how do you overcome low frequency radiation transition of line arrays?

thadman · 22nd December 2006, 05:30 PM

Ive been doing some research and have found a serious problem with line arrays...you're only going to be able to listen to them at a predetermined distance where the radiation patterns of the drivers are equal.

An 85.5" line of 12 RS180s (flange to flange) has a 424hz transition at 3 meters and 706hz transition at 5 meters (near field --> far field radiation pattern)

The 1/2 wavelength CTC spacing requirement for nearfield listening limits you to extremely small drivers with an extremely low crossover point (940hz crossover for RS180, 7.125" flange and 1357hz xover for the RS125, 4.9375" flange)

How do I overcome this? What tweeters come to mind that could meet this low xover point and execute it with high fidelity, while also having extremely close CTC spacing as to avoid comb filtering altogether except maybe in the final octave?

Dr.EM · 22nd December 2006, 07:24 PM

I believe large ribbons/arrays of ribbons are popular

thadman · 22nd December 2006, 08:31 PM

That still doesnt fix the issue at hand...

The line arrays polar response is going to transition from near field (sound is dominated by cylindrical wavefront and decreases at -3dB for every doubling of distance) to far field (sound is dominated by reflections and decreases at -6dB for every doubling of distance). This forces me to listen at a specified distance...and that distance only or the amplitudes get all crazy. I cannot correct this with equalization...

How do I combat this problem?

Tenson · 22nd December 2006, 09:03 PM

You build them with ribbons that don’t have much distance between them and then you build them BIG so whatever position you choose, you are in the near field. Even if you are starting to enter the far-field though, you are not going to get a completely messed up response unless you are far enough out to get more than a few dB of change.

thadman · 22nd December 2006, 09:22 PM

...I dont see how that solves the problem. The line simply is not long enough to allow nearfield radiation in the lower octaves. I dont see how using a ribbon or any other interrupted diaphragm could solve this problem...

At 100hz the line would have to be 4.47 meters or 176 inches! My ceiling allows maybe around ~90...tops.

Jim Griffin · 22nd December 2006, 10:07 PM

Thadman,

Relax and take a think about your issues. We can work through it. First, I assume that you are aware of my Near Field Line Array white paper:

http://www.audiodiycentral.com/resource/pdf/nflawp.pdf

1. Line height vs. near/far field transition distance for the woofer line.

If you only consider the line height (85.5"0 high in your case), then you would have low frequency transitions as you calculate. But as discussed in the white paper you have the benefits of in-room close coupling responses from the floor and ceiling at these lower frequencies. Only a padded chamber would not have reflections as we go down in frequency. My suggestion is that you assume that the line height is actually extended by a factor of 3 (assumes that you have reflections from the floor and ceiling which effectively triple the line height). Thus your effective line height is 3 x 85.5". Now if you compute the low frequency near/far transition frequency is on the order of 52 Hz. That means that you are well into the omnidirectional region so the room size dictates how the sound propagates with it. That means that effectively you have the near field line effect well below 100 Hz. Please read more details in my white paper.

2. The center to center spacing between drivers and its effect on near field listening. Primarily this is related to the spacing between the woofers and the tweeters in a two-way line array.

First of all, this isn't a near/far field listening situation per se but rather comb lining and line sensitivity vs. frequency issues. Your half wavelength c-t-c spacing criterion is good to prevent comb lining and line sensitivity roll off concerns. But as you realize this criterion is very difficult to meet in practice. In the white paper I suggest a wavelength c-t-c spacing or less and I have built several arrays with this spacing. My best answer is to actually measure the performance of the woofer and tweeter lines tha tyou expect to use. In my experience you can better live with some slight comb lining in the vertical plane a bit more than compensating for the roll off in sensitivity--at least within a passive crossover. I would suggest that you strive to have the crossover in the 1/2 to one wavelength spacing for success with your woofer line.

Now for your tweeter line you'll find a good discussion in the white paper on why I like ribbon or planar tweeters for the tweeter application in a line array. This is because the ribbon and planar tweeters have limited vertical plane sound dispersion so their sound field trends toward not overlapping as frequency goes up. Thus you don't have comb lining with these tweeters as compared to cone/dome tweeters. Again see the white paper for more details.

