The "Voodoo" of Vibration in Loudspeakers

Status
This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.
Not to long ago (on another forum) I participated in a "discussion" on vibration in loudspeakers. More recently on Shin's "Percieve 2" thread there was the topic of "gain clones", and in particular Peter Daniel's Audio Sector DIY offerings (.. Peter - a man that goes to extraordinary lengths to listen to different materials in different locations for what is an otherwise a "simple" chip-amp). Just now I came across Lynn Olsen's post here (the last response):

http://www.positive-feedback.com/Issue25/dsd_olson.htm

refering to this site:

http://mother-of-tone.com/index.htm
____________________________________________________

Specifically Lynn stated:

".. As a loudspeaker designer you make the painful discovery there's going to be a residue of coloration no matter what you do—this is why I get mad when I see reviews saying the "Great Man" has finally gotten rid of all coloration. Wrong. I don't care what technology is used, there's always some added coloration that's not in the original recording..

Rather than fight it with 300-lb. composite enclosures and $3000 power conditions, the smart—and artistic—choice is to shift around the coloration so it is musically consonant. Making things out of lacquered spruce wood makes a whole lot of sense to me—the pix of guitars and guitar amps say it all. Guitar players know tone—they make their living from it!

Based on this site alone, I'll probably be making my next speaker enclosures out of lacquered spruce wood, or at least comparing it to an identical enclosure made out of Baltic Birch (also a great material). MDF by comparison is "grayish" or dull sounding, and composites I suspect are unnatural sounding. People like wood for a reason—it sounds good. I suspect that wood and brass work well together—musicians have been using them for a very long time."
_____________________________________________________

Note that I've read through the mother-of-tone site before, with particularly pertinant pages here:

http://mother-of-tone.com/vibration.htm

http://mother-of-tone.com/speaker.htm
_____________________________________________________

OK then - exactly what am I getting at here?

That control of vibration is not "voodoo". Whether it is an electrical component "part" via chassis & heatsinks (which are coupled to other "parts" in some manner), active devices like tubes or semi-conductors, transformers/inductors, capacitors, resistors, and yes even wire/cables - control of vibration is important. Perhaps in these components, each "part" alteration is subtle, (..maybe even inaudible in double-blind testing), and yet the cumulative effect may be rather signifigant. Certainly as a component contains FEWER "parts", each alteration of vibration takes on a more significant character - becoming less subtle, perhaps to the point of signifigantly altering the sound (..in certain instances). (..and note that this is not to say other effects beyond vibration control are not effecting the sound.)

It is however of particular importance in Loudspeakers because unlike other components, (with the exception of LP playback), its operation is *based* on vibration.

Consider the movement of a driver's Voice Coil. Most drivers under normal conditions move less than a millimeter in length. There are of course exceptions: bass drivers and midbass drivers requireing greater excursion for increased spl at lower freq.s - particularly for bass transients. Still, even at moderat spl's around 80 db - these drivers often exhibit very little excursion (..depending on the program material being played back).

With that in mind - think about how even *very small* vibrations, (not endemic to the signal-related current controlled AC "pulses" from the amp), can effect the linear movement of a Voice Coil.
(..i.e. it shouldn't take a genius with outstanding logic reasoning skills to figure out that this "topic" is VERY important to achieveing better sound quality.)

The sad fact is though that AT BEST this "topic" is usually an afterthought for loudspeaker builders (..professional and DIY'ers). For instance, the proffesional loudspeaker designer I corresponded with not long ago, (in the "discussion" I mentioned above), actually dismissed this as being unimportant when "so many other things were "more" important to improved sound quality - that effectivly it isn't worth pursuing". ..And hey - this guy is NOT stupid, (just the opposite in fact). So does he know something I don't in this respect? Purely from a point of logic (as stated above) - I think NOT. (..and yes, beyond logic.. *I* have actually experienced substantial alterations in sound quality from controling vibrations with drivers.)

This leads me then to the perceptual question: "Whats the deal people? Why are some of you so resistant to the idea that vibration control can be so important to sound quality (at least the net/cumulative effect) - particularly in loudspeakers?" More over: "Why aren't you doing something more to improve this of aspect of loudspeaker design?"

Ok..Ok. I realize there isn't exactly a LOT of information on this subject. I also realize that your standard mdf is just plain EASY to use (..easy to afford, easy to source, easy to shape and finish.. well relativly anyway).
_____________________________________________________

On the information "front" - the fact is that you don't really NEED a lot of information. Just use those grey cells between your ears (..i.e. use some logic). For instance:

You have 3 basic forms of vibration to control for a driver:

1. "lossy" VC motion due to driver "rocking".
2. solid material vibration transmission from, and reflected to, the driver.
3. air born vibration transmited directly to the driver and indirectly to the driver via solid material coupling.
...........................................................

What do we do to improve #1?

1. *RIGIDLY* couple the driver to the baffle or another part of loudspeaker.
AND!
2. *RIGIDLY* couple the loudspeaker to the floor.

