Speakers w/Full-range Drivers

[Chrispy]

The whole one-way is something I know little about and some people really enjoy them so it's interesting that way.

One way is a poor term, full range driver is better.

 

[Doodski]

In a case of synchronicity, I have recently re-activated a much-beloved pair of mass loaded folded, tapered quarter-wave tubes (ML TQWT) enclosures using slightly tweaked Radio Shack 50-1354 twincone drivers. They are quite nice.

The notion of "undersized to purpose" drives is an amusing one -- look at the loudspeakers @amirm has tested over the past few years.

Undersized makes sense in that to put out high frequency and be lightweight a smaller driver is required so a one-way with small light drivers seems to make sense.

 

[mhardy6647]

Photo from 2013.


"before" closeup of the aforementioned ML TQWTs.

After years of meaning to do so, this past weekend I repaired damage to one of the two surrounds, replaced the rope caulk 'gaskets' (what was I thinking?!?! ) with proper gaskets, and 'repaired' the resulting oil stains on the baffles by washing them down with mineral oil, then sealing them with a marvy product from The Real Milk Paint Company called "Half and Half".

https://www.realmilkpaint.com/shop/oils/half-and-half/

Just a couple of comments about so-called "fullrange" (more reasonably, and realistically, "extended range") drivers. Asking for full audio bandwidth, reasonably flat, from one diaphragm and voice coil is a very tall order -- effectively impossible. I've been playing with extended-range drivers for a long, long time, and literally grew up listening to Electrovoice "Wolverine" LS-12 12" twincone drivers in large bass reflex cabinets that my father built in the late 1950s.

Some of them are very enjoyable, some are strident. "We" usually play tricks to enhance the LF output (e.g., back loaded horns, the aforementioned TQWT approach, etc.). Treble extension can be arrived at numerous ways (EDIT: "whizzers", dual compliance surrounds, the already mentioned dome-shaped and/or metal dustcaps, e.g.) -- some implementations work better than others.

Practically speaking, my opinion is the best use of a good "fullrange" driver is as an "extended midrange" driver, augmented in LF and treble with suitable drivers, but keeping the crossover frequencies out of harm's way!

I have mentioned the EV Esquire previously here at ASR. It does not have the LF XO as low nor the tweeter XO as high as I would prefer, but it still makes very good use of the 8" twincone midrange driver (EV Wolverine LS-8) that it employs, ending up with a respectably sensitive, easy to drive loudspeaker in a small-ish package. It was what Klipsch's Heresy could have been had the Colonel not been hell-bent on employing the nasty, shrill, ringy MR horn and cheap Atlas PA driver he used for midrange in the "classic" Heresy models.

 

[kemmler3D]

keeping the crossover frequencies out of harm's way!

Tend to agree... you can't really avoid crossing in the 1-2khz range with most drivers, but you can push it up with a 4" and still be credible down to 100Hz or lower. If you look at the Purifi 4" F10 is 50hz and you can sort of squint and see crossing it above 2khz, but it's usable up to 5khz, so maybe put it in some kind of giant waveguide and cross it higher...

 

[Heinrich]

Photo from 2013.


"before" closeup of the aforementioned ML TQWTs.

After years of meaning to do so, this past weekend I repaired damage to one of the two surrounds, replaced the rope caulk 'gaskets' (what was I thinking?!?! ) with proper gaskets, and 'repaired' the resulting oil stains on the baffles by washing them down with mineral oil, then sealing them with a marvy product from The Real Milk Paint Company called "Half and Half".

https://www.realmilkpaint.com/shop/oils/half-and-half/

View attachment 419470





Just a couple of comments about so-called "fullrange" (more reasonably, and realistically, "extended range") drivers. Asking for full audio bandwidth, reasonably flat, from one diaphragm and voice coil is a very tall order -- effectively impossible. I've been playing with extended-range drivers for a long, long time, and literally grew up listening to Electrovoice "Wolverine" LS-12 12" twincone drivers in large bass reflex cabinets that my father built in the late 1950s.