Jim

thadman · 22nd December 2006, 10:42 PM

Ok...thats reassuring. I never considered the ceiling and the floor.

Is CTC spacing extraordinarily critical such as crossing over below the 3rd harmonic of a giant 20dB breakup node, or is it just something thats recommended?

How would a stack of BG Neo8s fair?

thadman · 22nd December 2006, 10:52 PM

Would these remove bass/midbass nulls in the same way that dipole alignments offer because of their highly directive nature or no?

Im asking this, because I have the option of running these down to 50hz and porting them...or running them OB and crossing them up high (150-250hz) to a Dayton HF 15 on a 24" baffle.

auplater · 22nd December 2006, 11:15 PM

Quote:

Originally posted by thadman
Ok...thats reassuring. I never considered the ceiling and the floor.

Is CTC spacing extraordinarily critical such as crossing over below the 3rd harmonic of a giant 20dB breakup node, or is it just something thats recommended?

How would a stack of BG Neo8s fair?

rather than use an array of drivers, I've had pretty remarkable success (to my and many others ears at least) with single BG planars in a line source dipole setup...

Magnetar · 22nd December 2006, 11:56 PM

Hi again, the little Parts Express Neo domes with a 30 per side array is really pretty nice sounding and dynamic as a horn loaded compression driver. The problem is the 2.5 to 3k crossover point, not driver spacing interference or lobing .

I think 300-1.2K is a more suitable crossover point for 5.25" drivers (these DO interfere with each other !) that's why I was thinking of building those high efficiency stat tweeter lines to use with my 5.25" array mids.

22nd December 2006, 05:30 PM	#1
thadman diyAudio Member Join Date: Dec 2005 Location: west lafayette	how do you overcome low frequency radiation transition of line arrays? Ive been doing some research and have found a serious problem with line arrays...you're only going to be able to listen to them at a predetermined distance where the radiation patterns of the drivers are equal. An 85.5" line of 12 RS180s (flange to flange) has a 424hz transition at 3 meters and 706hz transition at 5 meters (near field --> far field radiation pattern) The 1/2 wavelength CTC spacing requirement for nearfield listening limits you to extremely small drivers with an extremely low crossover point (940hz crossover for RS180, 7.125" flange and 1357hz xover for the RS125, 4.9375" flange) How do I overcome this? What tweeters come to mind that could meet this low xover point and execute it with high fidelity, while also having extremely close CTC spacing as to avoid comb filtering altogether except maybe in the final octave?

Thread Tools	Search this Thread
Show Printable Version Email this Page	Search this Thread: Advanced Search

Similar Threads
Thread	Thread Starter	Forum	Replies	Last Post
NSB line arrays for PA use?	Spasticteapot	PA Systems	1	1st April 2007 12:32 PM
Do line arrays...	LizardBrain	Full Range	4	23rd March 2006 05:21 AM
WTB: Pair of line arrays	mazeroth	Swap Meet	1	15th March 2005 02:03 PM
Line Arrays Down Under	deaf monkey	Multi-Way	0	14th December 2004 06:03 PM
Line Arrays	Won	Multi-Way	33	10th February 2002 10:20 PM

New To Site?	Need Help?
Register to Participate Search Privacy Statement Contact Us	Frequently Asked Questions Did you forget your password? Mark Forums Read


	Please consider donating to help us continue to serve you. Ads on/off / Custom Title / More PMs / More album space / Advanced printing & mass image saving

22nd December 2006, 07:24 PM	#2
Dr.EM diyAudio Member Join Date: Oct 2006 Location: Swindon	I believe large ribbons/arrays of ribbons are popular

22nd December 2006, 08:31 PM	#3
thadman diyAudio Member Join Date: Dec 2005 Location: west lafayette	That still doesnt fix the issue at hand... The line arrays polar response is going to transition from near field (sound is dominated by cylindrical wavefront and decreases at -3dB for every doubling of distance) to far field (sound is dominated by reflections and decreases at -6dB for every doubling of distance). This forces me to listen at a specified distance...and that distance only or the amplitudes get all crazy. I cannot correct this with equalization... How do I combat this problem?

