About my "piston until 400hz" point is a bit off topic, but most people don't have access to Finecone. Luckily they provided a demo here:
At 1000hz, the cone is flexing very visibly already. but the frequency response is still looking perfectly innocent at that point.
In my opinion, when you see a dip at some frequency (like 1.2khz for MW16p), the cone has already started flexing at about 1/2 of that frequency. 1.2khz is just the frequency that the flexing is completely out of phase(inverted).
At 1000hz, the cone is flexing very visibly already. but the frequency response is still looking perfectly innocent at that point.
In my opinion, when you see a dip at some frequency (like 1.2khz for MW16p), the cone has already started flexing at about 1/2 of that frequency. 1.2khz is just the frequency that the flexing is completely out of phase(inverted).
Last edited:
Already have 4 x SP23NRX in my system and find them quite good. The curiosity simply drew me further, before I settle on the final design, like now, where I consider trying out the dedicated Purifi midrange instead of the MW13TX that I use now - even though I find it very good.
Compared 2 x SB23NRX against 2 x WO24P and 2 x RS225. For bass - all in around 60 liters closed.... I really could not hear any difference, when put in a combined system, with subwoofers, dedicated midrange and tweeter ( 3 way with subs).
There's more to losses than just parameters alone.Oh... I have been waiting for info like this for yearsAre you saying that - let's say two 8" woofers - with one having Qms of 1 and another maybe 5. That this is dominating at fs and actually means nothing at around 100-500hz?
Because I have been tumbling the thought for way to long, that a driver with lower Qms might have higher losses, and therefore sounded less detailed at higher frequencies than a driver with higher Qms and therefore lower losses.
But in reality it's way more about performance, much closer to resonance?
Like everything in audio, it is a trade-off between all components and - more importantly - the interaction between (sub)subsystems.
Qms should be considered in light of Cms and Rms as well as the application (to put it simply).
racingpht made some valuable comments about paper cones. The (adjustable) damping properties of paper are unmatched in my eyes.
Nowadays, all kinds of high-tech fibers are available for optimized stiffness/damping ratio, such as Aramid, Rohacell/Carbon, TeXtreme etc.
Ultimately, none of those modern materials appeal to me when it comes to reproduction quality (TeXtreme was particularly disappointing).
Aluminum is acceptable for smaller cones, but I have yet to hear the first speaker with Magnesium SEAS woofers that really makes me happy > and I've listened to more than a few, including 2 versions of the Grimm LS1.
That's my two cents.
Last edited:
Overly picky perhaps, but Textreme is not a ‘material’ but a type of fabric or weaving technique.
The fibers can be high-mid-low modulus and incorporate Innegra, aramids PBO etc., the resin systems can vary as well.
It’s early days still for the incorporation of carbon fiber tech in cone manufacturing, paper has a Vorschprung durch Technik of many decades.
The fibers can be high-mid-low modulus and incorporate Innegra, aramids PBO etc., the resin systems can vary as well.
It’s early days still for the incorporation of carbon fiber tech in cone manufacturing, paper has a Vorschprung durch Technik of many decades.
The losses reflect themselves in the higher current demand from amplifier because of lower impedance values in the resonance region.
Example.... ?
Are you saying that, as long as we are far away from resonance, then Qms matters less?
I find Textreme very neutral.... In active systems like mine.
About the LS1 - I thought they forgot the breakup of the 8" Seas - sounded shrill to me. I can easily make the DXT sound nice at home with a smaller midrange and narrow cabinet
How? I thought that Qms was a part of the suspension, meaning that the surround definitely has a say in the whole movement of the cone. And since the suspension definitely has a role in damping some of the cone's unwanted vibration... then how can you separate the two entirely?
Aren't they linked?
http://www.troelsgravesen.dk/Edge-coating.htm
It is quite simple, let's split it up:
A loudspeaker driver can be compared to a spring with a mass attached. The spring has a compliance, that is, for a given force, it will move a certain distance. This is Cms. The softer the spring, the higher Cms will be.
When this mass loaded spring moves, it will encounter some resistance. This is Rms.
A mass loaded spring will oscillate, just like a pendulum, when it is set in motion. A spring/mass system with a larger compliance will have a lower resonating frequency. This is quite intuitive. Just like adding mass would lead to a lower frequency.
The Q of a mass loaded spring is a measure of how long the spring/mass system will continue to oscillate after a disturbance. High Q means long oscillation. And it implies a low Rms. Because resistance would brake the oscillation.
Qms is a measure of how long a cone will continue to oscillate after a disturbance. Low Qms means that a lot of braking power comes from mechanical resistance, Rms.
This has nothing to do with resonances within the cone itself. You can have a very well damped cone, in a driver with a very high Qms. As a matter of fact, this is the best situation. You don't want resonances within the cone for obvious reasons.
But why do you want high Qms? Because it is directly related to Rms. And all mechanical breaking is highly non-linear. This non-linearity is a direct cause of loudspeaker distortion. By having low mechanical losses, you can minimize this contribution to loudspeaker distortion.
Therefore, the goal of maximizing Qms is not just a fad, it is essential for the development of low distortion drivers.
A loudspeaker driver can be compared to a spring with a mass attached. The spring has a compliance, that is, for a given force, it will move a certain distance. This is Cms. The softer the spring, the higher Cms will be.
When this mass loaded spring moves, it will encounter some resistance. This is Rms.
A mass loaded spring will oscillate, just like a pendulum, when it is set in motion. A spring/mass system with a larger compliance will have a lower resonating frequency. This is quite intuitive. Just like adding mass would lead to a lower frequency.
