Speaker "Speed"

[Theta]

I didn't see this old thread until Eric Alexander recently made a name for himself. He was trying to assert his skills as a successfull loudspeaker designer, but fell into the same trap as many people, and I posted this in the Legal fund for Reviewers/Erin? thread a few weeks ago.

Eric Alexander behaves as if he’s a great speaker designer, but he proved the opposite with his own speakers, so the evidence is out there. He makes a feature of using a tweeter array to perform mid-range duties. The argument is that they’re lighter, so they can mechanically respond to the electrical input more quickly, but there are a few problems with this:

1. Tweeters are small, so they have to move further than a mid-driver, which undoes the supposed advantage.
2. The ability to react quickly is reflected in the frequency response. If it's flat – it’s right.
3. Tweeters have very short travel, and will cause high distortion in the mid range, even when there are several of them.
4. A tweeters ability to play in the mid-range is limited by it's own system resonance.
5. Speaker drive units are dynamic transducers, not static, which is a crucial difference that few understand, including Eric.
6. If it really was a good idea, everyone would do it.

Here’s what dynamic means. At very low frequency, the time varying displacement of the drive unit diaphragm is roughly in phase and in proportion to the applied electrical signal. The resistance to motion comes from the mechanical and acoustic suspension. The displacement amplitude is roughly constant with frequency, and the acoustic output increases with frequency. In this state, a light drive unit would indeed respond to the electrical input more quickly than a heavy drive unit. This is static transducer behaviour, and this is what happens below the drive unit’s normal operating frequency range (the pass band).

Fortunately drive units don’t work like this in their pass band, otherwise you’d never get a flat frequency response. Within their pass band , drive units work as dynamic transducers, and the resistance to motion is predominantly the mass, not the suspension. The force generated from the electrical signal into the motor goes into accelerating the heavy drive unit back and forth, and it’s that acceleration of the diaphragm and hence the air that creates the loudness – not the position, nor the displacement, nor the velocity. The acoustic output comes from the second derivative of the displacement. When a signal is applied, the drive unit starts accelerating, and when the signal is removed the drive unit carries on what it’s doing – moving, stationary, whatever.

So – does reducing the mass must improve the “speed”? No – it increases the efficiency, not the speed or the bandwidth. The drive unit accelerates the instant the force is applied, not after it’s had time to move, it’s immediate. Reducing the mass will increase the acceleration, so you get a higher output for a given input, which is higher efficiency. The upper frequency limit is generally constrained by the driver’s ability to maintain pistonic motion, and not it’s mass.

It's just well that drive units are heavy because otherwise they would never have a flat frequency response - the displacement would be constant against frequency, but the acoustic output would be steeply uptilted.

So reducing the driver mass doesn’t make it faster. Proper speaker designers know this, and unless it’s all a cynical con, Eric clearly isn't one of them, and I guess that's why this thread woke up again.

what are the requirement to qualify as a

I didn't see this old thread until Eric Alexander recently made a name for himself. He was trying to assert his skills as a successfull loudspeaker designer, but fell into the same trap as many people, and I posted this in the Legal fund for Reviewers/Erin? thread a few weeks ago.

Eric Alexander behaves as if he’s a great speaker designer, but he proved the opposite with his own speakers, so the evidence is out there. He makes a feature of using a tweeter array to perform mid-range duties. The argument is that they’re lighter, so they can mechanically respond to the electrical input more quickly, but there are a few problems with this:

1. Tweeters are small, so they have to move further than a mid-driver, which undoes the supposed advantage.
2. The ability to react quickly is reflected in the frequency response. If it's flat – it’s right.
3. Tweeters have very short travel, and will cause high distortion in the mid range, even when there are several of them.
4. A tweeters ability to play in the mid-range is limited by it's own system resonance.
5. Speaker drive units are dynamic transducers, not static, which is a crucial difference that few understand, including Eric.
6. If it really was a good idea, everyone would do it.

Here’s what dynamic means. At very low frequency, the time varying displacement of the drive unit diaphragm is roughly in phase and in proportion to the applied electrical signal. The resistance to motion comes from the mechanical and acoustic suspension. The displacement amplitude is roughly constant with frequency, and the acoustic output increases with frequency. In this state, a light drive unit would indeed respond to the electrical input more quickly than a heavy drive unit. This is static transducer behaviour, and this is what happens below the drive unit’s normal operating frequency range (the pass band).

Fortunately drive units don’t work like this in their pass band, otherwise you’d never get a flat frequency response. Within their pass band , drive units work as dynamic transducers, and the resistance to motion is predominantly the mass, not the suspension. The force generated from the electrical signal into the motor goes into accelerating the heavy drive unit back and forth, and it’s that acceleration of the diaphragm and hence the air that creates the loudness – not the position, nor the displacement, nor the velocity. The acoustic output comes from the second derivative of the displacement. When a signal is applied, the drive unit starts accelerating, and when the signal is removed the drive unit carries on what it’s doing – moving, stationary, whatever.

So – does reducing the mass must improve the “speed”? No – it increases the efficiency, not the speed or the bandwidth. The drive unit accelerates the instant the force is applied, not after it’s had time to move, it’s immediate. Reducing the mass will increase the acceleration, so you get a higher output for a given input, which is higher efficiency. The upper frequency limit is generally constrained by the driver’s ability to maintain pistonic motion, and not it’s mass.

