Driver cone 'sound'

Reading some interesting discussion on the Harbeth user group recently about how cone material influences the overall sound of a loudspeaker system.

I often see comments about people not liking metal domes, or plastic mids etc. So what factors are at work here?

I have my views but I'm interested to hear from those who gravitate towards, or away, form certain types of driver and their reasons for doing so.

The only starter for ten I'll throw in, is that surely any effect depends on where you use the driver and whether or not you only run the driver tightly within its 'clean' area. That by implication surely brings the issue of the crossover firmly into the discussion.

 

I was chastised on the HUG for mentioning this as it was second hand info, but if I may I'll mention it here... For some years I supported a speaker firm called Musical Technology. Started by an old friend of mine, together with the designer of TDL drivers (and ex-Elac too), their speakerrs all featured metal-cone drive units, of which theirs were some of the best around and rather better than the earlier designed MA equivalents. Now, in their passband, these metal cones were extremely linear and quite low in mass, but the 6" unit took off at 5kHz badly (as do all metal cones at the top of their range). Two things were done to eliminate this, the first was to use steep third order filters (steep I think for a passive speaker) and the second was to bring the crossover dangerously down in frequency to around 1.2 to 1.8kHz.

In measurement terms these speakers were all but flawless, but sonically, there was a hardness I couldn't put my finger on, either from the "Vifa" metal-dome working too low down, or something else. Disconnecting the tweeter showed nothing untoward coming from the bass units at all, so at the time I put it down to the tweeter struggling hard. Their £1,000 model had a little "port-tube" put behind the tweeter to extend it down a touch more and this seemed to help a little.

However, and the bit that I had to edit out on a recent HUG post - On the Spendor Yahoo group, we occasionally have Derek Hughes post. Derek is son of SPENcer and DORothy and a traditionally trained speaker designer himself. He claimed in a recent post that when he experimented with metal cone drivers, he found that although the passive crossover could all but remove the resonance from measuring up at all, something in the tweeter's "acoustic" output seemed to set the cone ringing at the resonant frequency, despite no electrical excitement at all, the audible results not being pleasant...

So, in my case, having lived with the utmost natural and totally unforced nature of "BBC legacy" speakers in recent years and with huge success, my driver of suspicion has to be the often flash looking "Metal Cone" type driver.....

P.S. Metal dome tweeters can work superbly well I think, again, as long as they're working within their range and not asked to do anything they shouldn't (one of my favourite speaker brands uses these, of different types and the results are utterly beguiling and actually quite sweet in nature).

 

Good points Dave.
One of the (many) reasons that the Kensai runs the smallest of the Seas Magnesium drivers is that the main resonance is right up at 10kHz which makes dealing with it easier, assuming a fairly normal crossover point, 2.2khz in this case on a second order electrical (4th order acoustic) crossover. Go up just a little to the 5" model and that resonance drops down to 7k or 5k IIRC making it more difficult to suppress.

Interesting point from Derek about tweeter output exciting the mid cone. Presumably through transfer through the baffle as the drivers and physically connected?
Again using a driver where the resonance is placed as high as possible should bring benefits, as energy from the tweeter will fall with increasing frequency.

Just goes to show what a complex mix of compromises go into a loudspeaker.

 

Absolutely, but I think he was referring to "acoustic" output. Obviously, this is supposition on his part and hasn't yet been scientifically proven.

 

Perhaps it could be measured and then assessed.
Connect the bass driver to a sensitive line input (mic input perhaps) and record the output while driving the tweeter through its crossover. That should show how much the bass/mid cone is being excited by the tweeter output, and if there is any significant change around the resonant point.

 

Interesting points, and suggestions, made above.

For my own part, I think the sonic difference between cone materials is often overplayed and I seem to have heard examples of the implementation of all cone types that appear to contradict their supposed sonic signature. As suggested in an earlier post, I think much of the character comes from asking a driver to perform outside of its clean operating range.

As this is the Audiosmile forum, let me make a comment with regard to the Kensai. When I recently heard the Kensai speakers, one of the things that struck me was how well integrated the mid and top were and how little one was aware of the technology being used. Often when ribbon tweeters are used they seem to stand separate from the mid, 'spot lit' as it were (as demonstrated rather well by another speaker in a nearby room (Whittlebury show). So many ribbon equipped speakers have this discontinuity in their sound (to me) that it would be easy to draw conclusions about the sonic efficacy of the technology - that is, until one hears an implementation that contradicts ones prejudices (such as the Kensai).

 

Absolutely. Still an art to get it really right for as wide an audience as possible..

 

The point about excitement of resonances by other means is a very good one. The resonance/breakup mode exists so it will ring if energy can be coupled into it - which will be easy, because it represents a high electro/acoustical/mechanical impedance.

As a result I've always thought it's not enough to simply 'drive them less' electrically - e.g. by higher-order or lower-freq crossover filters, or even (probably worse) including notch filters. I lean towards the out-of-band response being key to 'the sound' - or lack of - rather than the in-band response.


I'm a fan of judicious bandwidth limitation in amplifiers for similar reasons.

 

A very interesting point. Assessing it should be easy if you simply disconnect the bass/mid driver and then drive the tweeter around the bass/mid resonance. Connect the bass driver to a mic input.

However, this does bypass the fact that some of the input signal would actually have fed to the bass/mid from the crossover, albeit at a low level. It would be extremely interesting if the level of the mechanical excitation from the tweeter was equal or more than that passed by the out of band area of the filter.

In the case of the Kensai that would be about -24dB from the main sound. I'm pleased to say that structural transmission from the tweeter should be minimal since the moving mass is incredibly low. Not that this was a main design consideration, lol.

On the general topic: I used to think driver materials had certain sounds. It's true, and it's not. It's true because certain materials do behave in a certain way. Polyprop cone drivers for example seem to have a fair bit of 3rd harmonic in the midrange. However, it's also not true in practice because it is entirely possible, as Rob says, to use drivers in a range where they work best. So if you use a polyprop driver as a woofer but not a midrange, does it sound like a polyprop driver? No, not at all.