Fast/Slow Speaker Article

Post by **GlassWolf** » Mon Sep 11, 2006 5:23 pm

I wasn't going to get ito fourier transforms, since that's typically third yer stuff, and most of the people reading this aren't really at that level.

solacedagony · Post by **solacedagony** » Mon Sep 11, 2006 5:30 pm

What is a Fourier transform? I found a Wikipedia on it, but I didn't quite understand what they were. It seemed sort of just like some kind of derived function, like how you can get the transfer function of your car, but I really am not sure.

thnking · Post by **thnking** » Mon Sep 11, 2006 6:54 pm

solacedagony wrote:What is a Fourier transform? I found a Wikipedia on it, but I didn't quite understand what they were.

It's a mathematical process to make a transformation from the frequency domain to time domain...I only mentioned it because it shows that an acoustical model of the loudspeaker "system" is directly related to the speed of the woofer- ie. there is more than inductance playing a role in woofer speed.

solacedagony · Post by **solacedagony** » Mon Sep 11, 2006 7:19 pm

I think that may be a little too advanced for me to actually figure out how to apply and understand the concepts yet, but what topics should I start delving into to start understanding these more advanced concepts? Glass, thnking? Also, thanks for everyone's contributions to this thread.

geolemon · Post by **geolemon** » Mon Sep 11, 2006 7:40 pm

thnking wrote:It's a mathematical process to make a transformation from the frequency domain to time domain...I only mentioned it because it shows that an acoustical model of the loudspeaker "system" is directly related to the speed of the woofer- ie. there is more than inductance playing a role in woofer speed.

It depends on what you think of when you say "speed".
Rise, or rolloff, for example?

A waterfall plot is a very common way to show rolloff (or "sustain").

Impulse responses are created to show not just rise time, but overhang - in other words, if you hit the speaker with a DC pulse, does the speaker excursion rise to that level, then stop and hold there? Or does the speaker overshoot the mark and oscillate about that level a bit instead?

These things play into a speaker's SQ, and certainly play into what you might perceive as woofer speed (certainly "clarity" ]ideally?[/i]
How does a speaker actually move (ie. how does it deviate from 'ideal')?

How does frequency affect motion?
How does a speaker move with respect to a single frequency (ie. sine wave)?
How does a speaker move with respect to a complex waveform?
How does a speaker move with respect to DC voltage rather than AC?

I'd bet that a quiet evening staring at the ceiling pondering these things, maybe drawing ideas on a notepad, might even answer what you are looking for, no "advanced research" required.
...and if not, at least there'll be a self-understood foundation there.

Post by **GlassWolf** » Tue Sep 12, 2006 12:37 am

and then you can take physics I, II, and III, tri, calc I - III, and so forth, then you can start entire college higher level classes dedicated to nothing but transforms :) It's absolutely mind-numbing. haha

Geolemon, where's my statement fifteen?
:P

thnking · Post by **thnking** » Tue Sep 12, 2006 3:54 am

[quote="geolemon"]It depends on what you think of when you say "speed".
Rise, or rolloff, for example?
A waterfall plot is a very common way to show rolloff (or "sustain").
Impulse responses are created to show not just rise time, but overhang - in other words, if you hit the speaker with a DC pulse, does the speaker excursion rise to that level, then stop and hold there? Or does the speaker overshoot the mark and oscillate about that level a bit instead?
These things play into a speaker's SQ, and certainly play into what you might perceive as woofer speed (certainly "clarity" ]
If someone is going to have an encompassing conversation on woofer speed, all aspects need to be looked at; the entire transient envelope (attack, decay, sustain, release).
I also believe that the article does a good job at beginning to discuss a speaker’s speed.
Of course the loudspeaker model is much more complicated than this, and force based on a voltage source has 3 main contenders regarding transient performance:
Static impedance
Motional impedance
External impedance (ie. air loading)

These are actually what forces contend with a speakers speed.
In terms of perceived fast vs slow woofers, I believe most people confuse the “speed” with the mid/sub pass band.