22nd December 2006, 09:03 PM	#4
Tenson diyAudio Member Join Date: Feb 2006 Location: Kent	You build them with ribbons that don’t have much distance between them and then you build them BIG so whatever position you choose, you are in the near field. Even if you are starting to enter the far-field though, you are not going to get a completely messed up response unless you are far enough out to get more than a few dB of change.

22nd December 2006, 09:22 PM	#5
thadman diyAudio Member Join Date: Dec 2005 Location: west lafayette	...I dont see how that solves the problem. The line simply is not long enough to allow nearfield radiation in the lower octaves. I dont see how using a ribbon or any other interrupted diaphragm could solve this problem... At 100hz the line would have to be 4.47 meters or 176 inches! My ceiling allows maybe around ~90...tops.

22nd December 2006, 10:07 PM	#6
Jim Griffin diyAudio Member Join Date: Jun 2003 Location: Tennessee	Thadman, Relax and take a think about your issues. We can work through it. First, I assume that you are aware of my Near Field Line Array white paper: http://www.audiodiycentral.com/resource/pdf/nflawp.pdf 1. Line height vs. near/far field transition distance for the woofer line. If you only consider the line height (85.5"0 high in your case), then you would have low frequency transitions as you calculate. But as discussed in the white paper you have the benefits of in-room close coupling responses from the floor and ceiling at these lower frequencies. Only a padded chamber would not have reflections as we go down in frequency. My suggestion is that you assume that the line height is actually extended by a factor of 3 (assumes that you have reflections from the floor and ceiling which effectively triple the line height). Thus your effective line height is 3 x 85.5". Now if you compute the low frequency near/far transition frequency is on the order of 52 Hz. That means that you are well into the omnidirectional region so the room size dictates how the sound propagates with it. That means that effectively you have the near field line effect well below 100 Hz. Please read more details in my white paper. 2. The center to center spacing between drivers and its effect on near field listening. Primarily this is related to the spacing between the woofers and the tweeters in a two-way line array. First of all, this isn't a near/far field listening situation per se but rather comb lining and line sensitivity vs. frequency issues. Your half wavelength c-t-c spacing criterion is good to prevent comb lining and line sensitivity roll off concerns. But as you realize this criterion is very difficult to meet in practice. In the white paper I suggest a wavelength c-t-c spacing or less and I have built several arrays with this spacing. My best answer is to actually measure the performance of the woofer and tweeter lines tha tyou expect to use. In my experience you can better live with some slight comb lining in the vertical plane a bit more than compensating for the roll off in sensitivity--at least within a passive crossover. I would suggest that you strive to have the crossover in the 1/2 to one wavelength spacing for success with your woofer line. Now for your tweeter line you'll find a good discussion in the white paper on why I like ribbon or planar tweeters for the tweeter application in a line array. This is because the ribbon and planar tweeters have limited vertical plane sound dispersion so their sound field trends toward not overlapping as frequency goes up. Thus you don't have comb lining with these tweeters as compared to cone/dome tweeters. Again see the white paper for more details. Jim

22nd December 2006, 10:42 PM	#7
thadman diyAudio Member Join Date: Dec 2005 Location: west lafayette	Ok...thats reassuring. I never considered the ceiling and the floor. Is CTC spacing extraordinarily critical such as crossing over below the 3rd harmonic of a giant 20dB breakup node, or is it just something thats recommended? How would a stack of BG Neo8s fair?

22nd December 2006, 10:52 PM	#8
thadman diyAudio Member Join Date: Dec 2005 Location: west lafayette	Would these remove bass/midbass nulls in the same way that dipole alignments offer because of their highly directive nature or no? Im asking this, because I have the option of running these down to 50hz and porting them...or running them OB and crossing them up high (150-250hz) to a Dayton HF 15 on a 24" baffle.

22nd December 2006, 11:56 PM	#10
Magnetar Banned Join Date: Feb 2004 Location: Portal 2012	Hi again, the little Parts Express Neo domes with a 30 per side array is really pretty nice sounding and dynamic as a horn loaded compression driver. The problem is the 2.5 to 3k crossover point, not driver spacing interference or lobing . I think 300-1.2K is a more suitable crossover point for 5.25" drivers (these DO interfere with each other !) that's why I was thinking of building those high efficiency stat tweeter lines to use with my 5.25" array mids.

Currently Active Users Viewing This Thread: 1 (0 members and 1 guests)