Of course rigidly in #1 means coupling to a material with a high tensile strength that has VERY low flexing properties for its given shape. It shouldn't come as a shock that mdf is pretty poor in this regard (..actually VERY poor). Worse still - most painted mdf (..in that the paint or clear coat is often softer than the mdf).
Various metals have high tensil strength - but its expensive and difficult to work with. So what is appropriate here for a DIY'er? Concrete - or specifically well cured Cement. Sure, it does require you to make a mold, but mdf isn't to difficult to work with for a mold - just think "negativly", and considering that only "one" speaker mold is required for a pair of loudspeakers the mold may actually require LESS mdf than a pair of loudspeakers would. The cement itself is relativly easy to work with and neither requires expensive tools or materials.

For #2, you will need both a rigid connection to the floor via inexpensive spikes, AND you will need a lot of weight to oppose the forward/backward motion of the loudspeaker caused by drivers with more force and mass (..typical of your lower freq. drivers).
..................................................

What do we do to improve #2?

Well to remove vibration from the driver we need something that will transmit that vibration more efficiently. Here again, mdf is relativly poor - it dampens vibration (..except for a mode or a group of modes the material displays because of its shape). On the other hand Cement is pretty good (because its a dense substance).

Unfortunetly what transmits vibration away - invariably also transmits vibration back to the driver quite well. Here then we need to "dampen", (convert vibration to heat), the material as it moves away from the driver. Connection to a dissimilar material (like mdf) can be quite usefull. Additionally, the use of very simple constrained layering (i.e. cement/lossy material/cement) can also acomplish this (and usually much more effectivly).
...................................................

What do we do to improve #3?

Well reducing air-borne vibration from being transmitted back to the driver requires a little more thought. We are not talking about overall pressurization here - which effects driver mechanical compliance. (..*THAT* is a subject all in itself.) Instead we are talking about air "disturbance" - principally as a result of higher freq. reflections and higher resonant modes within cabinets (..if used).

Now high freq. reflections can be reduced with cabinet wall dampening. Higher freq. modes in cabinets can be altered via "box" shape - think "non-parallel surfaces" in the cabinets interior. The level of these modes can also be reduced with damening in the form of fiber "fill". Note however that as a "side" topic, "fill" (even with minimal dampening on cabinet walls) can alter the motion of the driver in a similar, (yet different), manner to problem #1 - via air flow resistance. (..and again *THIS* is a subject all in itself). So if possible use a material that dampens while providing little air-flow resistance - in an area unlikely to cause air flow resistance (..i.e. the farther away from the driver the better).

To dampen solid material vibration transmission via air-borne vibration - use damping materials (particularly simple constrained layer dampening). ALSO use cabinet "shapes" that are less likely to flex and have modes (i.e. brace the cabinets).
 
continuation:

Fortunetly (for what ever reason), proffesionals and DIY'ers alike seem to seem to pay attention to problem #3 to at least some extent (sometimes a great deal - perhaps more than warrented) - and often use materials and techniques just mentioned. (..so this problem is something of a moot point.) The odd thing here is that IMO this problem (#3) is the LEAST significant to sound quality (of the 3 problems).
__________________________________________________

Ok then since I have some experience with these 3 problems I'll try to briefly describe the their subjective effect and importance in relation to one another. (..note that this is my personal experience, and the general experience of others I've "tested".)

Problem #1: poor vibrational control here leads to a loss in definition and "transparency", imaging isn't as "tight" or well defined. Additionally at lower freq.s and particularly with respect to loudspeaker/floor coupling - low freq. extension isn't as deep (..sometimes described similar to a loss of a half octave or more in apparent extension).
This problem seems to make the biggest difference of the three when considering traditional construction techniques.

Problem #2: poor vibrational control here somewhat "straddles the effects of #1 and #3. The second most important when considering traditional construction techniques.

Problem #3: Many of you have likely experienced this subjectivly for yourself. I'd describe it as an alteration in tone - particularly because it usually hits specific passbands - either because a driver is particularly sensitive to reflective sound penetration through the diaphram in a limited passband, or in the case of solid transmission - its usually mode specific and effects a limited passband. The least important when considering traditional construction techniques. (..and of course the reason for this is that most people apply at least some bracing and fill in their cabinets - or the drivers already have their own rear chambers.)
____________________________________________________

Well I hope that this has given you something to contemplate and perhaps incorporate in your next design.

;)
 
Hi ScottG,

" rigidly couple the loudspeaker to the floor"...spikes,etc
Unfortunately,I must to disagree.

Sure my english suck and I can't affront a complex-long discussion in
my conditions BUT
why you continue to work in elastic-reaction when you can
work in inertial-reaction ?
If you want minimize vibrations ( relative movements) you need
maximum dynamically rigidity ( not static) in the 20-20000Hz.
To obtain this target
you must decoupling your speaker in the best way.
What is best way? Air spring ( newport for example) or long elastic cables.
Speaker ( suppose whole for simplicity) have one mass, the spring have a compilance and you MUST obtain a resonant frequency
below your audio band ( to say 1-5 Hz). The system approximate the "symple harmonic oscillator".
The "spring" filter the floor's reactions to the speaker's movements.
You have isolated(decoupled) your speaker from " the rest of the world" and
vice-versa. You have minimized force's exchange and you have
maximized relative immobility. ;)

perdone me for my symplicistic language
Cheers,
Inertial
 
---*RIGIDLY* couple the loudspeaker to the floor.---

A low frequencies, mouting two units on opposite sides of an enclosure may be very beneficial to get a stable origin of the sound without necessarily having a perfectly rigid coupling to the floor.
 
inertial said:
Hi ScottG,

" rigidly couple the loudspeaker to the floor"...spikes,etc
Unfortunately,I must to disagree.

