"First Watt" Philosophy, for Drivers?

[cogitech]

The "First Watt" amplification approach seems to have been embraced and the fruits of that approach are enjoyed by many. I really do believe that the first watt matters the most.

So, I got to thinking. Could the same/similar approach be used in driver design?

The reason I am curious about this is because I have often received or viewed comments by others regarding cone excursion and/or distortion when pushing a full signal through full range drivers in specific cabinet tunings.

"Sure, that'll work, but as soon as you push 5 watts through it the cone excursion will be way beyond Xmax", or "at over 7 watts, the distortion will be off the chart".

These are not direct quotes, but examples of what I see stated so often.

In the back of my mind, I am always thinking "Uhhh, but within the 1st watt it is perfect..."

So I am wondering;

1) Just what compromises are made by a driver designer in an effort to satisfy the "15 watts is nothing" crowd?

2) What gains could be made if a driver designer set out with the goal of making a driver (or series of drivers) with the widest frequency response as the main priority, even if it meant producing a very "low watt" driver?

3) Allowing for headroom, could a "2 watt", 90dB driver with generous Xmax be designed to produce say 30Hz to 20kHz, +/- 2dB ?

4) Is there any better driver designer to consider and/or answer these questions than MarkAudio?

(OK, that last one is a rhetorical question)

 

[jcx]

read Geddes - has published peer reviewed listening test of amps, distortion preception, explored loudspeaker limitations

I believe a summary would be that loudspeaker drivers just don't have humanly audible objectional low level nonlinearity - they all pass the "first watt" criteria

and even many higher amplitude "soft" nonlinearities of loudspeakers aren't readily audible

still have to avoid bottoming the voice coil or cap against the pole pieces, and xmax may be very (over) optimistically defined by the manufacuterer

Geddes speaker designs do use extremely high power rated drivers by most diy speaker builder's standards
the Abbey 120 W rated compression driver efficency boost from the waveguide requires a ~10 kHz low pass XO to flatten its response - so its driven @ -20 dB @ the ~1kHz XO to the 12", 2 kW low frequency driver

 

[cogitech]

I believe a summary would be that loudspeaker drivers just don't have humanly audible objectional low level nonlinearity - they all pass the "first watt" criteria

Perhaps they do, but do they all pass that criteria from 20Hz to 20kHz?

In other words, can they all be put into a suitable enclosure and produce level response from 20Hz to 20kHz at 2.83v input?

I'm being silly, of course (?).

But really, the question I am asking is; can such a driver be produced if the "first watt" is all that is taken into account?

What is the point? Well, considering that the vast majority of listeners spend the vast majority of time listening below 1 watt....

To be clear, I am speaking of "the ultimate full range driver", something which does not exist. I feel that it does not exist because the assumption is that there would be no market for a 1 watt driver.

 

[markaudio]

The "First Watt" amplification approach seems to have been embraced and the fruits of that approach are enjoyed by many. I really do believe that the first watt matters the most.So, I got to thinking. Could the same/similar approach be used in driver design?

So I am wondering;

1) Just what compromises are made by a driver designer in an effort to satisfy the "15 watts is nothing" crowd?

2) What gains could be made if a driver designer set out with the goal of making a driver (or series of drivers) with the widest frequency response as the main priority, even if it meant producing a very "low watt" driver?

3) Allowing for headroom, could a "2 watt", 90dB driver with generous Xmax be designed to produce say 30Hz to 20kHz, +/- 2dB ?

Hi Cogitech, Guys,
Interesting comments, answering your questions:

1) I'm assuming your referring to those who are into "high power" applications, I've talked about some of the issues at some length here (post No 120): http://www.diyaudio.com/forums/markaudio/224477-future-woofer-production-12.html

2) The gains would be centred on the emitting surface substrate; Its ability to carry the micro-resonant signal variance within the main carrier wave, as it passes across the surface of the cone substrate. The key to any/all wide band emittance is the prime mechanical relationship of mass and operational loss. It does however result in power handling limitations.

