Quote:
Single driver speakers are not for everyone. I want to make this clear at this point, because the last thing I want to do is have a pair of speakers returned as "unsuitable".

Single driver speakers are most suitable for the more intimate genre and at moderate volume levels. This includes solo voice, small group jazz and 18th century classical and chamber music. For these genre, single driver speakers are superb. You will hear nothing better. They also do classic rock pretty well as long as the volume is kept to reasonable levels. By reasonable, I mean ~80dB at the listening position. (I set my LT-2000/DX3's 8 feet apart and my chair 8 feet back.) Now, if you are into heavy metal or want Mahler symphonies at concert volume, I suggest that you look elsewhere. You will be disappointed and I have to eat shipping costs and try to unload a custom pair of speakers at as little loss as possible.

Why Single Driver Speakers?

Single driver speakers have a unique sound. That sound comes from a lack of a conventional crossover. In a two-way speaker, the acoustic phase of each driver rotates in opposite directions around the crossover point. While the combined phase may be correct at the crossover point, away from that point, the phase will be constantly changing. Why is this important? The human ear/brain is designed to locate sounds primarily by phase. We are very sensitive to phase in the 300-3000Hz range. Outside on this range, sound location becomes increasingly difficult, but within this range, if the phasing is messed up, the sound stage lacks precision and depth. A normal two-way speaker has the crossover point in the middle of 1500-2000Hz range, which is smack in the middle of the frequency band where humans are most sensitive to phase changes. By not having a crossover and the attendant phase problems, a single driver speaker sounds much more natural.

A good single driver speaker is more articulate and detailed in the bass and midrange that a multi-way speaker. Because the driver in a singe driver speaker must work up to 10kHz and beyond, the cone is much lighter than the cone of a comparably sized driver in a multi-way speaker that is crossed out by 2kHz. Less cone mass means faster response to the electrical signal and better transient response. The lighter cones also promote higher efficiency. While my speakers don't qualify as "high-efficiency", all except the FTA-2000 are rated at 92dB/w/m or better, making them good matches to all but the smallest tube amplifiers. (Don't worry, they work great with high-powered solid-state amplifiers too.)

http://www.geocities.com/rbrines1/

The key to SDFR systems is coherence. If that means little or nothing to you, then go elsewhere for your music. Some listeners are sensitive to timbre and timing while others are not. Some are confounded by the sound of conventional ported systems while others are not. IMO you have to decide what it is you are trying to eliminate from your music before you are a good candidate for a SDFR. You might find a two way system with a very simple HP filter alone to be closer to satisfactory when compared to a SDFR with the two way offering slightly greater (measured) high frequency extension and almost certainly a wider sweet spot for listening. However, the phasing issues mentioned on Bob's web page will always exist with any crossover and it requires a designer who is thinking outside of the normal conventions of consumer loudspeaker design to make a simple two way system work well.

Beginning explorations of SDFR's need not be expensive though the best of the bunch can represent a substantial investment given the conflicting requirements placed upon a SDFR.

Don't limit yourself to the Fostex line, there are more than a few SDFR's to choose from (http://www.audience-av.com/parts/A3.php). Do some research before investing in any driver. This is a particularly nice SDFR that I have heard in several systems but it sacrifices sensitivity for smoothness (http://www.creativesound.ca/details.php?model=FR125SR). The size of the driver will be one consideration with obvious trade offs as you go between smaller and larger diameters.

Sensitivity is another consideration and currently many SDFR's are being built to suit the low powered SET and T-amp crowd (which also suits the character of a SDFR quite well - better IMO than a solid state, high powered amplifier would in most cases). However, if high sensitivity is not a requirement for your system, then you can trade off some other qualities by running a slightly lower spec'd unit. No matter what sensitivity spec you choose the efficiency of the system is generally higher than that of a two way since no energy is lost in the crossover components. System impedance remains quite constant in a SDFR so you are unlikely to stress the amplifier and that high current transistor amplifier will be out of place with a SDFR.

Most of the lower cost SDFR's (Fostex in particular) will show a rising midrange response which you must deal with in some manner. Some listeners choose to insert a baffle step correction network (http://search.yahoo.com/search?ei=utf-8&fr=slv8-hptb5&p=baffle%20step%20correction%20network&type=) while others simply deal with the position of the speaker within the listening environment or by manipulating the baffle dimensions. This approach would lead you to the possibility of an open baffle SDFR (http://search.yahoo.com/search?ei=utf-8&fr=slv8-hptb5&p=open%20baffle%20loudspeakers%20diy&type=).

If you should decide to try SDFR's, I would sugegst you not do so in the Bose cabinet. You can build or buy a cabinet for a SDFR for minimal dollars using minimal skills and tools that will be more to the liking of the driver and not add any coloration from a cheap Bose box. Plans for numerous enclosure types are available on line. It is not uncommon to see a SDFR paired with either a (modified) tansmission line or back loaded horn enclosure both of which can be used to extend the low frequency response of the driver.

