|
Post by johneppstein on Jun 14, 2018 19:16:18 GMT -6
Perhaps you could draw the free body diagram for yourself? The force imparted by the cone to the air to make the sound is reacted to the speaker frame, into the cab. The cab doesn't fly away, so that force is reacted on the base of the cab. We can model the base as having a rigid connection to the floor (which is practically true but not necessarily so in reality, because if you're on a board-on-joist floor or something the floor can resonate, too). That base has a stiffness and damping matrix in three dimensions (x,y,z - imagine a piece of plywood, which will have radically different stiffnesses between its three dimensions, but more or less the same damping). So, the force imparted by the cone is reacted down through the base and ultimately to the rigid floor. A static load (like you leaning on the speaker) will cause a static deflection of the whole structure - think like a diver standing on the edge of a springboard but not jumping, only sideways. A dynamic load, like playing a test tone or music, will cause an oscillating deflection of the whole structure. Everything moves, nothing is truly rigid, even if it is for all intents and purposes rigid. But deflection and load are not the same thing. I promise you, the mushy spring and the large solid mass react the exact same force. They have to, because otherwise the speaker would move. The stiffness of the foam pad may have a frequency component to it - I don't know. I am not familiar with this kind of behavior, but I'm not an expert in the field by any means. The structure below the cab - every structure - has a resonant natural frequency. We can practically decouple these by introducing very weak springs between structures that effectively isolate the structures from each other's oscillations or stiffness matrices. In electronic circuit design this is not unlike a buffer amplifier. In turbomachinery this is the purpose of a coupling - a spring of sufficiently low lateral stiffness to isolate lateral motion between rotors / bodies (while maintaining torsional coupling). These are not theories, this is fact. I can't help you with you theory and perfection hangup. As the young engineers that I train will tell you, I often say "all models are wrong; some are useful". But that is what engineers do - create useful models. We don't create useful models based on a diminished understanding of the underlying principles. And when results aren't as anticipated - when we made a bad model - we understand what is causing the observed results to improve our model. In this case, though, we're talking about a fairly simple structural design problem, compared to something like a hydrodynamic bearing or the modal response of a large steam turbine blade up to 8 kHz. I have to get back to my ivory tower now. No, not really. If the mass and density of the support is sufficiently great is simply will be too massive to absorb or transmit any energy - the energy stays with the box. Deflection is ZERO. Transmission and absorption are ZERO. Whereas with an elastomer there is significant deflection/absorbtion, particularity in the lower frequencies where transmission is measurably greater than in the higher frequencies where the amount of energy is insufficient to overcome the inertia of the box and never even reaches the elastomer.
Or, to phrase it differently, if mass and rigidity becomes sufficiently great the amopunt of force required to move it becomes infinite.
Why is this so difficult to understand?
There are not theories. These are well documented FACTS, widely known in acoustic design.
You need to study some basic texts on acoustic design.
In terms of studio design, if a wall is extremely dense and rigid it will not absorb energy (generally lows, which wold be transmitted by a less dense barrier), whereas if the wall is flexible and of less mass it will both absord energy and transmit it to the outside. Same thing as highly dense and rigid speaker stands.
It's really basic physics.
PS - As every university brat (of which I am one) will tell you, ivory is really not a very good material for isolation compared to, say, lead or granite. It allows all manner of bias to creep in.
PPS - please do not teach students your erroneous interpretations of physics and design. We have Ethan for that.
|
|
|
Post by matt@IAA on Jun 14, 2018 19:44:39 GMT -6
There is no such thing as a rigid body. It exists only in theory, it's an assumption to make life easier. Everything has stiffness, which means that there is deflection no matter how small of a force you apply. F=kx. And, the force of the speaker must be reacted or else it would move away. If you say "practically zero" you'd be fine. But, it's practically zero for a box sitting on a countertop without considering heavy stands or whatever. You're talking about really small effects to begin with. It's difficult to understand what you're saying because it isn't correct. Energy absorption and transmission from sound and air is not the same thing as energy transmission and absorption through mechanical coupling. In fact, the reason the wall and the speaker stand are different is actually kind of illustrative. The coupling between the speaker come and the wall is air - that is to say, an extremely weak isotropic spring with negligible damping. The coupling between the speaker and the floor is the stand - an extremely rigid but likely anisotropic spring with significant damping. I can explain it to you, but I can't understand it for you. But, I am not going to argue with you.
|
|
|
Post by matt@IAA on Jun 14, 2018 20:34:31 GMT -6
PPS - please do not teach students your erroneous interpretations of physics and design. We have Ethan for that. My friend, I don’t teach students. I train young mechanical engineers at my day job because I am the senior design engineer at my company. I work in the field of turbomachinery (rotating equipment), and if you’re ever in the Houston area I will be more than glad to give you a tour of our factory where we manufacture and service the finest steam turbines and centrifugal compressors made in the world, some of which I have been fortunate enough to have designed and modified. I’ll also be glad to buy you a beer or whatever your drink of choice is - as long as you’ll do the same for me if I’m ever in the Bay Area. We can disagree without being disagreeable.
|
|
|
Post by johneppstein on Jun 14, 2018 20:42:10 GMT -6
Also - the reason the wall and the speaker stand are different is actually kind of illustrative. The coupling between the speaker come and the wall is air - that is to say, an extremely weak isotropic spring with negligible damping. The coupling between the speaker and the floor is the stand - an extremely rigid but likely anisotropic spring with significant damping. But if the stand is sufficiently dense and rigid it completely DECOUPLES the speaker from the floor. And does it witrhout absorbining energy.