Sure my english suck and I can't affront a complex-long discussion in
my conditions BUT
why you continue to work in elastic-reaction when you can
work in inertial-reaction ?
If you want minimize vibrations ( relative movements) you need
maximum dynamically rigidity ( not static) in the 20-20000Hz.
To obtain this target
you must decoupling your speaker in the best way.
What is best way? Air spring ( newport for example) or long elastic cables.
Speaker ( suppose whole for simplicity) have one mass, the spring have a compilance and you MUST obtain a resonant frequency
below your audio band ( to say 1-5 Hz). The system approximate the "symple harmonic oscillator".
The "spring" filter the floor's reactions to the speaker's movements.
You have isolated(decoupled) your speaker from " the rest of the world" and
vice-versa. You have minimized force's exchange and you have
maximized relative immobility. ;)

perdone me for my symplicistic language
Cheers,
Inertial


I'm not sure I fully understand this.. however,

I believe you are looking at it from the viewpoint of vibration transmitting to the floor and from the floor back up to the speaker. (..and a perspective not dissimilar to tone-arm in a loudspeaker)

This is not what I'm referring to in this instance. (..and it really is under problems #2 & #3.) This of course is not to say that an air suspesension isn't a good idea - as long as the loudspeakes weight is so heavy that it opposes motion (..or use forr's method.)

forr has the gist of it - and yes a typical push-pull frame-to-frame contact between two drivers in a bipole configuration can achieve similar results in this respect.
 
Ex-Moderator
Joined 2002
Simple Newtonian physics.

Lets say a cone weighs 50 grammes, and the enclosure weighs 10 kilos, a not unreasonable assumption I feel. If the cone is moving, by Newton's conservation of momentum the forces must balance. The ratio of cone movement to to enclosure movement is therefore 50/10 000, which gives us a maximum possible movement in the box of 0.005% of the movement of the cone. And that's with a box that's not coupled to anything, and is completely free to move. I'm not that worried. ;)
 
pinkmouse said:
Simple Newtonian physics.

Lets say a cone weighs 50 grammes, and the enclosure weighs 10 kilos, a not unreasonable assumption I feel. If the cone is moving, by Newton's conservation of momentum the forces must balance. The ratio of cone movement to to enclosure movement is therefore 50/10 000, which gives us a maximum possible movement in the box of 0.005% of the movement of the cone. And that's with a box that's not coupled to anything, and is completely free to move. I'm not that worried. ;)

Well said that man! I had a similar discussion with Thorsten (mr. Wang something something ;) ) once about him using flexible acrylic for his OB speakers.

I think the main issue is with the pressurisation and even incidental sound inside the cabinet from the back-wave of the woofer causing the cab walls for flex and thus make sound. (is this problem no.3?) I don't know where I picked it up from but I have a figure in my head that in a typical loudspeaker as much as 30% of the sound emitted can be from the cabinet itself.

In my view one of the best ways to solve this is with thick walls and lots of bracing (one of the reasons I believe TL sound so damn good and un-coloured). However, it seems that an even better method is to use a material that has a resonance which is out of the pass-band of the driver in it.

Standing waves in the cab itself are also very important IMO.

Perhaps a interesting way to deal with these would be to have a flexible and damped wall inside the cab, and then outside of that is a very ridged one. This way the energy at LF would be used up very quickly (HF can be treated with fill) and anything that doesn't get turned to heat will have to try and excite a very ridged material with a high resonance.
 
Vibration.. ad nauseum

Those insisting that vibrations in wires, transistors, working capacitors, etc. would paint a more convincing argument that effects are significant if they'd actually provide some sort of rational empirical (aka actual measurements) or at least an engineering evaluation of the magnitudes of said effects.

If one does the math one would find that phonon noise in wires / electronics from thermal lattice motion would swamp any extra "vibration" from external sources at anything above absolute zero by many orders of magnitude.

With speaker enclosures, with resonance effects, there definitely are effects, but then, what live performance doesn't also suffer from such effects (concert hall modes / resonance... sounding board coloration in pianos / violas / guitars) etc.?

How about some actual MEASUREMENTS rather than speculation masquerading as fact and generalizations about what must be because it's intuitively obvious, etc...


auplater
 
Still, half a percent is small. Bigger than 5 thousands of a percent, though.;)

For some make-up points:

Assume a 1cm p-p displacement at 50 Hz with masses as in your example. There is a midrange mounted on the same enclosure. How much Doppler distortion will that 0.5% motion cause at 1kHz, with a 2mm p-p midrange displacement?
 
Status
This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.