Paper loudspeakers don’t burst into flames under normal use, or even loud use. The Who, known as the loudest rock band ever, didn’t manage to light their speakers on fire. If Pete Townsend couldn’t do it I very much think it’s impossible. Magnesium bursts into flames at 630C and paper burns at 233C. I think your magnesium speakers are fine, it’s paper we should worry about. Also, some alloys of magnesium don’t burn at all.
I am leaning towards domes for the 800 Hz and up range in my next build.
You seem to know A LOT about domes. Are there any domes out there now that you would suggest?
Well that may be what you want from a diaphragm, but that is not what I want. For applications where the highest possible performance is the goal, I have a strong preference to use drivers only within their pistonic range. This means I look for drivers which have a first mode resonance frequency which is relatively high for the size of the driver. For instance, a 10" woofer with a first mode breakup at 2 kHz, a 5" midwoofer or midrange which has a first mode breakup at 8 kHz, a 30 mm dome with a first mode breakup above 20k....
Not every design needs to strive for the highest level of performance, and a smooth controlled breakup may be appropriate for those.
For 1st order crossovers you probably don't want the stiffest cone possible. These tend to ring quite a bit when you get to their resonant frequency. I have some SEAS cones that I maybe want to turn into wind chimes. Poly cones are much better in that they have high internal damping.
At the end of the day, the question is whether you can make a good sounding speaker system out of dem units.
On that basis, da Beryllium domes in Yamaha NS1000 are 'good' ... though I might claim the speaker is good in spite of da Beryllium 😊
For treble units of 25mm or smaller, there are a several good hard dome units including our own. But I can achieve equivalent aural results more easily with a soft dome. I've done a couple of exercises where I've made a soft dome sound like a hard dome. But I can't do the converse.
Got some measurements of these mythical beasts? May be difficult to confirm first mode breakup without a scanning laser Doppler velocimeter but you can usually have a good guess from an accurate impedance plot ... at least for the cones.
Good sounding cones don't 'breakup'. The radiating area starts to shrink in a controlled manner.
BTW, we found this behaviour from scanning 'good sounding' units. The 'theory' came later when we were trying to understand why they sounded good. Then the confirmation is when we tried to make other 'good sounding' cones.
Last edited:
Since we essentially hear the material resonances, i.e. the material character...-)
In addition, drivers must also be installed in a physically useful way, including decoupling, and drivers can also be modified. Midrange drivers, in particular, should be given a partial overlay in order to optimize frequency-related horizontal directivity characteristics. But all this is not available to buy.
Only diy audio can help here;-)
In addition, drivers must also be installed in a physically useful way, including decoupling, and drivers can also be modified. Midrange drivers, in particular, should be given a partial overlay in order to optimize frequency-related horizontal directivity characteristics. But all this is not available to buy.
Only diy audio can help here;-)
today there are rigid cones which don't show much resonances any more like this one from Eton, they would work with first order network, too.
Similar technique is Kevlar on honeycomb cores. The newer Eton drivers managed to tame the hard cone resonances I think due to elastic glue. Their waterfalls look nearly perfect, too.
I would be so glad to see a fullrange driver using this cone.
Once I DSPed an Eton and it was quite good sounding. It had a metal dust cap which was directly fixed on the voice coil. Only pity was the voice coil was not for fullrange with low enough inductivity. So too much correction had to be applied.
It was this one
I would be so glad to see a fullrange driver using this cone.
Once I DSPed an Eton and it was quite good sounding. It had a metal dust cap which was directly fixed on the voice coil. Only pity was the voice coil was not for fullrange with low enough inductivity. So too much correction had to be applied.
It was this one
The Eton isn't normally what 'we' (well, 'I', anyway 😉 ) would generally mean by a rigid cone; generally speaking 'rigid' tends to be more a reference to homogenous or crystalline materials e.g. metals, ceramics, diamond [natch], with Hex & similar types falling under the 'semi-rigid' type. Which isn't to say it's bad -Eton usually have decent cones within their design context, sadly in many cases let down by mediocre motors.