Zero transmission = zero coupling. D'Oh!
Why is this so difficult for you to comprehend?
|
|
|
Post by johneppstein on Jun 14, 2018 20:46:33 GMT -6
PPS - please do not teach students your erroneous interpretations of physics and design. We have Ethan for that. My friend, I don’t teach students. I train young mechanical engineers at my day job because I am the senior design engineer at my company. I work in the field of turbomachinery (rotating equipment), and if you’re ever in the Houston area I will be more than glad to give you a tour of our factory where we manufacture and service the finest steam turbines and centrifugal compressors made in the world, some of which I have been fortunate enough to have designed and modified. I’ll also be glad to buy you a beer or whatever your drink of choice is - as long as you’ll do the same for me if I’m ever in the Bay Area. We can disagree without being disagreeable. Sure!
Something you need to understand is that since your industrial turbines and compressors are far more massive than even very large recording monitors the amount of mass required to decouple them adequately would be extremely, probably prohibitively, great.
|
|
|
Post by matt@IAA on Jun 14, 2018 21:02:36 GMT -6
I'm not gonna argue with you any more after this. What you're saying is not practically untrue - how you're saying it is wrong. I really wish I had a whiteboard to show you, because it would be immediately apparent. The stuff connected to the speaker can be modeled as a series of springs and dampers. In the simplest expression of this - one dimension (let's say normal to the speaker cone) the pad, the base, the floor, the cab itself, whatever, can all be expressed as a simple cantilever beam with a lateral load applied to it. This is going to move by Hooke's Law, F = K*X, where K is the equivalent stiffness of the structure. How it behaves dynamically (i.e., how the oscillation decays) is subject to its damping. The stiffness of the beam is is an equation based on young's modulus (material specific constant), the area moment of inertia (it's cross section) and the cube of its length. In other words, the material and its geometry (length, depth, whatever) constitute a fixed stiffness. If something is *really* rigid, the k value is extremely high. What that means is that for an applied force F, the deflection x is effectively zero: x = F/K, so if K is real big, x is real small. However, the force is still applied, and reacted, because for a static equation the sum of forces must be equal to zero or else something is accelerating. Good ole Netwon's Second Law, F=ma. Energy likewise must be reacted, but it is done so in very small dynamic motions because the deflection is exceedingly small, i.e., almost zero. In your example the speaker is not decoupled from the floor, instead it is perfectly coupled to a rigid body. In other words, the speaker effectively becomes part of a monolithic rigid structure. To decouple the speaker from the floor would be the exact opposite - having it float with nothing between it. This is easily demonstrated, because if you have an earthquake the speaker will jump up and down exactly with the floor if you have that big heavy stand. If it was decoupled the floor would move, and the speaker would happily not. Zero transmission = zero coupling. Yes, this is exactly right. But with a real heavy stand there is perfect transmission and perfect coupling. You're just sinking it down into a structure with effectively infinite stiffness. On the contrary, the underlying principles are identical, and if we understand Hooke's Law, Newton's second law, we can easily understand this entire situation. It is no different if it is a rotor on a hydrodynamic bearing, into a babbitted bearing, into a bearing pedestal, into a huge concrete support structure. The numbers change, but the equations stay the same. The only difference is because the loads are so big, we no longer get to assume that even a hugely massive concrete deck is rigid, like we can here. We actually want a very stiff spring between the rotor and the bearing, because we don't want things moving around. Even very, very minor effective eccentricities (effective center of gravity eccentricity in a well-balanced rotor is <0.0005" and yes I mean five tenths of a thousandth of an inch) can create incredibly high unbalance forces at operating speeds. Hydrodynamic bearing stiffnesses are high enough that these big forces give you small vibrations.
|
|
|
Post by Quint on Jun 14, 2018 21:28:48 GMT -6
Possibly, although possibly just before my time there. Is the post still there? What kicked it all off? Sorry for the OT Ethan bashing... No, the entire thread mysteriously vanished. Not locked, vanished overnight. A lot of people were rather upset about the way it was handled. IIRC, Mixerman retrieved a web archive of it, but with The Womb gone I don't know if it's accessible anywhere now, or if it's also been purged from the archive. If you're interested in the details, PM me and I'll tell you as much as I remember.
It involved Ethan's claims that Soundblaster converters produced results indistinuishable from high end professional converters.
Was that a Womb shit show or a Gearslutz shit show? I can remember both.
|
|