[dhwilkin]

I'm pretty familiar w/ dynamic and electrostatic speaker characteristics, thanks to headphones. But what about the more exotic speaker types, like ribbons, horns, and line sources? What are generally accepted strengths and weaknesses of each type? What specific models of each type are usually accepted as examples of speakers of that type done right? Was just thinking about future directions for my basic HT setup (or 2-channel setup, whenever I can afford a place w/ a dedicated listening room), and got curious about these other types.

 

[ooheadsoo]

Ribbons are electrostats, so you know the inherent qualities of them. I think, in addition to the ones you know, some ribbons have much wider dispersion than normal tweeters, creating a large sweetspot, but sometimes giving up the most absolute pinpoint image.

Horns are efficient, like megaphones. There are rear loaded horns and front loaded horns. Front loaded horns like the klipsch horns are extremely efficient. What it does is allow you to enjoy the best triode tube amplification at real life volume levels and dynamics. What you give up is neutrality and sweetspot size (not to mention listening room real estate.) The horn amplifies certain frequencies more than others, so there's what they sometimes call the "shout" and the horn also restricts the dispersion so beaming effects sometimes occur. Rear loaded horns are a different story. The horn is inside the cabinet and is used to smoothly extend the bass. It's basically a special transmission line design.

Line arrays are also very very efficient because of the multiple drivers. The strengths of line arrays are incredible imaging with sweetspots that extend pretty much to the speakers. (I've heard this with waveguide mounted tweeters too, like the mackie hr824 and 624.) The center image stays in the center until you walk beyond the speakers. Also, you can stand and still enjoy the frequency balance and imaging as when you are sitting, providing you are short enough/the array is tall enough. The dynamics are probably not as good as horns most likely because it doesn't have the megaphone shout effect. Also, sound decays at half the rate in a line array arrangement when compared to regular speakers so...I'm not sure the benefits of that. Certainly larger sweetspot, able to listen at lower volumes...stuff like that. I don't really know of any cons to line arrays except that they're really big and tall and probably have extraordinarily low WAF, plus they are expensive.

 

[tomek]

i understand how a sweetspot can be wide in terms of tonality, but how can it be wide in terms of imaging?

i thought that imaging depended on the sound getting to your ears at exactly the right time. that's why large speakers don't image well, reflections off the baffle. how can line arrays image so well that you can walk around the room?

i know with my speakers they sound great anywhere, but if i move on the couch then the images appear to shift over.

 

[halcyon]

Rather than talk about element types or configurations, I'd like to widen the scope to dispersion patterns:

- true omnipolar
- typical semi-controlled directivity
- uniform constant directivity

These have totally different room interations and thus, sound.

Mirages are a good example of omnipolar sound.

Most 2 and 3 way dynamic speakers with no waveguides or enclosure tricks (like open dipole design) fall into the semi-controlled directivity camp. This is perhaps 90% or more of the market.

Uniform directivity camp is where electrostatic panels, ribbon line arrays and some other speaker types (dipole Gradient, hypercardioid Amphion, some line-arrays) fall into.

This last group is my favourite group of speakers, because it minimises many of the detrimental room interactions, if done right.

In practise this means: better soundstaging (what's on the recording), less coloured sound, more accurate imaging/positioning and less room dependent sound. If done rigth with dynamic elements, the sweet spot can also be bigger and the sound doesn't become coloured when you move off-axis (as it does with almost all semi-controlled directivity designs).

Now, you can do all of these type with various element types, although some obvious restrictions do apply (omnipolar with electrostatic panels is a bit difficult for example). One can also use various element configurations and even mix and match element types.

However, the dispersion pattern type is a major contributor to the sound of speakers, once the basic characteristics of the speakers have been done right (like they are in most of the commercial speakers).

This is not just purely IMHO, btw. There are a few interesting AES papers on directivity issues, if anybody wants to take a look into this issue.

best regards,
Halcyon

PS It's perhaps no surprise that many if not most studio monitors are already or are becoming basic constant directivity designs with waveguides (e.g. Genelec, Klein+Hummel, Mackie, etc).

 

[ooheadsoo]

Quote:

Originally Posted by tomek i understand how a sweetspot can be wide in terms of tonality, but how can it be wide in terms of imaging? i thought that imaging depended on the sound getting to your ears at exactly the right time. that's why large speakers don't image well, reflections off the baffle. how can line arrays image so well that you can walk around the room? i know with my speakers they sound great anywhere, but if i move on the couch then the images appear to shift over.

The sweetspot for me also encompasses the zone where you get proper imaging. Things in the center stay in the center, left to the left, etc. With my speakers, I get the same imaging problem you do. With the line arrays and wave guide, etc speakers, even when you're standing directly in front of the left speaker, Norah's voice is still coming from the center between the two speakers, etc. Halcyon's made a good post on dispersion characteristics. I haven't read up on that like I should. Baffle size is not the only characteristic that affects imaging. If it were, all cheap computer speakers would image like the devil! They don't

Wave guide designs seem hard to voice well. I wonder why they don't any of them use ribbons.