solacedagony · Post by **solacedagony** » Tue Sep 12, 2006 4:34 pm

Geo: Yea, that's what I was referring to, where should I start learning to get that base. Thanks for the topics to look into.

two4fifteen · Post by **two4fifteen** » Tue Sep 12, 2006 5:00 pm

GlassWolf wrote:and then you can take physics I, II, and III, tri, calc I - III, and so forth, then you can start entire college higher level classes dedicated to nothing but transforms :) It's absolutely mind-numbing.
:P

i hate these classes! their killin me right now.

geolemon · Post by **geolemon** » Wed Sep 13, 2006 3:57 pm

thnking wrote:If someone is going to have an encompassing conversation on woofer speed, all aspects need to be looked at]
I agree - but I think "attack" is where people get confused.

Attack - the rise time of the speaker - and in it's most extreme form (and only in that extreme form), that would be the sudden presence of a waveform, requiring the speaker to rise to response at a given amplitude.
Ideally, the cone motion would follow the waveform exactly - and as such, the waveform will NEVER make the speaker move any faster than the highest frequency component of that waveform - and the highest frequency the speaker is capable of playing well is essentially dependent on the inductance of the speaker. (phew)
So, even in an idealistic sense, inductance DOES determine the "attack" (speed) of the speaker in this way.

However, there is an amplitude component to consider:
If you suddenly hit the speaker with a 100hz waveform, would the speaker actually reach full excursion right from cycle #1 [of those 100 cycles-per-second]? Or are there a few cycles worth of "rise time" before the cone excursion matches the amplitude of the waveform? There are some dynamics related to standard physics, such as momentum and inertia. But realistically, such a rise time is going to be inaudible, because of how many cycles really are happening "per second", and how few it takes to 'rise' - if there is any rise time.

The impulse response test of a speaker is interesting, both in the context of "attack" an "sustain". The speaker is simply hit with a DC pulse which is then sustained, and the cone motion is monitored. The speaker will rise at something less than "perfectly vertical" (the waveform does rise exactly vertical), and then does tend to overshoot that maximum peak and oscillate a bit. The 'rise' response is credited to the inductance of the speaker. The overhang/oscillation is both motor and suspension, and IMO doesn't really represent a scenario encountered with the speaker playing music - if the speaker is playing music, those are AC signals, and the speaker naturally will be oscillating - across the entire excursion range of the speaker. But it's a good segue into...

Sustain - I really don't think this is much of an issue. A Boss coaxial will "sustain" just as well as a Focal Utopia, once it's risen and is playing a sustained tone (That's NOT to say the Focal won't sound better ).

Decay - measured as per the waterfall plot, and here's where I believe oscillation matters. The speaker is played, and signal removed, to learn where the speaker fails to control its natural oscillations well enough. It's a very interesting plot, and one measurement where the speaker's natural resonance is it's own worst enemy.

Release - ...wouldn't really seem to be a function of the speaker, but rather the electronics.
...from the speaker's standpoint, when signal is removed, if the cone isn't moving already, it's not going to start moving on it's own.

"Static impedance" - ie. inertia (more recognizable physics term)
"Motional impedance" - ie. momentum (ditto)
"External impedance (ie. air loading)" - by "external" you mean to the speaker, not "the outside of the box". The greatest pressure forces affecting what really translates into the speaker's efficiency (not speed - in fact, rather the opposite... low frequencies become mechanically impeded) are actually the pressure forces INSIDE the enclosure (both static air and port pressure forces).

In terms of perceived fast vs slow woofers, I believe most people confuse the “speed” with the mid/sub pass band.

With most people, it's the impact of the bass - which is often the midbass component, not even played (I hope!) by the subwoofer.

Take away or EQ down the midbass in the car, and people are often tremendously disappointed, even describing the sound as slow and muddy.

Conversely, take a good system (with good midbass) and turn off the sub, and people describe the sound as having "great impact" - but "doesn't hit the low stuff".

It's interesting - but as they say... perception IS reality. :happy055: