• Welcome to ASR. There are many reviews of audio hardware and expert members to help answer your questions. Click here to have your audio equipment measured for free!

Pressed steel speaker chassis vs cast or plastic

OK then - how do you see the sound being transmitted through the system?

What are the moving parts?
The Voice-coil, cone and dust-cap?
How much does all that crap weigh?

I cannot imagine that turning a basket out of a billet of aluminium makes sense
Surely you can visualise that

Sounds generated by the voice coil travel through the former, through the cone material as well as being transmitted to the air. Through the surround. Via reactive forces to the magnet and hence the speaker basket into the baffle. From the rear of the cone into the enclosure and reflected sound through the cone.
The basket structure can resonate. The large mass hanging on the basket structure lowers its resonant frequency.
The moving mass has variable frequency.

Most of these considerations apply to car bodies as well. Which was your last remark on the previous post

A similar thing happens with cheap cars being steel, less stiff, less resonance damping etc. More expensive gets stiffer, stronger frame, better resonance damping
 
Actually, it should be easy to test, for those who have measuring equipment. Measurement A when the driver is as it is. Then, measurement B after having cut two load-bearing parts in the chassi /basket, i.e. metal plate between surround and front/top plate and measure. Any difference in FR and distortion between A and B?

Alternatively, with driver that has thin pressed sheet metal chassis. Stiff it up, somehow. It is more cumbersome to do than to cut off but it can be done. Measurement before and after the operation.
 
  • Like
Reactions: EJ3
Surely you can visualise that

Sounds generated by the voice coil travel through the former, through the cone material as well as being transmitted to the air. Through the surround.
^As one wants^
Via reactive forces to the magnet and hence the speaker basket into the baffle. From the rear of the cone into the enclosure and reflected sound through the cone.
The basket structure can resonate. The large mass hanging on the basket structure lowers its resonant frequency.
The moving mass has variable frequency.

Most of these considerations apply to car bodies as well. Which was your last remark on the previous post
I don’t think so. The car chassis is generally envisioned as stiff and the wheels are allowed to travel up and down, etc.
Maybe an Audi, Cadillac or MB is designed to be quiet… and just about every other car except for maybe a Mini Moke or Caterham.
Adding a bunch of damping materials can do that. But extra mass pretty much translates into needing extra force to maintain a specific acceleration.
And a car’s primary function is to get somewhere, and not to create sounds like a speaker.

A similar thing happens with cheap cars being steel, less stiff, less resonance damping etc. More expensive gets stiffer, stronger frame, better resonance damping
OK - but the massive motor compared to the light weight moving stuff means that to first order the motor is fixed and the moving stuff moves.
To second order we have a spring (basket) and mass (The motor) so can resonate at some frequency.
I suppose one can try and dampen the resonator.
Making it stiffer raises the resonant frequency.
And adding more motor mass lowers the resonant frequency.
The only thing it seems like one can reasonably do is to get the resonant frequency outside of the band that that driver plays, so that it is not very excited at the resonant frequency.

If the basket is too flexy, then why not just hold the motor to the box from the magnet side as well as from the front lip?
 
Last edited:
I wouldn't even get outta bed for stamped steel basket, you Lo-Fi peasants! :rolleyes:

Seriously though, I won't buy a speaker that uses stamped steel baskets. All it says to me is "Cheap".
 
It's interesting that a relatively low-cost lines like Kali Audio's LP-x and IN-x models have custom drivers -- but much higher-priced stuff from companies like Neumann and Kii use inexpensive off-the-shelf stuff. :rolleyes:
 
much higher-priced stuff from companies like Neumann and Kii use inexpensive off-the-shelf stuff. :rolleyes:
How did you come to this conclusion?

1740894749162.png

1740894772371.png

1740894792027.png
 
Aha, interesting. :)

You as a technical developer of drivers at KEF (which I didn't know you were when I wrote my previous post) of course already know what I wrote (pretty basic stuff).:)
Others who read my post above may find something useful or interesting.

BUT speaking of KEF and pressed metal sheet, stamped steel chassis. I wonder what the classic B200's distortion levels are? Perhaps unfair to compare a fifty year old design and performance with newly produced drives, but out of pure curiosity one can wonder. The B200 driver is still a classic. :)

You know it but for others reading. B200 an 8 inch bass driver. There were different magnet sizes, which went under the name B200. Then either SP1014:
View attachment 432128

..or B200 with the large magnet SP1039:
(maybe there were more models of the B200?)
View attachment 432129

View attachment 432123
This is how the B200 looks mounted in a KEF 104Ab:
View attachment 432130
If they were in full working order that buyer got a good deal I must say. :) Sold For: £170.00.

Hi DanielT,

Your question of new vs. old KEF drivers intrigued me.

I didn't manage to find any B200, at least not without dismantling any of our lab standards, but I did find something older. I found a B110 (SP1003). I believe that this is the first driver that we have in our current documentation. Production started on SP1003 on 20th November 1969, so the design is quite old...

Here it is compared to a modern small bass driver, which also has a steel chassis.

1741103086492.png


As I didn't have a suitable box, I performed nearfield measurements of both units. Both drivers have a nominal impedance of 8 ohms, and were driven with a 1Vrms signal.
1741104103809.png

The old driver is in red and the new in blue.
Fortunately, the new driver shows considerably less distortion. The levels in the bass are not quite the same, but I suspect that the new driver would still considerably outperform the old driver.
 
Hi DanielT,

Your question of new vs. old KEF drivers intrigued me.

I didn't manage to find any B200, at least not without dismantling any of our lab standards, but I did find something older. I found a B110 (SP1003). I believe that this is the first driver that we have in our current documentation. Production started on SP1003 on 20th November 1969, so the design is quite old...

Here it is compared to a modern small bass driver, which also has a steel chassis.

View attachment 433258

As I didn't have a suitable box, I performed nearfield measurements of both units. Both drivers have a nominal impedance of 8 ohms, and were driven with a 1Vrms signal.
View attachment 433260
The old driver is in red and the new in blue.
Fortunately, the new driver shows considerably less distortion. The levels in the bass are not quite the same, but I suspect that the new driver would still considerably outperform the old driver.
I am afraid of sounding too repetitive or like a fanboy but I find it so great that people in such a big company still go into such effort to answer and demonstrate their statements, this shows not only utter professionalism but also high passion for their field and respect to the customer.
 
Hi DanielT,

Your question of new vs. old KEF drivers intrigued me.

I didn't manage to find any B200, at least not without dismantling any of our lab standards, but I did find something older. I found a B110 (SP1003). I believe that this is the first driver that we have in our current documentation. Production started on SP1003 on 20th November 1969, so the design is quite old...

Here it is compared to a modern small bass driver, which also has a steel chassis.

View attachment 433258

As I didn't have a suitable box, I performed nearfield measurements of both units. Both drivers have a nominal impedance of 8 ohms, and were driven with a 1Vrms signal.
View attachment 433260
The old driver is in red and the new in blue.
Fortunately, the new driver shows considerably less distortion. The levels in the bass are not quite the same, but I suspect that the new driver would still considerably outperform the old driver.
Very interesting. Thank you for taking the time to do this. More than I could wish for from you. :)

The result was as I could guess it would be. You say: Fortunately, the new driver shows considerably less distortion.
Which I think shows that development has moved forward in the right direction.:)
I am afraid of sounding too repetitive or like a fanboy but I find it so great that people in such a big company still go into such effort to answer and demonstrate their statements, this shows not only utter professionalism but also high passion for their field and respect to the customer.
I completely agree with what you just said.:)
 
Hi DanielT,

Your question of new vs. old KEF drivers intrigued me.

I didn't manage to find any B200, at least not without dismantling any of our lab standards, but I did find something older. I found a B110 (SP1003). I believe that this is the first driver that we have in our current documentation. Production started on SP1003 on 20th November 1969, so the design is quite old...

Here it is compared to a modern small bass driver, which also has a steel chassis.