Some of them are very enjoyable, some are strident. "We" usually play tricks to enhance the LF output (e.g., back loaded horns, the aforementioned TQWT approach, etc.). Treble extension can be arrived at numerous ways -- some implementations work better than others.

Practically speaking, my opinion is the best use of a good "fullrange" driver is as an "extended midrange" driver, augmented in LF and treble with suitable drivers, but keeping the crossover frequencies out of harm's way!

I have mentioned the EV Esquire previously here at ASR. It does not have the LF XO as low nor the tweeter XO as high as I would prefer, but it still makes very good use of the 8" twincone midrange driver (EV Wolverine LS-8) that it employs, ending up with a respectably sensitive, easy to drive loudspeaker in a small-ish package. It was what Klipsch's Heresy could have been had the Colonel not been hell-bent on employing the nasty, shrill, ringy MR horn and cheap Atlas PA driver he used for midrange in the "classic" Heresy models.

I‘ve not seen a thourough discussion of the crossover issue yet. In consquence no reason was given why a single driver is better. It seems to be a sentimental thing mostly, which would be so far so good.

Likewise it could exemplify on the hypothesis, that an essential topic with audio is the trust the listener invests into believing it. Once there is trust and belief the attention goes to the music, mission accomplished. Could be a strong idealistic expectation, or a well taken parcours of fine measurements, or just tremendous expenses. All do not actually need to realize, just imagine it would, and switch focus to the simple purpose of that nasty thingy.

 

[Doodski]

I‘ve not seen a thourough discussion of the crossover issue yet. In consquence no reason was given why a single driver is better. It seems to be a sentimental thing mostly, which would be so far so good.

Somewhere back in this thread I thought it was said that the seamless integration of a full range driver is better in ways. I can understand that although I am inclined to thinking full range = lacking stuff.

 

[Everett T]

New member here, though long-time beneficiary of the forum, particularly DAC reviews (I have a Topping E70 and Okto DAC8 in two of my systems).

Anyway, with regard to speakers, over the last few years I’ve increasingly been drawn to the sound of crossover-less (or minimalist) designs by companies like Fern & Roby, Rethm, Voxativ, Songer, etc. I know there are many on this site who believe that such a design is fundamentally flawed, or limiting, such that it is unlikely to ever compete with a well-designed two- or three-way speaker.

The most obvious challenge to a design using a full-range driver is at the frequency extremes. But some of these companies using full-range drivers have come up with interesting approaches for addressing this challenge. Rethm, for example, essentially uses a powered subwoofer - 4 drivers in an “isobaric configuration” to fill in the low-end - but of course most of these companies do not provide detailed specifications on their speakers’ performance.

So finally to my point: At Capital AudioFest this year I heard a pair of the Audience 1+1 speakers and found them remarkable. They use two full range drivers and two passive radiators in each small box. But here’s the kicker - unlike some of the other examples I’ve heard, Audience publishes info that includes some detail regarding their frequency response. I’d be very interested in others’ views on this, as it appears as if they have accomplished a remarkably flat and extended frequency response with these little speakers:

Audience white paper

I appreciate everybody’s patience with such a long, first post!

Mark Audio has some excellent value drivers to research. The top octave won't be an issue these days as .ich but you may want to consider an active low frequency driver setup.

 

[mhardy6647]

The Foster/Fostex FE-103 4" "fullrange" driver has been in production in various morphs since the early 1960s, with only minor "improvements" (changes). The biggest one was perhaps going from a large AlNiCo magnet to a nominally equivalent ceramic magnet many years ago.
It is an OK driver within its passband, although suffering from a bit of the typical (but not universal) Fostex "fullrange" driver "shouty-ness".
Gazillions of these have been sold under numerous brand names and catalog numbers. In the US, they Radio Shack 40-1197 was/is probably the best known.

