Metal drive units. What's all the fuss about?

Can anyone explain to me why speakers with metal drive units are popularly supposed to sound "bright"? Is it just a word-association thing, or is there a good reason why metallic drivers should give a particular sound quality?

From my personal experience I've never really associated metal tweeters with a bright sound, any more than I associate silk domes (or whatever) with a laid-back sound, but then I haven't exactly heard a lot of speakers to compare.

 

I understand it to be something to do with the character of the breakup mode with metal dome tweeters in general, although the Bandors sound smooth even when the breakup is clearly measurable below 20khz.

I don't find them neccessarily bright, but have yet to hear one I really enjoy listening to for long.

 

So is driver breakup generally significant in determining how a speaker sounds then? - ie. non-ideally pistonic behaviour is commonly encountered in ordinary operation?

 

Hopefully not in bass/midrange units Pete. I assume the deigner crosses over well before any effects become audible. Most metal domes I know of breakup at around 25khz (rather suddenly and violently), so the affect of this on the percieved treble quality is as open to debate as supertweeters and digital filters.

Sorry no answers - maybe one of our resident speaker gurus will be able to help out. What are the supposed advantages of metal though, and why do we need it?

 

This is what I don't understand. AFAICT if you've got a piston moving backwards and forwards and pushing air then as long as it's not porous and it's not flexing then it doesn't really make a difference what it's made of - its shape and topology, not to mention size, clearly will make a difference, but I really can't understand how people attribute different characteristics to different materials.

I understand that the different materials used - polymers, whether natural (silk) or not (Kevlar), or metals like Al or Be - are clearly chosen to be stiff and light so as to maximise the ideal-ness of the behaviour, ie. responding as quickly as possible to an applied impulse (light) without flexing (stiff). And presumably the choice of different materials may allow different construction - for example, you can weave or chemically manipulate polymers to change their stiffness properties in various directions, whereas with metals you're pretty much stuck with their (anisotropic) physical properties as you find them unless you make a bespoke alloy.

However, this doesn't explain why what material it is that pushes the air should particularly make any great difference, as long as it's not flexing to any significant degree - so I'm assuming there's something I'm missing from my knowledge of materials science (very limited) and loudspeaker design (zero).

 

The monitor Gr series only becomes bearable, with x/over reworking, far better caps and resisitors (that only just fit!!!), a rewire and some foo foo dust via the quantum filters, though not anyway near the same league as the lower B&W gear 603's etc, they can be harsh.
Ally has benfits over standard paper cones, stiffer/lighter/increased cone acceleration and ease of control, I believe easier driver assembly & manufacturing, possibly a price issue, though you always know when a metal driver is present, though a coat of C37 can work wonders, to me they never seem that natural, they are exceptions, but very few

 

It's because the materials are subject to stresses when they are moving at all those different frequencies. Different materials respond differently to stresses.

 

Metals and other high stiffness materials [kevlar] have low self damping. You use them in thinner sheets and they are therefore more prone to ringing. The problem is worst in bass-mid drivers, since you have a lot of driver area and a lot of audible range to filter out with a x-over. It just means they are harder to make good drivers with, than with other materials. The good ones are the ones that the manufacturer doesn't discreetly stop making after a year. So presumably Bandor and Jordan are pretty sorted!
Do remember, a x-over does not 'remove' unwanted frequencies, it 'FILTERS' them. A low pass at 2kHz has removed 3dB by 2kHz, ~6dB by 3kHz, ~12dB by 4kHz etc..Thats if the components are in tolerance. A driver that is ringing may well be annoying at -20dB or whatever

 

I'd say metal tweeters are often on the bright side but there are exceptions - the Harbeth Compact 7's are exceptionally smooth (but detailed) to my ears.

Matt.

 

FWIW I heard possibly the best sound I've ever heard from a hifi system this weekend at the Lisbon Audioshow. It was an all Classe top of the range setup with B&W Nautilus speakers....which are 100% metal drivers

Here are some pics of the setup:







Now I know the usual suspects will already be quietly chuckling away that anyone could find a system with Classe electronics good but believe me, it was that good. For the first time ever I heard a system that could make you believe a full orchestra was playing infront of you. The scale, dynamics and power were all there. Quite incredible. My only criticism was that perhaps the sound was a teensy bit too clean but it's a very minor one. The believable-ness of the system was incredible.

btw, the Nautilus speakers require separate amping for each driver (hence the army of power amps) and they come with their own digital crossover.

I don't think you can make generalisations about metal drivers any more than any other kind. There are so many other factors involved that can have an equal or even greater effect on the sound.

