, you’ll probably notice that the “old” ones are more flexible in the places where your foot bends, because you’ve been bending them.
It turns out (according to the colleague who suggested the pink noise trick who also used to design and make loudspeaker drivers for a living) that the suspension (the surround and spider) of a loudspeaker driver becomes more flexible by repeated flexing – just like your sneakers. If you take a pair of loudspeakers out of the box, plug them in, and start listening, they’ll be stiff. You need to work them a little to “loosen them up”.
This is not only true of new sneakers (and speakers) but also of sneakers (and speakers). For example, I keep my old sneakers around to use when I’m mowing the lawn. When I stick my foot into the sneakers that I haven’t worn all winter, they feel stiff, like a new pair, because they have not been flexed for a while. This is what happened to those speakers that I brought upstairs after sitting in the basement storage for years. The suspensions became stiff and needed to be moved a little before using them for listening to music.
A small problem that compounds the complexity of evaluating this issue is that we also “get used to” how things sound. So, as you’re “breaking in” a loudspeaker by listening to it, you are also learning and accommodating yourself to how it sounds, so you’re both changing simultaneously. Unless you have the option of playing a trick on people like I did with my colleagues, it’s difficult to make a reliable judgement of how big a difference this makes.
Millemissen says:
Hi Geoff – still alive and kicking :-)
So, the big question – which you probably are waiting for – is:
Does that go for a BeoLab as well, or???
Greetings Millemissen
geoff says:
Hi MM,
The answer is “probably”. This issue is different from loudspeaker to loudspeaker due to differences in the behaviour of the materials in the loudspeaker driver suspensions over time. So, it is probably the case for all loudspeakers – but to varying amounts depending on the details of the construction(s).
Cheers
-geoff
Millemissen says:
Just a thought – if you are tuning a loudspeaker and find it finished for a release to the costumers….and the drivers over time react (maybe just a bit) different, this must have an influence on how the speaker performs?
Or is the DSP solution – used in modern B&O speakers – able to compensate for these differences?
Greetings MM
geoff says:
Hi MM,
Yes, the behaviour/performance of a loudspeaker changes over time. Some changes are compensated for in the DSP (an example of this is Thermal Compression Compensation in the BeoLab 5 – although this is a relatively “short term” effect). Some changes are not. For example, if you leave your loudspeakers for 2 years without turning them on, and then, one day, start listening, the tuning in the filtering will probably not be appropriate, since the surround and spider are probably a little stiff. Then again, as things “loosen up”, the speaker becomes more and more like itself, which is more like the speaker we tuned.
Then again, there are other materials that deteriorate over time – some things more than others. Foam surrounds tend to disintegrate, for example. As this happens, the speaker is also less like itself, and therefore its tuning is less and less correct.
However, exactly how much it changes over time is an unknown, since it depends on too many variables like time, temperature changes, humidity, how much the loudspeakers have been played, and so on…
We do some monitoring of this over time. For example, all of out “målenormale” loudspeakers (“measurement reference” loudspeakers) are put back in the Cube each year and checked against our measurements from the previous years. This is done primarily to ensure that our measurement references for the production line are still reliable. However, these are “special” cases, since the only time they play sound is when they’re measured. They’re certainly not used to play music every day – so the rate at which they deteriorate (assuming that they do) is not the same as a normal loudspeaker.
Cheers
-geoff
oluv says:
I am still not quite sure about this “burn-in” legend. Because how can you tell the “burn-in” makes the speakers sound better and not worse? If they have been constructed and measured to sound good without burn-in, the sound could change much for the worse after many hours of listening, because the bass is not “tight” anymore, but “loose”?
The human ear is the last I would trust in, therefore I would have loved to see some measurement as well to underpin this theory. One taken directly after the speaker was unpacked, and another one after hours of “burn-in”.
Unfortunately I don’t have the ressources nor the chamber for this to make this on my own, as I cannot leave my equipment standing around in the same position for days, but for exact measurements it has to be done like that, without moving any object, otherwise slight differences in placement that will show up in the measurement might be taken for “burn-in effect”.
Also it would be interesting what changes one has to expect after the “burn-in”, frequency response? impulse? is this effect measurable at all? If the human ear can hear it or think that it can hear it, it should be measureable as well.
Please could you do away with this whole burin-in myth and try some real scientific measurements? My idea went even so far to check how much influence temperature would have to the sound of a speaker. A fresh unpacked speaker is often cold because of transport, so let’s measure a speaker that is around 0°C and the same speaker at 50°, because it was standing outdoor in the sun…
What do you think about this?
geoff says:
Hi,
Some responses:
All I’ve said here is that, speaking objective-perceptually (and not subjectively), that some loudspeaker drivers’ suspensions may stiff over long periods of non-use. Using them will reduce this stiffness. This makes them audibly and measurably different. Although I will admit that I have not done any physical measurements, I would certainly trust the listening team in this case. Same tunes, same room, same listening positions, same loudspeaker positions, same playback equipment, same playback levels – everything was as controlled as in a “real” listening experiment. This makes the results as reliable as a physical measurement.
