Line Array Speaker vs Point Source Speaker

Johnk: No, virtual line source. Sound Lab full range electrostatic. You can read about it here.

I have heard excellent line arrays, but one respected engineer has told me that arrays have different arrival times, since the drivers are different distances from the listener, and thus could not recommend them. I have seen a curved line array online that solved the distance issue, but I can't seem to find a link.

Here is a an interesting discussion on line arrays.

http://www.diyaudio.com/forums/multi-way/17948-line-source-speaker.html

"Is there any inherent advantage to either of these speaker designs [Line array or point source]?" - Mitch4t

The main difference is that output from a line source or line array will fall off at 3 dB for every doubling of distance, instead of 6 dB for every doubling of distance for a point source, under anechoic conditions.

Some time ago I measured a line-source and a point-source speaker at 1 meter and again back at 8 meters, in a large living room. The reverberant field contribution was thus included.

Anechoic theory would predict a falloff of 18 dB for the point source and 9 dB for the line source (going from 1 meter back to 8 meters = three doublings of distance). The actual measured falloff (using broadband pink noise) was 11 dB for the point source and only 4 dB for the line source!

Now why this matters is, that uniform sounfield is significantly closer to what you normally experience at a live performance, and so it better conveys the "feel" of listening to live music. Also, the sweet spot is a lot wider than normal because the SPL doesn't change much as you move from side to side.

"I have heard excellent line arrays, but one respected engineer has told me that arrays have different arrival times, since the drivers are different distances from the listener, and thus could not recommend them." - Emailists

A true line source beams severely in the vertical dimension, but the beam is a coherent wavefront the height of the speaker, so different arrival times is not an issue.

The more a line array (vertical stack of drivers) departs from behaving as a true line source, the more different arrival times becomes a reality. However, different arrival times in the vertical plane is far less detrimental than different arrival times in the horizontal plane. The ears will cue off of the first arrival sound for imaging purposes, and so you may well find that the image height is at ear level whether you are sitting or standing. If there is any degradation of clarity, it is minor or negligible with a good line array.

One of the reasons I don't build line source or line array speakers is, I don't see any window of opportunity to "build a better mousetrap". I happen to be a dealer for the same line-source speaker that Essentialaudio is, and when I informed their designer (Roger West) that I was embarking on my own commercial speaker venture, I told him that my goal was to build the second-best speakers in the world. As far as the basic concept goes, I do not know how to improve on what he's doing, nor do I have the skills to even try to equal it.

Duke

That's quite a compliment Duke!

How do line arrays deal with "comb filtering", or is it not an issue?

Hce4, comb filtering in the vertical plane is dealt with by having the inter-driver spacing small in relation to the wavelengths produced, and/or using drivers that inherently have limited vertical dispersion anyway and therefore little or no overlap vertical overlap (like the full-height ribbons in the modules that make up the Dali Megailne).

In the horizontal plane, we want the inter-driver spacing as small a possible in relation to a wavelength at the crossover frequency. A steep crossover also helps to prevent lobing in the horizonatal plane by narrowing the frequency region covered by two drivers.

Some prosound line arrays have slot-loaded compression drivers flanked by a pair of midwoofers. The slot loading of the compression driver gives it a narrow horizontal footprint but wide horizontal coverage, along with narrow vertical dispersion (which minimizes vertical comb filtering) in a format that allows a fairly low crossover frequency so that the flanking midwoofers don't beam too badly in the horizontal plane in the crossover region. With the right crossover, it should be possible to avoid any deep nulls in the horizontal plane within the coverage angle of the array.

Duke