 

[tomek]

are waveguides used in any 'audiophile' speakers?

so many manufacturers claim to have researched the best design and try to distinguish themselves with some sort of innovation but I don't think I've seen this before.

I wonder if the tweeter used in my speakers had this in mind.

http://www.usspeaker.com/beyma%20t2030-1.htm

 

[ooheadsoo]

Absolutely there are!

http://www.4sptech.com/

Some studios do use these speakers as monitors though

Your tweeters have the basic idea of the wave guide.

 

[Calanctus]

Quote:

Originally Posted by ooheadsoo Ribbons are electrostats, so you know the inherent qualities of them. I think, in addition to the ones you know, some ribbons have much wider dispersion than normal tweeters, creating a large sweetspot, but sometimes giving up the most absolute pinpoint image.

Can you clarify what you mean by saying that 'ribbons are electrostats'? Typically, ribbons and electrostats are considered different. Electrostats use a thin mylar (or similar) membrane, while ribbons use a strip of metal. Electrostats are typically far wider than ribbons--e.g. Sound Lab electrostats which are 2-3 feet across (and 6-7 feet tall), far wider than any ribbon.

Also, their sound qualities are similar but not identical. Ribbon tweeters typically have a much more extended high frequency response than electrostats--e.g. 40,000 KHz at -3 db on my ribbon tweeters, versus 20,000 to 27,000 at -3db for most commercial electrostats.

They do share the qualities of a very transparent midrange, with fast transient response.

Also, there are what are called 'planar magnetic' drivers which share some of the qualities of ribbons, but not all. As I understand it, planar magnetics are fixed on the sides and top, while true ribbon drivers are fixed only at the ends (top and bottom).

 

[ooheadsoo]

What you say is very true. But I consider these direct drive systems as a group fundamentally different from dynamic drivers with voice coils. There are of course many different types of these systems.

 

[halcyon]

I think the reason why many consumers in USA have a bad connotation with waveguide is due to BOSE and their marketing.

However, let me assure you that proper waveguide application is not only acoustically, but also perceptually more accurate than not using them. Bose is just one company mainly using the term in their marketing and not really implementing them properly.

Waveguides are used in audiophile speakers as well, but only by those who understand their merits and can design around their possible pitfalls.

All horn loudspeakers are of course with waveguide (the horn is the waveguide). One of my favourites is the Avantgarde Duo 2.2.

The coaxial elements are also partial waveguides themselves, although not perhaps the best in terms of shape (contour). See Tannoy and KEF for some examples of this type of element as a waveguide for another element.

Then there are other makers, who use waveguides that they've calculated/fitted themselves (Amphion, Chorus, Zingali, etc.).

More waveguides are starting to get used in even the highest quality high tweeters because it makes sense (BG, Dynaudio, Scanspeak, etc.). Of course, in most of these the waveguide is relatively shallow and the dispersion cannot be tailored.

However, waveguides are not the only way to control dispersion nor are they enough for dynamic done types. The enclosure directivity must also be considered. Two examples of this are the open baffle (leads to dipole directivity pattern in the bass region, as used by Gradient) and controlled resistance enclosure (can give hypercardioid dispersion pattern if designed properly). In addition, with dynamic elements the vertical directivity is also often controlled with MTM element configuration.

Sorry if I ventured too much off the topic.

 

[tomek]

i love reading your posts halcyon.

although they do worry me sometimes and i question the speakers i purchased.

 

[darkclouds]

Quote:

Originally Posted by tomek i understand how a sweetspot can be wide in terms of tonality, but how can it be wide in terms of imaging? i thought that imaging depended on the sound getting to your ears at exactly the right time. that's why large speakers don't image well, reflections off the baffle. how can line arrays image so well that you can walk around the room? i know with my speakers they sound great anywhere, but if i move on the couch then the images appear to shift over.

Tomek, I found that if I toe in my speakers just right, the image is more stable and doesn't seem to shift much even when I'm beyond the boundary of either speakers.

 

[edstrelow]

The most exotic speaker I have seen was an ionophone, basically a system to create a plasma in the air and then to dirve the plasma as the speaker source. the advantage would appear to be a "diaphragm" weighing only about as much as air istelf. The unit I saw had a big pressurized cylinder attached. Alas I never heard it in operation.

I just checked this out on GOOGLE and see that the concept is quite old but I don't see any current models out.

 

[m0nKeY bUsiNeSs]

Quote:

Originally Posted by edstrelow but I don't see any current models out.

Acutally i belive Acapella has some speakers out with plasma ion tweeters
Here is there homepage.
You can find a review of there speaker atEnjoy the Music

 

[ooheadsoo]

Yeah, halcyon's like an encyclopedia! You could run a website and archive all your knowledge. It would be a great help.