The Q of a mass loaded spring is a measure of how long the spring/mass system will continue to oscillate after a disturbance. High Q means long oscillation. And it implies a low Rms. Because resistance would brake the oscillation.
Qms is a measure of how long a cone will continue to oscillate after a disturbance. Low Qms means that a lot of braking power comes from mechanical resistance, Rms.
This has nothing to do with resonances within the cone itself. You can have a very well damped cone, in a driver with a very high Qms. As a matter of fact, this is the best situation. You don't want resonances within the cone for obvious reasons.
But why do you want high Qms? Because it is directly related to Rms. And all mechanical breaking is highly non-linear. This non-linearity is a direct cause of loudspeaker distortion. By having low mechanical losses, you can minimize this contribution to loudspeaker distortion.
Therefore, the goal of maximizing Qms is not just a fad, it is essential for the development of low distortion drivers.
I'm not fully agree with that. In my understanding, electromechanical damping is linear. So, a driver can have a low-loss surround and spider, but an aluminum former, it's will be a low QMS driver, but it's probably better than a high-loss surround driver with non-conductive former, all else being equal.
Also, the biggest "RMS" of a driver is the voltage-driven(shorted) voice coil. It's electromechanical damping, and Qes is many times lower than Qms in a typical driver.
I would suggest they are "weakly linked". For example, a high-loss surround can absorb more edge resonance, hence increase cone-damping, but also increase RMS.
For example, Seas MU10RB-SL uses high-loss surround, resulting a bit smoother response than FU10RB, and much lower QMS. Same as Seas W12CY006's edge coating leads to a low QMS driver even with Ti former.
Also, Mark Fenlon talked about difference former affects cone breakup
https://hifiduino.wordpress.com/2013/02/02/markaudio-alpair-speaker-drivers/
"Two types of voice coil were tested before selecting the aluminum body type:
...
Alu coil bodies have a higher material damping factor, so we’ve the better looking frequency response of these 2 prototypes.
...
"
Last, from https://audiotechnology.dk/faq/
"Drivers with Kapton voice coil former, can have a tendency to a more pronounced break-up before roll off, because of the lack of mechanical brake."
Last edited:
I did not want to make my long post even longer, but indeed, part of Rms is electrical braking. Which is also highly non-linear and for that reason aluminium voice coil formers are typically not found in good drivers. All aluminium voice coil formers are slitted in order to minimize eddy currents in it, but even then. Their main advantage is cost and thermal.
Break up is an entirely different issue. It may be impacted by the voice coil former, dunno, but cone termination by the surround is obviously important.
Edit: just read the links you posted. The Danish firm has it all wrong imo. I don't know of any recent professional bass driver with an Alu voice coil, it's mostly fibre glass. Drivers shouldn't bottom out, but if they do, with Alu it is one strike and out because it will deform immediately. A fibre glas vc former might survive.
Mark audio is a special case. On the whole, you would want the voice coil former to transmit all forces generated by the voice coil without modification through flex.
Break up is an entirely different issue. It may be impacted by the voice coil former, dunno, but cone termination by the surround is obviously important.
Edit: just read the links you posted. The Danish firm has it all wrong imo. I don't know of any recent professional bass driver with an Alu voice coil, it's mostly fibre glass. Drivers shouldn't bottom out, but if they do, with Alu it is one strike and out because it will deform immediately. A fibre glas vc former might survive.
Mark audio is a special case. On the whole, you would want the voice coil former to transmit all forces generated by the voice coil without modification through flex.
Last edited:
If I were a Dane, I would consider that in insult ;-) on the Skaaning family (Ejvind and Per) which are regarded to be top notch driver designers of all times.
Their composite former, Kapton center section with the VC winding on top and conductive "shorting ring" ends, is one of few designs that tackle an overlooked property of VC drivers: once the coil leaves the gap, electrical braking (from a termination with a low impedance) becomes nonlinear, getting reduced. But these termination "semi-shorts" counteract that perfectly, keeping the electrical damping uniform as they are designed to have about the same (BL)²/Re (== same damping) than the main coil.
By this, while the driver still starts to distort when overdriven excursion-wise it provides a truly constant damping of any cone motion which means that recovery from overdrive is nice, fast and clean, notably it is stable and non-chaotic. Technically, step response after any excitation, be in internal or external (the motor/cone distortion error signal must be regarded as external here) is always the same, and well-damped exactly to the design target even when overdriven and that pays off.
Slotted conductive formers are seldom used in PA / power drivers because they reduce top end and they heat up, both things not what you want here. In HiFi realms, nothing speaks against slightly conductive slotted formers making use of the rather linear (but frequency-dependent) eddy current braking, in an attempt to additionally damp cone-breakup parasitic motion and convert it into heat.
btw, another overlooked factor in this discussion is linear resistive damping from air load, moving air in a dipole fashion when measuring T/S-parameters. With large drivers of very lightweight construction of cone and VC assembly this creates a Qms penalty as the resistive air load is brickwalling it. True mechanical Rms should be measured in vacuum and under current drive conditions. For a smaller size high excursion bass driver with half a kilo of moving mass air load is irrelevant
Hi, actually I would love to hear why you think electromagnetic brake is non-linear, other than position-dependent issues similar to BL(x) caused Qes(x) non-linearity(KTSR has a good point on this).
Another issue I have talked before, is modulated/distorted eddy-current can be induced in ALU former, generating distorted cone motion. but that's a side-effect of conductive former at higher frequencies. The mechanism is not quite the same as electromagnetic brake (motion induced).
Any other theories for why electromagnetic brake is nonlinear?