I mention this mostly due to this well-known Electronics Illustrated article from Nov. 1965 with a design for an implementation of this driver in a small box including - for better or worse - a somewhat rudimentary contour filter (meant as a BSC, at least I presume). The plans were included with the purchase of a 40-1197 from R/S for at least several years.

https://www.worldradiohistory.com/Archive-Electronics-Illustrated/Electronics-Illustrated-1965-11.pdf

I actually have a mismatched "pair" of these, found (separately) at the good ol' Harvard, MA town "transfer station" ("dump", although not actually a dump per se) decades ago now. One of them is missing its back, and thus its coil and capacitor network, too.

 

[Will33]

All this and no mention of Eclipse yet?

 

[kemmler3D]

All this and no mention of Eclipse yet?

Interesting. I was aware that the teardrop shape was considered better than sphere and I have a vague thought of building some like this.

What's kind of cool is it seems they deal with beaming by having a really tall dustcap that I guess acts like a phase plug. $2K each for the floorstanders is a lot, but I have certainly seen worse...

 

[Ktacos]

nor do they have major dispersion issues

Idk most of the ones I tried had pretty whack dispersion, tc9fd, a few mark audio, definitely nothing close to a good multiway.

One thing I don't like about them is the really narrow top end, it seems to make the HF information sound really tiny and cheap.

 

[Selkirks]

One thing I don't like about them is the really narrow top end, it seems to make the HF information sound really tiny and cheap.

You had this issue with Markaudio drivers on-axis?

 

[kemmler3D]

One thing I don't like about them is the really narrow top end,

I think this is unavoidable with cones and gets worse as they get bigger.

 

[thewas]

Many wide range divers exhibit pretty low distortion figures in the mids

Which don't necessarily correlate though to audibility as multitone distortions usually rise when using a driver that plays mids and highs also to reproduce bass and is together with the directivity problems of the most the reasons why full band driver usage is a strong compromise unless for special listening cases like for example small listeing distances and levels.

 

[mdalton]

So I guess I’m wondering if standard measurement techniques capture everything there is to understand about what makes a full-range driver sound different. As I understand it, and as has been alluded to by others, the absence of a crossover in a critical frequency range seems to be the most obvious, salient feature. So what should we look at when we measure a 3-way with crossover points at 250 and 2k? Proponents of full-range drivers talk about the benefits of eliminating the comb effect and phase issues, dynamics not being sacrificed by a crossover, the tonal consistency of one driver, etc. And I think I can hear a difference with full-range drivers - I would characterize them as being more “immediate”, though I realize that’s not a very ASR-like description (and no, I haven’t done a double-blind). But I would analogize the experience to hearing live music. I first heard the Audience speakers from the hallway outside the room, and thought I could hear something special, more natural, about their sound. I wasn’t even planning on going into the room originally, it was just uncanny. So back to my live music analogy: it’s really easy to distinguish live music from recorded music, even from another room. Can we measure that? I honestly don’t know, but would love others’ thoughts.

 

[kemmler3D]

So I guess I’m wondering if standard measurement techniques capture everything there is to understand about what makes a full-range driver sound different. As I understand it, and as has been alluded to by others, the absence of a crossover in a critical frequency range seems to be the most obvious, salient feature. So what should we look at when we measure a 3-way with crossover points at 250 and 2k? Proponents of full-range drivers talk about the benefits of eliminating the comb effect and phase issues, dynamics not being sacrificed by a crossover, the tonal consistency of one driver, etc. And I think I can hear a difference with full-range drivers - I would characterize them as being more “immediate”, though I realize that’s not a very ASR-like description (and no, I haven’t done a double-blind). But I would analogize the experience to hearing live music. I first heard the Audience speakers from the hallway outside the room, and thought I could hear something special, more natural, about their sound. I wasn’t even planning on going into the room originally, it was just uncanny. So back to my live music analogy: it’s really easy to distinguish live music from recorded music, even from another room. Can we measure that? I honestly don’t know, but would love others’ thoughts.