That aside -when it comes to multiway design I tend to be happy enough with decent examples of all varieties depending on what I'm doing. If I'm accepting cone TL modes are going to be in the passband for whatever reason (lower order filters, specific requirements) then softer / soft types with a [hopefully] controlled / progressive decoupling of outer portions are great -big fan of the Altec biflex, [coated] Scan Revelators etc. And the Curv material Seas have used in a handful of units (woven polypropylene strips) isn't a bad compromise, ditto the Hex types, albeit they do let go a bit more at the top end. Some metal types e.g. the Markaudio, Jordan etc. can work well too, with somewhat different priorities. If I want pure piston though & circumstances allow it, metal or ceramic all the way, though you have to make very sure that's exactly how you're using them; most of the time for, say, a 2-way, that means low & steep filters, potentially stopband resonance suppression to prevent the distortion amplification lower down etc. Which can = big and expensive filters since if the latter aspect is needed, you can't actually do it actively: it requires passive components. Which can be big asks in many cases. 😉
That aside -when it comes to multiway design I tend to be happy enough with decent examples of all varieties depending on what I'm doing. If I'm accepting cone TL modes are going to be in the passband for whatever reason (lower order filters, specific requirements) then softer / soft types with a [hopefully] controlled / progressive decoupling of outer portions are great -big fan of the Altec biflex, [coated] Scan Revelators etc. And the Curv material Seas have used in a handful of units (woven polypropylene strips) isn't a bad compromise, ditto the Hex types, albeit they do let go a bit more at the top end. Some metal types e.g. the Markaudio, Jordan etc. can work well too, with somewhat different priorities. If I want pure piston though & circumstances allow it, metal or ceramic all the way, though you have to make very sure that's exactly how you're using them; most of the time for, say, a 2-way, that means low & steep filters, potentially stopband resonance suppression to prevent the distortion amplification lower down etc. Which can = big and expensive filters since if the latter aspect is needed, you can't actually do it actively: it requires passive components. Which can be big asks in many cases. 😉
Last edited:
I think Vandersteen gets around it.... but, besides that, I can't remember a speaker I liked, where first order filter was used. I do use first order HP on my woofers, so they cross more easy to my subwoofers.... but that's about it.
No big fan of 1st order here either. Can have its uses, but not generally my preference.
On Vandersteen I'm a bit dubious about whether I'd call theirs rigid cone units or not, despite the 'pure piston' terms. With 1st order filtering I suspect it might not be how I'd define it, but in fairness, it's not a fixed definitition & without a full set of measurements (which might exist -I haven't gone looking & their website isn't especially forthcoming), I'll keep an open mind.
On Vandersteen I'm a bit dubious about whether I'd call theirs rigid cone units or not, despite the 'pure piston' terms. With 1st order filtering I suspect it might not be how I'd define it, but in fairness, it's not a fixed definitition & without a full set of measurements (which might exist -I haven't gone looking & their website isn't especially forthcoming), I'll keep an open mind.
For a 5" driver, the Dayton RS125-4 aluminum cone driver has a 1st mode cone resonance at about 8k. This is not an exotic high priced driver.
https://hificompass.com/en/speakers/measurements/dayton-audio/dayton-audio-rs125-4
For a bit more money, the SBA 5.5" aluminum cone SB15NBAC30-4 has excellent performance. These are my own measurements:
Here is a 10" woofer with a first mode cone resonance well above 2k
https://hificompass.com/en/speakers/measurements/scan-speak/scanspeak-26w/4534g00
I have the RS125 too... But I definitely find the MW13TX much more detailed in just the right way 👍
Depending on a person's definition of performance, a smooth controlled breakup could certainly result in a high level of performance. I think there are many ways of defining performance in a loudspeaker.
Last edited:
Yes I agree, that is why I said
This is my opinion about high performance drivers, or more precisely, my preference.
I don't expect everyone to share my preferences, and I am not going to argue that my preferences are best for everyone. I am going to push back a bit when personal opinions are expressed as if they are facts.
This is my opinion about high performance drivers, or more precisely, my preference.
I don't expect everyone to share my preferences, and I am not going to argue that my preferences are best for everyone. I am going to push back a bit when personal opinions are expressed as if they are facts.
FYI Beryllium is toxic. I once interviewed at a company that made Beryllium domes for compression drivers. All the employees had to get yearly medical exams! It is really bad if the dust gets in your lungs. I'm all for great sound but not if it kills you.
It damn well should do 😉 -as well as being about 25% larger, the TX cone in that size is probably almost as rigid as the aluminium RS, and it's got a much more advanced motor. I really like the RS series midbass units -most are still very good value for money (in some markets), but in fairness, you can beat them if you've the money.