Line Arrays

[johngalt47]

I heard some line arrays for the first time in my life. I was disappointed with the diffuse sound field. I have two way speakers that have pinpoint imaging and so was disconcerted with what I head from the arrays.

So, my question is, are line arrays supposed to have good imaging?

[billfitzmaurice]

Re: Line Arrays

So, my question is, are line arrays supposed to have good imaging?

No more or less than any other configuration. But as with any alignment you have to get it right.

[ewetho]

Re: Line Arrays

I have built Bill's SLAs (short line arrays) both the verticals and curved baffle center channel and imaging is fantastic.

Thanks Bill.

Parts are available from here except the box yo have to build.

I have also built larger derivatives and those are very good too.

[Sydney]

Re: Line Arrays

I heard some line arrays for the first time in my life. I was disappointed with the diffuse sound field. I have two way speakers that have pinpoint imaging and so was disconcerted with what I head from the arrays.

So, my question is, are line arrays supposed to have good imaging?

You didn't mention in what context ( setting, distance, and specifics etc )...
With multiple drivers with overlapping propagation the interaction has to be dealt with.
Line arrays that are not implemented correctly can have a "diffuse sound field" and a very hot center lobe that gets tighter with increasing frequency.
FWIW: On several occasions Mr fitzmaurice mentioned: that of all his designs; his Davids with a MTM top provided the best imaging.
A single HF source.

[r-carpenter]

Re: Line Arrays

I've had line arrays in my shop for a couple of years. Build a dozen versions and about to start a final "kill them all" monster "Alice". They do image differently but I wouldn't call it diffused, more "life like". In my experience line arrays provided superb imaging to a conventional semi point source loudspeakers.
having said all that, I'd go back to what Bill mentioned.

[Sydney]

Re: Line Arrays

I've had line arrays in my shop for a couple of years. Build a dozen versions ...

As I recall ( from prior threads ) your implementation is considerable different from that of the BF implementation.
That is why I wondered about the specifics: orienting drivers along an axis makes it a line by definition, but the details of implementation is a more important matter and makes an enormous difference.

[r-carpenter]

Re: Line Arrays

As I recall ( from prior threads ) you're implementation was different from that of the BF implementation.
That is why I wondered about the specifics: orienting drivers along an axis makes it a line by definition, but the details of implementation is a more important matter and makes an enormous difference.

yes, agreed.
I was trying to answer a general question as to how Line arrays (and I suspect we'll be going in to line sources) image.
Because there are few types of arrays, the implementation and executions are of cause more important.
Original question was very broad and can be equated to: I've heard this MTM and imaging was very bad, do all MTMs have bad imaging?"

[Sydney]

Re: Line Arrays

Original question was very broad:

Sure enough:
I'm guessing your building path was one of improvement and refinement with each version, perhaps even uniquely custom tailored to your particular listening area.
It would be hard to generalize when the implementations can be very simple: X number of drivers wired for safe amp load, to complex: A floor to ceiling line or using the various means of physical and/or electrically tapering the individual drivers or segments.

Anecdotally: When the Peavey Versarray was publicized heavily as a lower cost entry into the LA type systems used in touring - the first setup I heard was dreadful.
According to the owner of the system it was the fault of their employee for poor setup.

[r-carpenter]

Re: Line Arrays

Sure enough:
I'm guessing your building path was one of improvement and refinement with each version, perhaps even uniquely custom tailored to your particular listening area.
It would be hard to generalize when the implementations can be very simple: X number of drivers wired for safe amp load, to complex: A floor to ceiling line or using the various means of physical and/or electrically tapering the individual drivers or segments.

For me it was more trial and error.
It was like: Let's build a line of 6 drivers and listen. Throw em on a wide baffle. Add a tweeter. Increase to an array of 8 then 9 then 12. Cut the baffle in to 3 sections and try to focus it. Measurements were taking place as well but it's a bit difficult with an 8 ft tall speaker having multiple drivers.
At this point I kind of narrowed the desirable characteristics.
Modeling helps (Edge) and confirms quite a bit of subjective impressions as well.

[John L. Murphy]

Re: Line Arrays

The "magic" happens when you run a line array between parallel floor and ceiling. The floor and ceiling reflections create the perfect acoustic illusion of an infinite line array just as two parallel mirrors would. This places the listener in the near-field of the array with all the benefits of near-field behavior. This includes constant vertical directivity with consistent frequency response from floor to ceiling. Readers should be careful to distinguish this infinite line array system from truncated far-field line array systems designed for concert sound.

Here is how Floyd Toole depicts it in figure 18.2 of his excellent book "Sound Reproduction" where he shows just the first reflection:



Because the vertical directivity is constant in the near-field of an infinite line array the vertical imaging changes from that of a point source. Unlike a point source, the image from the infinite line array is always located at ear height, that is, normal to (90 degrees to) the array. Some listeners report that this lends to the impression of a "tall" image but my own personal experience is that the image has just the right height and is overall more convincing than the vertically flattened image projected by point sources. The horizontal directivity of an infinite line array is the same as that of the individual array components. So you tend to get "normal" imaging in the horizontal plane with "non-flattened", imaging in the vertical plane. When you sit-down or stand-up the sound of the infinite line array is unchanged. You can even walk about the room and experience a very convincing image of the performers.

For an infinite line array system it is not appropriate to taper the array in any way. Tapering is only appropriate for truncated line sources used in the far-field in free space (i.e., not in rooms). You only need to consider the reflected images to realize that tapering applied to the real array would be inappropriate. With an infinite line array setup the reflections are already naturally tapered as each generation of reflection is progressively attenuated according to the absorption characteristics of the floor and ceiling. Tapering applied to the individual real array would result in the unintended effect of periodic loud/quiet tapering of the complete array consisting of several repeated reflections.