I think if you wanted to visualise the differences, an un-smoothed frequency response and phase response graph could show what you're talking about.

I usually assume the "sounds live from outside the room" effect is due to radiation pattern and probably higher dynamic range. But it would be cool to dig deeper into the full range mystique.

 

[MaxwellsEq]

So I guess I’m wondering if standard measurement techniques capture everything there is to understand about what makes a full-range driver sound different

Just so you are forewarned, this sounds a bit like a classic strawman that people introduce and then get an unexpectedly negative response - "measurements don't tell us everything". ASR is not the easiest place to trial this line of thought. Mostly, you will be given a suggestion to read: Floyd Toole's "Sound Reproduction: The Acoustics and Psychoacoustics of Loudspeakers and Rooms". Or you will be asked to identify the gap in current measurement theory and how it should be fixed!

 

[kemmler3D]

Just so you are forewarned, this sounds a bit like a classic strawman that people introduce and then get an unexpectedly negative response - "measurements don't tell us everything". ASR is not the easiest place to trial this line of thought. Mostly, you will be given a suggestion to read: Floyd Toole's "Sound Reproduction: The Acoustics and Psychoacoustics of Loudspeakers and Rooms". Or you will be asked to identify the gap in current measurement theory and how it should be fixed!

I think a good way to think about this is "what measurements could explain what people seem to find compelling about single-driver speakers"... if we think comb filtering is the culprit as @mdalton proposes, it's not hard to measure.

I agree that it sounds like the typical "what are we missing about amps" tomfoolery but I don't think this is that kind of question.

 

[mdalton]

Yes, I want to be very respectful of the forum and its members, and I am intensely aware of my own technical limitations. Having said that, what I found interesting about the Audience speakers is that the designer has taken the trouble to publish on his website a 10 page(!) white paper that includes, among other things, what appears to these untrained eyes as a very smooth frequency response (page 2), and a step response graph (page 9) with what appear to be relatively comprehensive explanations of the measuring techniques. Given that some folks here - including Amir - have done a significant amount of thinking (and doing) on speaker measurements, I’m curious how the Audience measurements measure up (sorry for pun), and whether folks think those measurements are sufficient to capture fully what’s going on with full-range drivers. Full disclosure: my intuition leans towards “no”, but I acknowledge that intuition to be relatively uninformed, hence my posts.

 

[kemmler3D]

Yes, I want to be very respectful of the forum and its members, and I am intensely aware of my own technical limitations. Having said that, what I found interesting about the Audience speakers is that the designer has taken the trouble to publish on his website a 10 page(!) white paper that includes, among other things, what appears to these untrained eyes as a very smooth frequency response (page 2), and a step response graph (page 9) with what appear to be relatively comprehensive explanations of the measuring techniques. Given that some folks here - including Amir - have done a significant amount of thinking (and doing) on speaker measurements, I’m curious how the Audience measurements measure up (sorry for pun), and whether folks think those measurements are sufficient to capture fully what’s going on with full-range drivers. Full disclosure: my intuition leans towards “no”, but I acknowledge that intuition to be relatively uninformed, hence my posts.

To your point about comb filtering, the measurements provided by Audience would not show that at all. Comb filtering shows up as tightly spaced, narrow dips and peaks in the frequency response. Usually it's filtered out either way by smoothing the graph, because it's not considered to be very audible in practice.

However, if you took a measurement of the Audience speaker and it showed a lot less comb filtering than a comparable 2-way speaker, that would be a hint that comb filtering could explain some of the subjective impressions people have of single-driver speakers.

This would actually not be hard to do at home, you could just do a windowed nearfield measurement of two speakers and see what you get. No special equipment beyond a UMIK or similar would be needed.