View attachment 433258

As I didn't have a suitable box, I performed nearfield measurements of both units. Both drivers have a nominal impedance of 8 ohms, and were driven with a 1Vrms signal.
View attachment 433260
The old driver is in red and the new in blue.
Fortunately, the new driver shows considerably less distortion. The levels in the bass are not quite the same, but I suspect that the new driver would still considerably outperform the old driver.
WOW - that REALLY IS a very well-behaved 'white belly' B110 I must say :D Thanks so much for the contribution :)

It's well documented, but the later samples in the mid 80s had a necessary (for health reasons?) change and suffered a massive peak at 1.5kHz, which the current Falcon re-makes seem to exhibit too, if the dire response of their stock BBC LKS3.5A are to be believed. Some of the more wacky UK speaker makers using this driver, damped the steel chassis with a bitumen 'underseal' kind of substance I remember and all but glued the things into the enclosure. I gather steel chassis' can resonate at audio frequencies in some designs. I do note that many modern drivers as well as tweeters, seem to go ballistic when they do take off, even if said frequency is higher than the usual passband for such a driver. My ears are beyond it all now, but I still have lingering doubts, as to whether the performance in-band might be better if these massive out of band peaks weren't there at all, rather than smothered by the crossover. Mind you, I've not done any proper tests to prove or disprove this old hunch, so forgive me if it's a load of bull ;)
 
Hi DanielT,

Your question of new vs. old KEF drivers intrigued me.

I didn't manage to find any B200, at least not without dismantling any of our lab standards, but I did find something older. I found a B110 (SP1003). I believe that this is the first driver that we have in our current documentation. Production started on SP1003 on 20th November 1969, so the design is quite old...

Here it is compared to a modern small bass driver, which also has a steel chassis.

View attachment 433258

As I didn't have a suitable box, I performed nearfield measurements of both units. Both drivers have a nominal impedance of 8 ohms, and were driven with a 1Vrms signal.
View attachment 433260
The old driver is in red and the new in blue.
Fortunately, the new driver shows considerably less distortion. The levels in the bass are not quite the same, but I suspect that the new driver would still considerably outperform the old driver.
This triggered some nostalgia.
I bought KEFkit 3 when I was in my 2nd year at Imperial College on 1970, I got my only taxi ride of the year back from Audio T which was on the 3rd floor iirc of 119 Oxford street with 2 heavy boxes. I loved those speakers for years.
 
Actually, yes - from an engineering standpoint it's better in pretty much every sense. Stiffer, less resonant, reduces distortion. Whether or not they're audible is a different question... Sometimes it is, sometimes it isn't.
Hmmm. Aluminum is an elastic material just like steel. It has about one-third the density of steel, one-third the strength, and one-third the modulus of elasticity. But it is still elastic up to its yield point. And with aluminum, you don't want to ever get anywhere near its yield point or it will fatigue badly and crack.

Cast material is often more brittle than cold-worked material.

So, an aluminum basket that is lighter than a steel basket will be less stiff, though the shape of the casting might put more material (and stiffness) where it does greater good and less where it isn't needed. I have this feeling that this is rarely designed based on material principles but rather on aesthetic (and cost) principles and then tested with adjustments as necessary to fulfill functional requirements.

Plastic can be light and strong but it also tends to creep, even when reinforced with, say, glass fibers. Carbon composites can be designed to minimized creep. A lightweight basket suspending a heavy magnet out in free space seems like an invitation for long-term creep to me.

Material science is far less about the material properties than about the shape and design of the structure and how it accommodates those properties. This is why speakers should be evaluated on their performance rather than their structures.

(I'm not commenting on whether the iron in steel affects the voice coil EM field favorably or not compared to alternatives--not data in my possession--but I'll bet even that can be compensated for.)

But aesthetics matter to some folks, and de gustibus non est disputandum. I suspect that's especially true at higher price points, though the cabinetry and exterior finish would normally be the primary part of that consideration. I've managed to avoid higher price points fairly well in my choices, and been happy to live with speakers that look okay but are not works of aesthetic art.

