That's it... I want an OB line array

So I've been reading up for the last several days about the NSB's and line arrays and I have to say I am extremely intrigued by the whole idea. Out of curiosity regarding open baffles and dipoles and such I made an OB speaker today out of scrap 1/2" MDF and a coaxial car speaker. It sounded like nothing I've ever heard. Music all the way around and off the walls and very loud.

I like cheap, I like extrema, and I love DIY'ing things only people with tons of cash can afford to buy off the shelf. Enter the exotic and ubiquitous line array :devilr:

I have made some "sketches" in powerpoint of some ideas and am looking for opinions. They are quasi 1:10 scale. I like the black, but piano black is probably the hardest finish to do "right." I would love to do a real cherry or walnut solid front (no MDF, my Dad has a wood shop) but I am concerned about the strength of the baffle after cutting 16-48 holes in it.

I am also deciding between a line array of uber-cheap tweeters (2 tweeters for each NSB) or just a single super horn ($20 pyle unit from PE) in the middle.

Here are some pics...

Some early ones
Some later styles
Some black ones

I can post the ppt file if anyone is interested, there may be others who find ppt good for sketches before making real plans in CAD.
 
If you read below this message it has my golden rule of DIY. Dont get too cheap on drivers here, It will effect the end result. I've heard alot of people using NSBs for line arrays, and I've heard alot of things that people have done to try and make them sound better. If you're going to use less expensive drivers, I would suggest taking it easy on the finish. You might decide you want to change the entire thing after a while...
 
Yeah, I'm looking for something cheap and easy, but still somewhat civilized looking. You can dress up a pig all you want, but at the end of the day it's still a pig :)

I have heard, though, that the NSBs sound much better in an array than one would expect for 49 cents each, and since my dad has a shop I figure I could build a set for very little $$$ even including things like speaker terminals and some nice sensitive tweeters. He has wood, MDF, and all manner of saws, routers, jigs, etc.

I don't like the look of mids and tweeters side by side, but the functionality and price of some of the closeout onkyo 25 cent car tweeters makes it appealing. I don't think I'd ever be getting more $$$ drivers for a line array like this, the cost goes through the roof very quickly. 32 drivers at $15 each is almost $500 in drivers alone. the WAF plays too much into it at that point. I want to hear music, not complaining :)

It is hard to decide how to proceed. I like the black ones pictured with a single tweeter in the middle, but maybe with 16 NSBs per side, not 12.

With a single tweeter in the middle do line arrays loose any of that "filling the room" that comes from a line array?
 
remember you shouldnt use a center-center distance greater than 1/2 wavelength of highest frequency produced, so a line of tweeters becomes difficult unless you use peizos or ribbons. If you exceed the 1/2 wavelength lobing issues will result.

I've never really heard of using peizos in line arrays before, but due to the fact that they dont have a magnet structure the distance from center to center could dramatically be reduced. They're generally cheap too. You might want to look into that
 

thylantyr

Member
2001-02-19 10:38 pm
Mars
I have info here on my NSB project;
http://www.caraudioforum.com/vbb3/forumdisplay.php?f=37
/// Come to the dark side :devilr: ////

The NSB sounds better with the cone treatments, I prefer them in
a ported box as I've done some listening tests recently between
ported and sealed. Many people like open baffle but they need a
large room and many feet from the wall to get the right sound, but
you have no bass. In a sealed box you have weak bass, but in a ported box tuned correctly it's pretty good for 49 cent speakers
and I'm amazed still at what can be done when you do the porting
correctly and you place the cabinets in the corner of a room.
People that audition the array think I have some hidden woofers
in the cabinets - nope ...I have no woofers and I have no
subwoofers but the upper bass is strong enough to fool them.

I'm also surprised that my NSB array is handling my 70hz high pass
crossover point well and I drive them with a QSC RMX 2450 amp
near clipping, rated for 1200 watts @ 2 ohms per channel, the NSB's
are wired for 2 ohm per channel.

I'm convinced that you can get really good sound out of an NSB
array if the design is done properly meaning the system as a whole
has to be designed well. It seems that people that cut corners will compromise the sound which gives the NSB a bad name when in fact it's not the NSB's fault it sounds ok or bad, it's the fault of the designer.

If you are on a budget get a good Eminence compression driver
and horn lense for the center of the array and use an Lpad.
If you need more info, follow the link above :devilr: :smash:
 
nsb mods

Hi all,

I made a 3' wide nsb 16x array. The low end was defeted to much. Also I ended up placeing a coil to bring down the highs. They sound good when there is just a fraction of a watt driveing each driver. The main mod was removing the dust cap and replaceing it with a 3" long dowell. A noticible clarity improvement.

I recomend a true ribbon tweeter for speed and clarity. I am making a pair to save $$ and for the fun of it. The first iteration is 7" x .32 ". I made the transformers also.

I am going to place the nsb 16x array in a vented box to get more bass and see if the results are livable.

Some of the open baffle appeal may be due to the rear fireing wave. It seems to fill the room and make the sound source seem less apparent. I like the idea of no stored energy and no back wave comeing through the paper cone.

