Why Crossoverless Single-Driver Speakers Are (Mostly) Crap! (Ep:84)

In this video, SoundStage! founder Doug Schneider explains why single-driver speakers do not work well and crossovers are often necessary to build better-sounding speakers. After you watch, feel free to comment as to who you feel is right.

#audiophile #stereo #hifi

[karl-heinzfink]

Life is a compromise, and speakers are no exception.
Crossovers are adding phase problems to a speaker. Whatever you do, a crossover will do that in the crossover area. Even first-order crossovers, which could be better in theory, don’t help because you have to look at the combination of the acoustic behaviour of the driver in the cabinet plus the electrical filter. So, fewer ways mean better phase response. However, it is impossible to make a driver that handles the whole spectrum, and even so, if we could find one, we would have to deal with the effect of the cabinet dimensions. And that needs some eq filtering. So a minimum of 2 drivers is a fine compromise. 
In the old days (before Klippel), drivers often showed changes in inductance over excursion, modulating the connected crossover plus generating IM and harmonic distortion (I worked hard to make a crossover-less speaker when I started with EPOS, but found it did not work well). With this sort of driver, a simple crossover would be beneficial. However, with well-compensated magnet systems that are standard nowadays, the crossover does not hurt. We also have nice tools to design the crossover - no more boxes of components and soldering iron for a good result (OK, after simulation, we need it again).
So we always have to find a compromise. One of my teachers was Ted Jordan, and his full-range drivers could sound very nice. I did some work with BMR drivers that could sound nice as well. But yes, all had some problems, like insufficient energy in the top area or no deep or full bass. So I do 2 and 3-way today and live with the fact that a crossover adds some group delay distortion but gives us other benefits. It’s a compromise.

[rvaillant]

Having built single driver speakers using Fostex 4” drivers I can say they do have some virtues.
1. Is imaging. Better than most speakers.
2. The sense of a live performance.

Cons.
1. The baffle step loss and resulting 6 db loss of lower frequencies below aprox. 700 hz. I fixed it with digital e.q.
2. No low bass. Fixed with subwoofers.
3. Although not readily apparent, grainy highs and lack of high frequency speed. There is no fix unless you crossover to a tweeter and the lower the frequency you can do it and work properly the better.

Bottom line:
A well designed multi way speaker will always outperform a single driver design in most areas.

We should skate some time Doug, ha.

[audio_acoustic_engineering]

Cube audio are very good single driver speakers, and Songer Audio. I’d love to hear the Pearl Acoustic Sibelius, HiFi Cave gave them a rave review. There are many poor and meh single driver speaker, there are fewer great ones.

[meowhsu]

I like my Omega speaker with my deceare setup. It has a more emotion/liviness to the setup and I feel more connected. It might not be good for all kinds of music but I don't think it's easy to find a speaker that's good for all types of music regardless of the design

[gaborozorai]

All design concepts are compromises. Single driver speakers just choose different ones. Those that are well-executed have strengths that are very difficult for multi-driver ones to achieve. Multi-driver speakers try very hard to behave like a single point source with no crossover or phase issues. Singles don't need to try, they just ARE like that. Furthermore a single driver speaker may be simple in terms of components but the design of the driver is anything but. There often is - especially with larger drivers - a deliberately designed flexibility in the cone or between the center, directly attached to the voice coil (whizzer cone or dustcap/dome) and the outer parts of the diaphragm. A mechanical crossover filter essentially.
Anyway, my references being live acoustic music and other (multi-driver) speakers, if I hear good full-ranges to be very good indeed, in many ways, I just don't need anyone telling me that they're crap. It's just childish.

[MaartenVisser]

Maybe you should consider that a single driver speaker needs a completely different design of the enclosure to sound good. I designed a speaker that sounds (and measures!) extremely good with a single driver. It just needs some thinking in stead of calling something crap, and saying that someone who tries to explain something is talking BS. And yes, I also used to build multidriver speakers.

[thecommish]

Having worked for and with some brilliant and creative loudspeaker designers, I absolutely agree. Most of us want simpler to be better, but sometimes it just does not turn out that way.

[eldloppa]

This will ruffle some feathers, I think.

[sonusancti]

Not the full 20-20k perhaps but say 80-14k is definitely common. Many brands have been making single drivers for decades and they would've shut down if they sound crap as you say.

Theoretically, it is an excellent idea. Think of your tiny eardrum, isn't it perfectly full range?

[solomonstewart]

As one gets older one loses some high frequency capability so some of us are more tolerant of the pros of single drivers.

[paulmorris]

I use a 2 watt SET amp and single 6.5" driver speakers (albeit with a sub) because I LIKE the sound. I'm not a recording engineer, I'm not interested in accuracy (to what anyway?). I just want it to sound "nice" and don't give a damn about distortion. Each to their own and all that.

[SonicFlare]

This reveals the corner that "high-end" audio has painted itself into over the past 40-50 years. Facts matter little, measurements matter little and it's all a grand old "subjective" family claiming to have objective authority to defend their positions. Except of course when it all starts to fall apart and you actually start revealing truth and the crowd then predictably goes wild with claims that measurements don't equal sound quality.

[connorduke]

Fully agree. To me, most Dynaudio standmount speakers display a mastery of engineering compromise across all of the engineering parameters I am aware of.

[tyeadel]

Going from 30 Hz to 20 kHz with a reasonably flat response is a big ask. The closest I can recall was the Goodmans axiom 301 in a 250 L (about 8 cu ft) ported box. Now the best solution is probably bi (tri) amping. The sound source in your video with vinyl and moving magnet is highly suspect though.

[Phloored]

My solution is to get all the filtering done before the amplifiers with active crossovers. I bi-ampa two way.
A stack of two heil amt in free space (mounted on a pole) for each side with wave guides will play down to 500htz. Then a 9.5 inch mid bass in a 170 litter transmission line cab will easily pick up from 500htz down to the teens. So bi or tri amping with electronic crossovers gives that direct injection. It's a really really great way to listen.

[ColocasiaCorm]

as a single-driver fan, these flaws and shortcomings were unbeknownst to me

thank you Doug for exposing the truth!

i will convert to a multi-driver box fan now

[uncola]

I tend to think things that are EASY to measure, over time became believed to be more important to performance because they are easy to market. The simpler things like single driver crossoverless design have benefits that are harder to measure, time alignment and other stuff

[IanKnight]

You are spot on Doug. As a professional loudspeaker engineer I agree with you totally. There is so much crap out there misleading people.

[jimmatheos]

If this is even remotely true, why are the top single drivers companies still in business with many of them not even able to keep up with production demands? Many of these single driver companies are out of stock or are backed up in orders. This clearly shows that audiophiles want single driver speakers and will wait long periods to get them rather than buying boring multi driver speakers.

[ColocasiaCorm]

So Doug when’s the cube nenuphar review coming out?