Many loudspeaker designs "break" in unexpected ways when the speaker is placed in a listening room with six unanticipated boundary surfaces; the infinite line array, in contrast, depends on two of the boundaries (floor and ceiling) for its basic operation.

The loudspeaker-room interface is truly the last (but still largely wild and uncontrolled) link in the listening chain. An infinite line array system goes a long way toward solving the loudspeaker-room problem by anticipating two of the surfaces of the room.

For more on infinite line arrays and a design that fully anticipates four of the room boundaries see my DIY line array project here:

Corner Line Array DIY Project

Regards,

John

[Jim Griffin]

Re: Line Arrays

I've resisted jumping into this thread but line arrays come and go and if not well designed the results can be less than satisfying. The guidelines that I derived several years ago are still the keys to success. Download my white paper at:

http://audioroundtable.com/ArraySpeakers/messages/2226.html

This array is one that I designed five years ago and I'm a daily listener:

http://www.greatplainsaudiofest.com/Photos/CSSLineArray01.jpg

While I not a big fan of full range driver arrays, John Murphy's response is right on about room effects and the need to have an array that is at least close to floor to ceiling height.

Jim

[Sydney]

Re: Line Arrays

You only need to consider the reflected images to realize that tapering applied to the real array would be inappropriate. With an infinite line array setup the reflections are already naturally tapered as each generation of reflection is progressively attenuated according to the absorption characteristics of the floor and ceiling.

I've not know any who had the room and implemented a floor to ceiling speaker.
Since the absorption coefficients of different flooring choices vary ( and are not linear with frequency ) along with the effect of furnishings; which are going to attenuate reflections differently:
What do implementors do?
( a heavily carpeted floor will have a greater attenuation than a hardwood or tile floor )

but line arrays come and go and if not well designed the results can be less than satisfying.

That's for sure as does the variations on attempts to control focus.

[OlderMongrel]

Re: Line Arrays

The "magic" happens when you run a line array between parallel floor and ceiling. The floor and ceiling reflections create the perfect acoustic illusion of an infinite line array just as two parallel mirrors would.

So John, what would the theoretical ideal be? Do I need a concrete ceiling to go with the concrete floor? I currently have open joists (some insulated, some not) which I think help with most types of speakers. I suppose sheet rock would be sufficient to get decent "mirroring"?

John

[r-carpenter]

Re: Line Arrays

Because the vertical directivity is constant in the near-field of an infinite line array the vertical imaging changes from that of a point source. Unlike a point source, the image from the infinite line array is always located at ear height, that is, normal to (90 degrees to) the array. Some listeners report that this lends to the impression of a "tall" image but my own personal experience is that the image has just the right height and is overall more convincing than the vertically flattened image projected by point sources. The horizontal directivity of an infinite line array is the same as that of the individual array components. So you tend to get "normal" imaging in the horizontal plane with "non-flattened", imaging in the vertical plane. When you sit-down or stand-up the sound of the infinite line array is unchanged. You can even walk about the room and experience a very convincing image of the performers.


Regards,

John

This a wonderful observation and precisely why I got hooked up on the Line Arrays. The image is very "life like". The perception while you are walking around the room is exactly that: you are walking away or to the stage. My observation also is that with the arrays, this effect takes place the a height of the line approaches of exceeds the height of the instrument (stage, group, orchestra, performer).

I've not know any who had the room and implemented a floor to ceiling speaker.
Since the absorption coefficients of different flooring choices vary ( and are not linear with frequency ) along with the effect of furnishings; which are going to attenuate reflections differently:
What do implementors do?
( a heavily carpeted floor will have a greater attenuation than a hardwood or tile floor )

That's for sure as does the variations on attempts to control focus.

I have never implemented a true floor to ceiling array. Fist of, my ceiling in the shop is about 16ft.
But I have taken an 12 driver array (8ft tall) home where I have 12ft ceiling. I would center array in relation from the floor to the ceiling on the 2ft MDF box. My impression is that centered array sounded a bit cleaner but it's really hard to say. DBTs are hard then you doing everything yourself.
Focused array had probably the best imaging but only in the sweet spot. You could literally walk through the soundstage. Room effects appeared less important to me. My subjective opinion is that if you are locked in you listening spot, focused array is probably the best. I like to move around some, so for me it's a straight array, no tapering.

This array is one that I designed five years ago and I'm a daily listener:

http://www.greatplainsaudiofest.com/Photos/CSSLineArray01.jpg

While I not a big fan of full range driver arrays, John Murphy's response is right on about room effects and the need to have an array that is at least close to floor to ceiling height.

Jim

That looks yammy! Line arrays get you addicted very quickly.
I am doing line of 9 Jordan xj92 with 6 Fountek2.0. Bass will be carried by 4 Peerless 10" drivers TL loaded. So, it's 3 way.

[Sydney]

Re: Line Arrays

... Focused array had probably the best imaging but only in the sweet spot. .

That's the reason I wouldn't want an array with a fixed focal point.
The ability to adjust array dispersion ( to fit the particular operating circumstances ) is more useful.
Keele's CBT work and his 5 segment Bessel arrays I find interesting.
( Wish there were more - high impedance drivers suitable )

One personal assessment of arrays ( in their numerous forms ) is the significance of the driver quality: Arrays made from the cheapest drivers didn't sound as good as those from quality drivers.
For those who don't hear much difference in driver quality it may not be an issue; but I've owned some of the drivers often used in the simpler arrays.
These "close-out" drivers just didn't sound and perform as well as more expensive drivers.
Some argue that the sum result of an array of cheap drivers is a greater/better aggregate - not to my ears; versus using and matching quality components.