Pay attention to the links on this page;
http://www.fostexspeakers.com/fostex.html

http://audioroundtable.com/forum/index.php?t=thread&frm_id=13

http://www.t-linespeakers.org/design/classic.html

http://www.madisound.com/kits/index.php

http://www.quarter-wave.com/Gallery/Gallery.html

Dave Merrill designed and had custom built for him some of the best SDFR's by anyone's expectations, certainly for the price they killed the Fostex line. Dave passed away not long ago so the availability of his drivers is in question;
http://diyparadise.com/shop/index.php?main_page=product_info&cPath=4&products_id=101

http://www.timaliforman.com/13.html

http://www.tnt-audio.com/clinica/bfb_e.html

Pay attention to the Brines systems, the Maxxhorns, Hawthorne (http://www.positive-feedback.com/Issue31/hawthorne.htm) and John Busch's room which offered a slightly different take on SDFR's - http://positive-feedback.com/Issue37/lonestar_08.htm

One of my favorites; http://www.passdiy.com/pdf/KleinHorn.pdf

You can get on the DIY forums to do more research or ask questions, http://www.diyaudio.com/. Be aware this is a rather heavily moderated forum.

With a single full range driver you always have the same speed and sound character at all frequencies. It maintains the same tonality at different distances and listening levels.

Artur wrote:

With a single full range driver you always have the same speed and sound character at all frequencies. It maintains the same tonality at different distances and listening levels.

No. It does not mean that at all.

A single driver will almost always have different characteristics at different frequencies, which is the reason for using multiple drivers.

A phased array, such as is used in most Vandersteen speakers, solves most of the problems of time/phase alignment.

commsysman wrote:

A phased array, such as is used in most Vandersteen speakers, solves most of the problems of time/phase alignment.

I love Vandersteens. But this just simply isn't the case. It's an improvement, but it's not the same. Closer to solving the x-over/phase issue it the Tannoy, but it's still not as coherent as a large panel, such as a Martin Logan. That said, the ML needs more powerful, usually PP Solid State amplification, that often negates the coherence advantage. Usually the best solution is a good full range single driver with a mounted, meaning movable and rearrangeable super tweeter. Which for the most part, I believe, means DIY.

in addition, some companies are working with crossover-less designs for the woofers, supposedly also reducing the time-phase alignment issues. Totem does that with their Torrent drivers. I am aware Totem's are not everyone's taste, the highs are too sharp for some that like a more relaxed sound - but i tried them and was hooked. i find them very engaging and entertaining, and the Torrent driver must be part of the reason.

One of the main problems with the single driver loudspeaker concept is extreme beaming at frequencies greater than the driver's physical dimensions... Floor-to-ceiling line sources and large curved panels can reduce the beaming, but at the expense of size, inconvenience and cost.

With multiple driver systems, the diversity of electro-acoustic transducer centers in space and frequencies causes the complete destruction of the musical signal image integrity and the stability of virtual sources in the front, depth and height of the soundstage.

Yes, it does mean that...when we are talking about a Single Real Full Range driver that covers the entire frequency range from 20Hz to 20kHz.
The reason for using multiple drivers is the difficulties of designing and producing a single transducer that reproduces all the frequencies.
But, systems with this type of drivers are available in the market today. Finally.
It will be hard to go back to multi-driver speakers after listening to a "proper" Single Real Full Range driver system.

It seems that everyone agrees that single full-range drivers will beam unacceptably at some frequencies, and that multi-driver speakers have phase and imaging problems at the crossover point(s), so neither is perfect. Is that correct..?

ejchr2 wrote:

It seems that everyone agrees that single full-range drivers will beam unacceptably at some frequencies, and that multi-driver speakers have phase and imaging problems at the crossover point(s), so neither is perfect. Is that correct..?

What you said is correct about multi-driver speakers only. There are some Single Real Full Range dynamic drivers (even 15") with almost flat frequency response (+/-2dB) from 20Hz to 20kHz and with impulse response better that of electrostatic speakers, but without any beaming.
Welcome to the 21st century!

ejchr2 wrote:

It seems that everyone agrees that single full-range drivers will beam unacceptably at some frequencies, and that multi-driver speakers have phase and imaging problems at the crossover point(s), so neither is perfect. Is that correct..?

I suspect single driver enthusiasts do not consider the beaming to be unacceptable at high frequencies. I also suspect that most people do not consider competently designed multi-driver speakers to have substantial phase and imaging problems at crossover. Where I think there is agreement is that neither is perfect although one of them is sufficiently far from perfect to have little to do with high fidelity sound.

Could you give us the makes and models of the up to 15" dynamic drivers that you mentioned that operate "...without any beaming"?

ejchr2 wrote:

Could you give us the makes and models of the up to 15" dynamic drivers that you mentioned that operate "...without any beaming"?

http://www.stereophile.com/content/taves-2014-part-2

Hmmm... I am beginning to wonder what relationship some of the posters in this thread might have with commercial speaker companies.

andy_ You are right. Direct relationship. Don’t want to hide behind another name to post comments here. I just want other audio enthusiasts like me to know more about new groundbreaking technology. Sorry if I hurt anyone’s feeling.

http://www.stereophile.com/content/r2r-audio-real-full-range-self-powered-15-inch-loudspeaker-system

Does anyone here know what happens to the crossover problem in a two way speaker when the jumpers are removed from the speaker inputs and separate power amp outputs are connected to each speaker terminal?I have Polk RTia5's with removable jumpers on the input terminals and an amplifier with separate power amp circuits in it which can be connected discreetly to different speaker drivers.

Not much would happen, actually. You would still be going thru the passive crossover inside the speaker...