Michael.

 

i think i've just embarassed myself. that is one fantastic looking system there - i wish i'd gone to the lisbon show instead of crappy old heathrow. who cares what it sounds like - but it's a bonus that it does sound good.
cheers


julian

 

TBH Julian that B&W / Classe room was the highlight of the show. There were only about 4 or 5 other rooms with proper stereo systems and the rest was just AV demos with loud explosions .

The other room of note was the Sonus Faber / AR room:



Sonus Faber Stradivaris with top of the range AR system (CD3, Reference 2 pre, Reference 600 monoblocs). As with the B&W system, well above the £50K mark. I've heard the SF Strads on one previous occasion with Halcro amps and then they sounded shite. This time though the system was pretty impressive but not quite up there with the Classe / B&W system. The sound had a bit more life to it perhaps than the B&W system but there was something missing that prevented the suspension of disbelief so in the end you knew you were just listening to two speakers.

I think the Heathrow Show would probably have been a lot more interesting. Not quite sure why they exihibit these mega-systems in Lisbon. It's not like there's loads of people here who've got £50-100K to blow on hifi.

Michael.

 

IIRC (one of?) the original metal domed tweeters was fitted to the Celestion SL6/600 and that had a reputation for being a bit dull.

Paul

 

Now that would have been quite a system if they had used proper speakers instead of those rather daft looking Sonus Flabbers. A pair of D100s would have fitted the bill nicely I feel.

 

Surely you are meant to say that they were really dull and sent you to sleep? And how they couldn't time or boogie or something? I though that this was the general idea?

 

There was a slightly (only slightly) less pricey AR system driving some ProAcs, don't remember which ones and I didn't get a chance to listen to it properly unfortunately as I got distracted and then forgot to return to the room. Others at the show say it wasn't a patch on the SF + AR and Classe + B&W systems.

Here's a pic:



Michael.

 

Who was that, Antonio?

That looks like the CD3, LS25 and I dont recognise the power amps, they look to be a different configuration to the VS110 powers. Whatever, they are very considerably cheaper than the ref2/600s in the SF room!

 

Prehaps Mike should take a trip to Titains, " Would be a great experiance for him"

 

Not in any macroscopically significant fashion if the stress is insufficient to cause deformation of the sample, which is AFAIK the case for real-world operation of speaker cones (ie. the cone isn't flexing).

But presumably that's not a characteristic peculiar to metal drivers, or indeed any other - drive units are necessarily made out of light, stiff material for the reasons we've already covered.

All I'm getting at here is that you often see particular qualities ascribed to metallised drivers on this forum and elsewhere, and I'm trying to understand why this should be.

And yes, I'd happily take a pair of the Model Nautilus, although I'd be prepared to make do with D100s too

 

First part, yes and no. Ringing isn't just a function of the thickness of the material used. It is a function of the material itself. Why do you think bells are made of metal? because they ring. Think of those things you can get to hang in the wind made of tubes. How does the sound of the metal and woodern ones differ? Apart from the obvious difference in tone, there is considerable difference in the sustain of the note.

As Graham has stated, (although to the best of my knowlege hes wrong about kevlar - IIRC it has very good self damping), it's not just about cone break up it's also about internal resonances of the material and how well damped they are. These resonances are very often "in band" and aren't nesasarily shown up by a frequency sweep. It's perfectly possible for the dome/cone to be ringing at a particular frequency but not do so in a way to significantly affect the amplitude of the drive units frequency response at that point. It will however, colour the sound as the unit will continue to output these frequencies after the input has ceased. This is best revealed in "waterfall" plots of drive units. where the ringing shows up as a continued output in the time domain of the speaker. IMO, this characteristic is significant in determining the characteristic "sound" of differing cone/dome materials. As if you think about it, it will directly affect the timbre of the decay of notes.


Here's an example of a typical aluminium dome tweeter:

http://www.seas.no/seas_line/tweeters/H398.PDF


Take a look at the "cumilative spectral decay" plot. Notice the four distinct back-front ridges on the plot (from below 10khz up to 20khz) ? This is ringing. What is significant is that they are there when the other frequencies around them have ceased to show on the plot. Hence they are "standing out" for our ears to hear.

Here is the same thing again for a fabric dome tweeter:

http://www.seas.no/seas_line/tweeters/H519.PDF


Of course it doesn't mean that all metal or fabric dome tweeters have these issues. Even amongst SEAS own range there are aluminium dome tweeters with less ringing than some of their fabric dome tweeters.


GTM