OK, "Rip van Woofer" and "Unregistered guest", here are a few facts from the speaker end of the chain written by someone who makes his living designing hi-fi speakers currently and car speakers previous to that -
A moving coil loudspeaker drive unit consists of a cone of whatever material, supported by a surround which is usually cloth, foam or rubber. However, underneath that cone is a suspension of flexible material that is normally made from cotton, conex or some other type of cloth, heavily impregnated with a resin to stiffen it. The type, hardness and quantity of resin is generally what sets the mechanical properties of the suspension and thus the drive unit.
I can assure you all that this suspension starts changing its physical properties as soon as you put a signal through it and it's behaviour changes drastically over the period of its initial operation and will have a profound effect on the sound of the speaker.
One of my customers when i was designing car speakers required a "break-in" of every sample we provided, which consisted of playing short duration pulses with peaks of up to 35V through the speaker for an hour to loosen things up - their reason for this was to replicate the behaviour and performance of a speaker that had been sat in a car for a year or so, operated and exposed to the variations in temperature and humidity.. One of the products I designed was a high performance 2 Ohm 9x6 subwoofer and it needed a quantity and density of resin that had never been required before by our suspension supplier, in order to pass all the necessary environmental validation testing - when one of these speakers was taken from the end of the production line and swept with a sine wave, you could actually hear the resin quietly cracking, flexing and loosening for 20-30 seconds before it stabilised! Measuring one fresh from the line and then later after the above mentioned one hour break-in would show lower drive unit resonance, lower distortion figures and in the case of full range speakers, smoother frequency response, particularly at high frequencies.
In my current company we always run our prototype speakers in at least overnight and sometimes over a weekend before making any measurements or carrying out listening evaluations. We actually use a very simple method (pink noise, speakers facing each other, one wired out of phase) and i can assure you it DOES make a difference, however it would not be practical to do this for every production pair from the point of view of both physical space and cost - a quick end-of-line sweep test will NOT run the speaker in.
The loosening of loudspeaker drive unit components has nothing to do with perception, imagination or witchcraft - it is a physical fact and, whilst i can't vouch for other manufacturers, in both of the companies i have worked for we design our speakers to take into account how their behaviour and sound changes as they run in - it isn't rocket science!
Adam.