3) Difficult. Most of you know I'm generally heading in an "efficiency" design direction, but not at the expense of handling LF (low frequency). The traditional method that most (virtually all) genuine Full-Range driver designers (there's only a few of us) use is to increase SPL is to (a) reduce mass and (b) increase magnetic force. However, this primarily results in drivers with little or no LF emittance capacity. I've swam against this tide for several years, firmly in the belief that it is possible to design Full-Range pure single cone drivers capable of both mechanical oscillation and semi-static to static resonance with a wide operational bandwidth. Those members using the latest crop of Markaudio drivers (I hope) are finding my efforts effective. Certainly the key design element to getting a driver to an operational threshold of 2W,90dB + wide Xmax @30Hz-20kHz (+/- 2dB) is the reduction of mass while designing-in wide linearity into compliant suspensions; A real challenge, which I've been working on for some time.

However, there's possibly an even bigger challenge for such a super-douper driver, its operational application. All input signals would have to be super clean and near perfectly balanced against the severe power handling limitations of the driver's power-train. One wiff of an extra watt, its power-train will over-load, distortion will be become audible. I suspect only the most experienced end users could effectively design suitable box systems and operate the driver safely within its limits. I'm glad Markaudio drivers have gone along way to overcome such problems with the introduction of the Arrestor. Certainly the new Alpair 10P will be using this device as the prototypes are proving to be very emittance sensitive, evan under small driven loads while retaining the same Xmax compliant suspension systems as its sister drivers.

Hope this helps

Cheers
Mark.

 

[cogitech]

Mark,

Fantasic reply. Thank you.

It is your excellent drivers, of course, which have inspired me to post this question in the first place. I haven't asked for Fostex or Jordan's opinions on it.

It's really a matter of "how good could it (will it) actually get?"

I would be happy to be one who commits to the self control of limiting my signal output to 1 watt for the ultimate full range driver. For one pair of speakers anyway.

BTW, I did notice you opted to use the word "difficult" rather than "impossible".

 

[cogitech]

BTW, your addition of BSC to drivers is awesome. This type of innovation could be taken further in a "1 watt" speaker by adding just the right amount of "loudness" compensation. In that "first watt" area, we are less sensitive to bass than the rest of the frequencies.

Imagine a "1 watt" driver, with built-in BSC and built-in "loudness" compensation, inside a box designed (by one of our local geniuses) to specifically reproduce an ideal "first watt" listening experience. If the premise is that only 1 watt will be delivered to the system, the design could be that much more precise, I would think.

I suspect that first watt experience would rival just about any pair of speakers out there.

In the meantime, I eagerly anticipate the Alpair 10.3 !

 

[markaudio]

Mark,

Fantasic reply. Thank you.

It is your excellent drivers, of course, which have inspired me to post this question in the first place. I haven't asked for Fostex or Jordan's opinions on it.

It's really a matter of "how good could it (will it) actually get?"

I would be happy to be one who commits to the self control of limiting my signal output to 1 watt for the ultimate full range driver. For one pair of speakers anyway.

BTW, I did notice you opted to use the word "difficult" rather than "impossible".

Glad to help out.
I think its very interesting re other makers. Within the "makers community" here in South China (Guangdong Province), I've raised a few eyebrows over the years. Making mechanically compliant low mass drivers is a risky business. Tolerances on all major components is very tight, handling these items during assembly is difficult, worker skill levels must be good, reject rates remain a little higher than average although I've managed to bring it down to <1% per run of +1000 drivers. Overall, each designer/maker (I'm both) will (or should) bring their own philosophy into their work, so I doubt there will be unanimity, nor should there be. Diverse choice is the nest way to go.

Yes, "difficult" but I'm always interested in an engineering challenge, hence why I like to read about members thinking and receive feedback. This is the prime driving force for the development of Markaudio drivers. Rightly so as your the folks who live with our products.

I think the 10P-Evan just might surprise you (1 watt et al)

Cheers
Mark.

 

[chrisb]

I think the 10P-Evan just might surprise you (1 watt et al)

Cheers
Mark.

can't wait to get my hands on a pair of those - got an ETA yet?

Fedex air direct was awesomely fast on the MAOP7s, which I keep on threatening to resurrect my vinyl rig to fully appreciate.