It's just well that drive units are heavy because otherwise they would never have a flat frequency response - the displacement would be constant against frequency, but the acoustic output would be steeply uptilted.

So reducing the driver mass doesn’t make it faster. Proper speaker designers know this, and unless it’s all a cynical con, Eric clearly isn't one of them, and I guess that's why this thread woke up again.

"Proper speaker designers"
What are the qualifications for a proper speaker designer? Can you name a few.

 

[Purité Audio]

Salford University’s acoustic department is held in pretty high regard,
Geoff Martin
Jack Ocklee Brown
Bruno Putzeys
Keith

 

[welwynnick]

Salford University’s acoustic department is held in pretty high regard,
Geoff Martin
Jack Ocklle Brown
Bruno Putzeys
Keith

Absolutely.
Add Mark Glazer and Kevin Voecks to that list.

 

[Theta]

Salford University’s acoustic department is held in pretty high regard,
Geoff Martin
Jack Ocklee Brown
Bruno Putzeys
Keith

With the exception of Geoff Martin these highly capable engineers are assigned by the marketing department of the corporate office to design the best speaker they can come up with as long as it is a in the form of a column and the narrowness possible.
This criteria's only purpose is wife acceptance.
Of course these engineer will tell you that it is a technical requirements to avoid early diffraction and other bs that fit their purpose
This is like telling a formidable jockey to win a horse race with a 20kg saddle. He might win in a field of nags but not a major derby.

 

[MaxBuck]

I think the fastest speakers probably are the ones that Ferrari puts into their FXX Evoluzione. I believe 249 miles per hour.

 

[Old_School_Brad]

I think the fastest speakers probably are the ones that Ferrari puts into their FXX Evoluzione. I believe 249 miles per hour.

I believe the fastest speakers were those used in the communication equipment of Apollo 10 during its return to Earth from the Moon.

I do remember useless things, including that this was the fastest recorded speed at which any human has ever traveled. I had to look up the the speed though. It was about 40.000 km/h (24.820 mph).
In space, no one can hear the Doppler effect!

 

[Purité Audio]

With the exception of Geoff Martin these highly capable engineers are assigned by the marketing department of the corporate office to design the best speaker they can come up with as long as it is a in the form of a column and the narrowness possible.
This criteria's only purpose is wife acceptance.
Of course these engineer will tell you that it is a technical requirements to avoid early diffraction and other bs that fit their purpose
This is like telling a formidable jockey to win a horse race with a 20kg saddle. He might win in a field of nags but not a major derby.

Bruno was a jobbing engineer ( I believe ) when he came up with the Kii Three and presented it to Eelco Grimm and Guido Tent , a speaker doesn’t have to be capable of 124dB continuous to be considered ‘good’.
The D&D8c is easily capable for the majority of listeners, the new flagship will play louder, lower , extended cardioid etc etc but it’s size and cost will obviously make it less attainable for the majority.
Keith

 

[BDWoody]

This criteria's only purpose is wife acceptance.

Uh huh...

Of course these engineer will tell you... other bs that fit their purpose

Uh huh

This is like telling a formidable jockey to win a horse race with a 20kg saddle. He might win in a field of nags but not a major derby.

Is that what it's like?

 

[Theta]

Uh huh...



Uh huh



Is that what it's like?

There is that element that can't be ignored.

 

[antcollinet]

In space, no one can hear the Doppler effect!

But they can measure it

 

[Zwepp]

Me trying to understand things here.. The driver's membrane is mechanically coupled to the voice coil. The voice coil starts moving through the force of the magnet when a signal flows through the voice coil, however the mass of the membrane, spider and other passive parts will form some resistance for the electrical force. Is this what we measure as impuls response? And is this resistance also correlated to distortion?

 

[Philbo King]

Isn’t group delay the time it takes to start playing (“lag”), not a measure of how it plays it for (“hang time”)?

Nope. Group delay refers to unequal delays from a system (e.g., speaker) across different frequencies. This can cause temporal smearing.

It can be calculated as the first derivative of phase vs. frequency response.

Group delay and phase delay - Wikipedia

en.m.wikipedia.org

 

[Vergiliusma]

"The fastest sounding bass is when there is no bass." Josh Ricci

 

[dfuller]

Nope. Group delay refers to unequal delays from a system (e.g., speaker) across different frequencies. This can cause temporal smearing.

It can be calculated as the first derivative of phase vs. frequency response.

Group delay and phase delay - Wikipedia

en.m.wikipedia.org

Group delay is arguably the most telling thing for how bass "moves" other than frequency response itself. All other things being equal lower group delay is better until you get to truly low frequencies.

I think some of why big ported speakers sound quite different from smaller ones is that the port tuning is much, much lower. I want to say the KH420s have a port tuning around 30hz, which is a great deal lower than small two-way bookshelves.

 

[Vergiliusma]

Group delay is arguably the most telling thing for how bass "moves" other than frequency response itself. All other things being equal lower group delay is better until you get to truly low frequencies.

The two are linked. Group delay, which is the rate of change of the phase response, will rise if there's a change in the frequency response.

For example, if you take a ported speaker that is underdamped, meaning peaking at resonance, it will have a higher group delay than the same speaker that is tuned maximally flat for the driver Q.

Similarly, a ported speaker at its rolloff will have a higher group delay than its sealed counterpart, because it is falling at 24dB/octave vs. 12dB/ocatve--different frequency responses, different phase responses, and different group delay.