-Linc
 

thylantyr

Member
2001-02-19 10:38 pm
Mars
no back wave comeing through the paper cone.

Regarding the back wave. Referencing this picture;
http://home.pacbell.net/lordpk/robarray/Picture20.JPG

There is 4 chambers, 4 NSB per chamber and a 6" diameter
port in the rear of each chamber with sound conditioning pillows
on the walls. The sound conditioning is only on the walls leaving
the center of the chamber open. While it's not a perfect
implementation, I'm willing to bet that some of the back wave
exits the port. So, here's a ported box with some benefits of
getting rid of the back wave - :dead:

You can get less coloration with this method and you also get
the benefits of a ported design, more bass. If you want a ported
box tune it higher because the NSB won't move alot of air and
tuning it too low might be moot. The higher tuned box gives
you a peak in response if you chose the volume of the box
correctly and you may offset some of the bafflestep. I haven't
measured the array yet so this is speculation based on what I
hear.
 
thylantyr said:

If you are on a budget get a good Eminence compression driver
and horn lense for the center of the array and use an Lpad.
If you need more info, follow the link above :devilr: :smash:

Well, I have read your whole writeup and looked at your pictures many times :) I was really impressed by the sound of the open baffle though. I think I could make something after these to handle the low midbass and super low bass using a TL and subwoofer combination. If I go with ported enclosures for the NSB's I am worried about the weight of such big boxes, as I have small kids and have to carry the cabinets upstairs to listen to them.

With an L-pad and a tweeter like this I could cross the NSB's over at around 5K and try to avoid the 7k bump. I am still uneasy about using the cheapy tweeters...

If I screw something up I can chalk it up to a learning experience, at least it will be fun :)
 
PE has a buyout on some $3.95 3.5 inch Bravox drivers with a very high Qts of more than 0.8 which if i'm not wrong would make it ideal for OB use. I did a search here but there isnt much discussion about these driver. IMO they would be an interesing semi-cheap experiment in Line Arrays and would probably sound a lot better than the NSBs. 30 of those drivers would cost 120...not an unreasonable price. I wish i lived in the US, i would have bought them...the shipping for 30 drivers to australia costs 225 USD:eek:
 
If I go with ported enclosures for the NSB's I am worried about the weight of such big boxes, as I have small kids and have to carry the cabinets upstairs to listen to them.

I'm guessing my tower is 200 - 250 lbs each.

With an L-pad and a tweeter like this I could cross the NSB's over at around 5K and try to avoid the 7k bump. I am still uneasy about using the cheapy tweeters...

I don't know what that tweeter will sound like.

The problem with picking drivers {tweets, mids, etc.,} is not
knowing what they sound like before purchase. I bet if you had
access to all these tweeter candidates and auditioned them you
would have a different opinion on which one to choose. Graphs
only tell a small tale. The way I find peace of mind in my projects
is to buy samples of drivers I'm interested in. I only buy one of each
and test only 1 channel to hear what it sounds like.

I don't know what your sound quality expectations are
and how much work you want to invest in the project. If you
want to crossover the midranges that high you should treat
the cones, but before you do this you should test one NSB
with treatment vs. one that is not and compare the sound.

If you array the NSB's the crossover should be closer to 3khz
but people cheat all the time in line array design. Look at the
McIntosh array -- lol

I've run them full range at low - medium volume and it works,
the SQ is subjective, the gremlins heard is subjective. But if
I crank up the sound it's too busy and smeared.

You should build a scaled down version test box to understand what
the drivers can offer. Make a 4 NSB test box ported, plug the hole to
hear it sealed, then do the OB test to make sure you are 100%
satisfied that you made the correct decision.
 
nsb mod testing

To test the single drivers listen to them at perhaps 1-2' or closer (not loud of course ). Jim Griffin goes over the distance calculation in his paper. D= ~1.5 x the array height at 1khz. When starting with my nsb project I went with a single , then 4, then 16. Each time there was a different type of sound that the drivers made. They did not just get more efficient. I think the main factor was that as more drivers are added the near field region becomes larger.

Does any one have an idea why the nearfield has a different sound than farfield. I think it is more than room bounce.

-Linc
 
Re: nsb mod testing

lcholke said:
Does any one have an idea why the nearfield has a different sound than farfield. I think it is more than room bounce.

-Linc
Comb effect? The ratio of listening distance to the distance between each driver. As you move away from the array, the individual piont sources seem closer to each other and merge together to form the line array.

Max
 
maxro ,

I was once listening to a set of fostex fx200 drivers (ob mounted) and was wondering why they did not have the punch of my 16 array. Well punch may not be the right word, but they had horn dynamics without the horn tones. Then I moved about 1 or 2 ft from the fostex driver and they were there. A while back when I had a chance listened to the needles line array it had that line array punch also. Jim Griffin gave a talk at the Dayton DIY meet and noted the two different sound types. He did not go into the cause and I did not think of asking him.

-Linc