Speaker Break In: Fact or Fiction? - page 2
What about the case of a vented box loudspeaker system? What sort of pre/post break in changes in amplitude response occur?
Graphic 5 & 6 illustrate that pre/post-break in amplitude response and system impedance differences are, much like that for the totally enclosed box, very small indeed. This comes as no surprise. Just as in the case of the totally enclosed box, the compliance of the vented box (for most common, practical implementations of that design now in production) has the dominant influence, minimizing the effect of variations in driver suspension mechanical compliance.
To put in to perspective just how small the variances were between the pre/post break in systems modeled above, a further 4 simulations were run. The purpose of this series was to see how normal production unit-to-unit driver spec variances would affect final amplitude response of a system.
Four mid-range drivers, selected from the same production run, were measured and the data taken were used to work up the simulations. A totally enclosed system was designed for the first of the four drivers. The amplitude response of that system was selected as the baseline to which all others would be compared. Then the simulation was run again, each time substituting one of the remaining three drivers for the first in the cabinet designed for the first driver.
When the test series was run to completion, the resulting amplitude response graphs indicated that an end user would likely encounter larger system-to-system amplitude response differences (~1.04 dB Spl) owing to normal driver variances than would be encountered breaking in raw drivers. In Graphic 7 we see the results of that run.
Further perspective was gained by running four additional simulations, this time modeling amplitude response variations arising as a consequence of altering suspension compliance 5%, 20% and 30%. The amplitude response in each case was compared with a baseline response, derived from the system designed around the data pulled from direct measurement of the first driver. From the resulting plots it can be seen that even compliance changes on the order of 20% - 30% have very little (~.65 dB spl @ max. passband variance) effect on a system's amplitude response.
From the foregoing analyses, it's reasonable to conclude that suspension compliance changes arising as a consequence of initial driver burn in has little effect on the performance of a loudspeaker system.
But once a driver is past the initial suspension-stretching break in, do further changes occur?
Regardless of how completely broken in a driver may be, while under operating conditions many well-known & understood mechanisms come in to play that affect several of a driver's measured parameters, including suspension mechanical compliance.
When under normal operating conditions, its not at all unusual to see in a driver that has already been broken in a further 5% to 20% drop in a fs , due largely to an increase in the mechanical compliance of driver's suspension. In this case, typically it's the temporary elastic deformation of the surround that's the cause of the increase in driver compliance.
Cease stimulus and the driver's compliance will return - in most cases within seconds or minutes depending upon surround design, material composition, ambient temperature and so on - to its pre-stimulus value; the compliance changes are temporary. So too are the changes that occur in all the other driver parameters that are effected by compliance, hence the changes in fs , vas, etc.
Anectdotes & A Few Useful Observations
It was the late 80's and one day I found myself in possession of 8 newly-delivered JBL 2226H LF drivers along with some JBL 2447H MF/HF drivers that were bound for a FOH system I was designing for a small theatre in New York. In those days, my practice was to burn in every driver that came across my bench for extended periods of time, then use the data derived from post-break in measurements in the design phase of the project.
Round about that same time, though, I was growing disenchanted with all the time I was spending burning in various drivers because my pre/post break in parameter/performance assessments consistently turned up next to nothing in terms of measurable differences. Where were the changes I should have been seeing?
In those days I didn't understand the process of driver burn in well enough to realize it's the initial burn in that makes the largest difference in a driver's performance and anything done subsequently along those lines does next nothing in terms of lasting effects. So it was with my incomplete understanding I set up a weekend-long burn in session of the 2226Hs, along with the 2447s.
I'm not sure now whether it was because I was becoming convinced that extended burn ins were all but useless or because I had just acquired a bunch of new test gear that I went ahead and measured the drivers pre-burn in. Anyway, I set up the burn in session that Friday and didn't return to do the post-burn in measurements until the following Monday.
The following Monday I measured each driver twice: first immediately after disconnecting it from the test rig, then again about 4 hours later. Aside from differences attributable to changes in each driver's voice coil temperature, in none of those drivers did I uncover any significant changes in any of the measured parameters or performance.