Watches are a whole other thing (he says, wearing his Breguet XX today, which can have no functional justification at all but who cares?). And my tuba is beautifully made (and Swiss) and was priced such that I expected beautiful construction as well as superior musical capability. So, what do I know?

Rick "whose speakers had to pass the WAF but doesn't even know how the baskets are constructed" Denney
 
Last edited:
  • Like
Reactions: MAB
What matters is how the entire system performs as a finished unit. Obsessing over what materials are used where is... silly. Particularly for internal components where aesthetic appeal isn't even a factor.
Agree. Intuitive "premium-ness" of materials is not a good guide to performance.

MDF is a humble, cheap, and common building material that also happens to be excellent for speaker cabinets. It's simply hard to beat even if you are willing to spend more. This is not intuitive, and if you follow your intuition and go for (say) a thinner plywood or even aluminium cabinet, the sound might actually get worse.
 
MDF is a humble, cheap, and common building material that also happens to be excellent for speaker cabinets. It's simply hard to beat even if you are willing to spend more. This is not intuitive, and if you follow your intuition and go for (say) a thinner plywood or even aluminium cabinet, the sound might actually get worse.
MDF dust is horrible. I made my last MDF speakers over a decade ago, and never looked back. Quality plywood is also an excellent material for speaker building. Plywood is more reflective that MDF, this can be an advantage in some cases as well, particularly horns, waveguides or transmission lines.
 
  • Like
Reactions: MAB
I suspect that the new driver would still considerably outperform the old driver.
I'd be shocked if it didn't, to be honest. Driver design is much better than it was 56 years ago.
Hmmm. Aluminum is an elastic material just like steel. It has about one-third the density of steel, one-third the strength, and one-third the modulus of elasticity. But it is still elastic up to its yield point. And with aluminum, you don't want to ever get anywhere near its yield point or it will fatigue badly and crack.

Cast material is often more brittle than cold-worked material.

So, an aluminum basket that is lighter than a steel basket will be less stiff, though the shape of the casting might put more material (and stiffness) where it does greater good and less where it isn't needed. I have this feeling that this is rarely designed based on material principles but rather on aesthetic (and cost) principles and then tested with adjustments as necessary to fulfill functional requirements.

Plastic can be light and strong but it also tends to creep, even when reinforced with, say, glass fibers. Carbon composites can be designed to minimized creep. A lightweight basket suspending a heavy magnet out in free space seems like an invitation for long-term creep to me.

Material science is far less about the material properties than about the shape and design of the structure and how it accommodates those properties. This is why speakers should be evaluated on their performance rather than their structures.

(I'm not commenting on whether the iron in steel affects the voice coil EM field favorably or not compared to alternatives--not data in my possession--but I'll bet even that can be compensated for.)

But aesthetics matter to some folks, and de gustibus non est disputandum. I suspect that's especially true at higher price points, though the cabinetry and exterior finish would normally be the primary part of that consideration. I've managed to avoid higher price points fairly well in my choices, and been happy to live with speakers that look okay but are not works of aesthetic art.

Watches are a whole other thing (he says, wearing his Breguet XX today, which can have no functional justification at all but who cares?). And my tuba is beautifully made (and Swiss) and was priced such that I expected beautiful construction as well as superior musical capability. So, what do I know?

Rick "whose speakers had to pass the WAF but doesn't even know how the baskets are constructed" Denney
I don't have any data on what's used for speaker frames... But if we compare 6061 to mild steel your numbers are pretty right on the money. Of course the advantage with casting vs stamping is those more complex shapes are easier to make with casting.
 

Material science is far less about the material properties than about the shape and design of the structure and how it accommodates those properties. This is why speakers should be evaluated on their performance rather than their structures.
It is about the materials a lot of the time too.
Anything that touches corrosion, space, the ocean, heat, or strength and stiffness limits.

I agree for a bicycle or a bridge it is more about shape.
 
Is that resonance in blue at about 1.3k Hz the bracket resonance?
It could be.
As this is designed to be crossed around 450Hz, I didn't do much investigating.
From my simulations, I know that the suspension resonance is right around there, so I would more confidently guess it's that.
 
Back
Top Bottom