Whether “different” means “better” or “worse” is not part of the discussion. I’m talking about objectively perceptual changes – not subjective ones. “Better” and “worse” are hedonic ratings – which are based on personal opinions. The sound of the loudspeaker changes. Whether you like it more or less will change from person to person.
Our loudspeakers are tuned after many, many hours of use. So, they are not tuned to “sound good without burn-in”. If the suspension of the drivers in a loudspeaker in a customer’s home change in time the same as the drivers in a similar loudspeaker that we’ve tuned, then the customer’s loudspeaker becomes more like the one we tuned. However, different suspensions undergo different changes. Some suspension types change much, much less than others – it depends on the materials of which they’re made. Also, if the compliance of the driver in the enclosure is primarily determined by the enclosure volume, then changes in the compliance of the loudspeaker driver will have less of an effect on the total result.
Also, I would ask that, until proven to be a myth, you be careful about using the word “myth” to describe the effect.
As to the question of temperature effects on the compliance of a loudspeaker driver, this is a well-known problem for automotive systems which see ranges of -30 to +140 deg C. However, at those temperature ranges, even the change in the speed of sound will affect things like room modes – so you have plenty of variables to lead you astray…. :-)
Cheers
-geoff
oluv says:
Well, of course you implied that after that burn in the unkown speaker indeed sounded better, not only “different”, please let me quote you, after the first listening session “not many of us had anything nice to say”, and then after the pink noise burn in “everyone in the room agreed that the sound was quite good – much better than what we heard last week”.
Therefore I was asking why the burn-.in always has to make the speaker or headphones sound better not worse. It is the same in various headphone forums or when reading user reviews, let me take the H6, a can where you were deeply involved. Many complain about lack of bass with this model, they say it sounds anemic and nearly dismiss it only to discover after a certain burn-in period that the bass became much better etc. I also own these headphones and I liked them from the very beginning, if the bass became “much better” over time, I probably wouldn’t like them that much. But you can read the same story again and again, everything seems to sound so much better and bassier, clearer, more controlled etc after a burn in, it never sounds worse, this is what bugs me.
If there was some scientific reproducible data, documented with measurements or even recordings (before/after) that prove the improvement of sound after a burn-in, I would believe. Of course I follow your argument about change of materials etc, I just cannot believe that this change always leads to some improvement. I know it from my sofa, and I am sure that it was more cosy at the beginning, than now after years of usage and because I became fat the Sofa definitely doesn’t have the same qualities as before, the “sitting-in” in my case led to some worsening, unlike your sneakers that seem to fit you better now.
geoff says:
okay – fair enough. I should have qualified my original story by saying that: “for that pair of loudspeakers, with those particular drivers, in those particular enclosures, for that particular period of storage time, the sound quality improved.” However, for a different loudspeaker driver or driver/enclosure, things may be more or less different, and “different” can mean either “better” or “worse” or just “different”.
It would certainly be interesting to do this experiment. I’ll put the same pair of loudspeakers that were in this story on a shelf and wait a year or so and measure them, once off-the-shelf and once after driving them for a day or so (and after letting them cool down to room temperature).
I’ll be back in a year.
Cheers
-geoff
oluv says:
ok, fair enough, I’ll remind you then 1 year later, in case you should forget ;-)
btw, will you share which particular speaker we are talking about here? you can also send me an email, if you are afraid of sharing this info in public!
geoff says:
sorry. that info I will not share. all I will say is that it was a pair of passive. 2-way consumer loudspeakers.
geoff says:
Hi again,
As I sit here, listening to a pair of headphones, I just came to realise that your comment also included the word “headphones” which have an additional parameter that needs to be considered in this discussion.
The timbral balance of a pair of over-ear closed headphones, particularly in the low-mids and low frequency bands is extremely sensitive to the sealing of the ear cups to the sides of the head. This can be easily demonstrated by just pressing on the headphones to improve the sealing – you’ll probably get a change in the bass level – most likely an increase.
If you put on a pair of over-ear closed headphones out-of-the-box, the foam is still quite new and stiff, and as a result, you will likely have leaks around the ear cups, particularly just behind your jaw. As the foam and leather (or whatever the external material is) softens, the sealing improves, reducing the leakage and changing the low-frequency balance.
This isn’t “burn-in” as such, since you could just wear the headphones without listening to any signal to get the same effect – but it certainly changes the sound.
This is also why it doesn’t make sense for me (when I’m tuning) to work on a new product. They have to be worn and softened in order to get the same sealing that a customer will get after some time of wearing them.
This also means that, if you take a brand-new pair of closed headphones out of the box and listen to them, you might not be hearing what you’ll hear in the future – even if there are no changes to the diaphragm and motor.
Cheers
-geoff
oluv says:
Yes, thanks again for this reply. I am aware of the ear cups problem. It is particularly noticeable with the H6, but in my case the bass doesn’t increase if I press it harder to my head, but I rather get some kind of “vacuum-effect”, bass or even most of the sound tends to disappear then. I don’t get this effect with a Beyerdynamic DT770 for example. Therefore I cannot use the H6 in bed, because as soon as the cushion presses against the headphone the sound becomes different.