Four years after the system had been installed in the theater, I had the opportunity to go back and take some measurements. I availed myself of the opportunity and found that the amplitude response plots made that day matched to within a fraction of a dB those made when the system was first installed. Four year's worth of "breaking in" hadn't affected the system to any significant degree.
In hindsight, its clear the 2226s had been sufficiently broken in long before I took possession of them and it's doubtful any amount of burn in would have materially effected the 2447s, which sport titanium diaphragms.
A few years later, I found myself, literally, peering into a warehouse filled with raw drivers. Most of these drivers were the product of various Asian manufacturers, none of which I had any previous experience with.
At that point in time, I was working on the design of a 3-way system and was at that stage in the process where I had to select an appropriate woofer for the system. The head of engineering recommended I take the time to burn in each candidate driver before doing any measurements as some would likely never have been burnt in at all. Being familiar with the driver manufacturing process, it was hard to imagine a driver never being exposed to a suspension-stretching stimulus, but as it turned out, some of the candidates did indeed exhibit symptoms of never having been burnt in.
At that particular establishment, driver burn in was a quick process: a sine wave, delivered at the driver's pre-burn in fs and at an amplitude large enough to stretch the driver's suspension, but not so large to cause damage was used. Following initial burn in, suspension compliance would show the usual expected shift, then eventually drift back to and settle at a value something on the order of ~5% to 10% greater than than that measured pre-burn in. In this case, taking the time to burn in drivers was necessary.
In an electrodynamic driver featuring the usual surround-diaphragm-spider construction, driver suspension mechanical compliance plays a key roll in determining the measured value of various driver parameters. All of these parameters will shift as the mechanical compliance of the driver's suspension shifts in value. The bulk of a driver's compliance shift will occur at the time of initial burn in.
Subsequent shifts in compliance are largely temporary in nature. An example of one such mechanism contributing to such temporary shifts is that which arise from the elastic deformation of butadiene-styrene surrounds. Given sufficient time to recover, these changes tend to reverse themselves and the driver returns to its pre-stimulus state.
As the enclosure compliance in both totally enclosed boxes and vented cabinets dominates that of the driver for most practical implementations of either type enclosure currently in production, any potential changes in system amplitude response attributable to changes in driver suspension mechanical compliance tend to be minimized. Normal production unit-to-unit driver spec variances can affect final amplitude response of a system to a larger degree than that expected from normal pre- post-burn in driver suspension compliance changes.
Thiele, A. N.: "Loudspeakers in Vented Boxes: Part I", Journal of the Audio Engineering
Society, Vol. 19, #5 May 1971
Thiele, A. N.: "Loudspeakers in Vented Boxes: Part II", Journal of the Audio Engineering
Society, Vol. 19, #6, June 1971
Small, Richard H.: "Direct Radiator Loudspeaker System Analysis", Journal of the Audio Engineering
Society, Vol. 20, #5, June 1972
Small, Richard H.: "Closed-Box Loudspeaker Systems Part I: Analysis", Journal of the Audio Engineering
Society, Vol. 20, #10, Dec. 1972
Small, Richard H.: "Closed-Box Loudspeaker Systems Part II: Synthesis", Journal of the Audio Engineering
Society, Vol. 21, #1, Jan. /Feb. 1973
Benson, J. E.: "Theory & Design Of Loudspeaker Enclosures", Howard W. Sams & Company,
Indianapolis, IN, 1996
Button, Douglas, J.: "Heat Dissipation & Power Compression in Loudspeakers", Audio
Engineering Society Preprint #2981
Gander, Mark R.: "Dynamic Linearity and Power Compression in Moving-Coil Loudspeakers",
Journal of the Audio Engineering Society, Vol. 34, #9, September 1986
Klasco, Mike.: "Voice Coil Design in Loudspeakers", Sound & Communication, March 1997
Klasco, Mike.: "Woofers and Subwoofers", Sound & Communication, July 1994
Confused about what AV Gear to buy or how to set it up? Join our Exclusive Audioholics E-Book Membership Program!