PDA

View Full Version : How much volume needed for 9 NSB's



ballstothewall
10-27-2005, 03:23 PM
Ok, I have a couple quick questions here regarding my future line array project. I think I have decided on 9 NSB's per tower, with one tweet in each (haven't decided on which tweet or where to cross it over at...ideas?:naughty: ) This will be powered with my new toy of a Yamaha RX-V480 receiver that I got off of ebay. It puts out 70wmrs at 8ohms....

How much volume does 9 NSB's need?

Does this idea look like it will work?

http://img394.imageshack.us/img394/3756/linearrays8ha.jpg (http://imageshack.us)

The box will probably be triangular instad of what is drawn.

What type of wood would be best?....1/2" mdf, baltic birch, or what?

Do I need to cover the inside of it with foam of some sort or stuffing like I see in pictures, or what. Please educate me

Any help is appreciated

-Ryan

thylantyr
10-27-2005, 04:40 PM
9 NSB's per tower

That works if you need 8 ohms. If you are able to drive 4 ohms I'd do an 8 line
with an impedance switch to select 4 ohm or 1 ohm only to offer more punishment if you had a beefy amplifier.

with one tweet in each (haven't decided on which tweet

A 9 line NSB array is ~ 95.5dB sensitivity so ideally you want a tweeter of
equal or greater sensitivity if you do a passive crossover design. If active,
then it's moot. You can use a horn tweeter if it has the response and SQ
you like, but they may be louder so you'd need a variable L-pad to adjust the
tweeter volume to mate with the midrange volume {fine tuning}.

or where to cross it over at...ideas?:naughty: )

Low pass -> whatever you want and it will work. But your tweeter might
dictate this. ie, if your tweeter chosen sounds good at 2khz, then do 2khz for
NSB's. If your tweeter sounds good only at 5khz, then do 5khz on the NSB,
but if you want to keep the design 'proper' use the line array spacing formula;
13560/driver center to center spacing.

If your NSB's are 4.5" c-t-c spaced, 13560/4.5" = ~3Khz .. try to keep the crossover under 3khz on the NSB's.

High pass -> Depends on slope and power you drive into the NSB. Steeper
slopes can allow lower crossover point to keep cone excursion low. For example,
an 8th order {48dB/octave slope} can be used at 65hz, a 18dB/octave slope
try to keep up over 100hz, maybe 150hz. For 6dB/12dB, even higher.
These are just guidelines. If you don't push the array hard, run the NSB's with
no high pass and just keep the volume real by not overdriving the cones,
a judgement call.

How much volume does 9 NSB's need?
The 95.5dB sensitivity means that at 1 watt/1 meter you produce ~95dB SPL.
Is this loud enough? if not drive 2 watts, 10watts, 100watts until you either
blow up the drivers or reach a SPL level that you are happy with. /hehe

Does this idea look like it will work?
You can make it work.

What type of wood would be best?....1/2" mdf, baltic birch, or what?
If you are skilled, you can use any wood. Particle board, plywood, MDF, etc.
If you want better performance then I'd make a deeper cabinet to help
disperse standing waves, use MDF to lower cabinet resonance and use
a chambered design where you have three chambers, three NSB per chamber
as the two internal shelfs will help lower cabinet resonance because they act
as braces. Within each chamber, add a 1" dowel in the center to couple the side
walls, but this dowel so it's a very tight fit when you install it and screw/glue it in
place. This reduces cabinet reasonance lower and it's cheap to do.

If sealed or ported box, poly fill loosely. If you made a hybrid ported/open baffle,
then don't polyfil, rather make sound conditioning pillows for each wall, it's pretty cool the hybrid box - but not all designs can be made to use it.

ballstothewall
10-27-2005, 05:25 PM
Ok, would this tweeter work for a passive setup?
http://www.partsexpress.com/pe/pshowdetl.cfm?&DID=7&Partnumber=270-050&scqty=2

On the box making, would it be benificial to move the tweeter up one from my current position to three down from the top in order to have only three NSB's in each chamber and not have the tweeter added into one of the chambers to get consistent volume (and just add one more brace) or does it really matter?

So every three NSB's are in their own seperate chamber, with some loose polyfill (so no air can pass between any of the chambers)? With a 1" dowl rod horizontally through the chamber to add extra support? Ok, how much Airspace is needed for the NSB's?

(and yes I consider myself pretty skilled in a woodshop, just uneducated about box building, I've only built one sub box)

And do I need to chamfer the inside side of the hole for the NSB so I don't choke the airflow (I think that is what I read, am I right?)

Ok, how do I know what the tweeter will sound good at, or is this just a trial and error sort of thing? Would this crossover work also? http://www.partsexpress.com/pe/showdetl.cfm?&DID=7&Partnumber=260-174
(one of the cheaper ones at 3khz)
Or would it be better to make my own? I can solder, I just don't know anything about making a crossover.

Thanks a lot with your help on this project, It will prob be going together next weekend I believe.

-Ryan

thylantyr
10-28-2005, 12:17 AM
Ok, would this tweeter work for a passive setup?
http://www.partsexpress.com/pe/pshowdetl.cfm?&DID=7&Partnumber=270-050&scqty=2


Any tweeter can work, I just don't know your SQ threshold. They don't have
response charts either so who knows what it does. You can test one as is
to see if you like it before construction begins.

If your budget is higher try Eminence compression drivers and choose a
horn lenses with the size you like. They have bullet tweeters too but many
have chaotic response curves. Poke around the PE site and make a list
and we can evaluate your choices.

On the box making, would it be benificial to move the tweeter up one from my current position to three down from the top in order to have only three NSB's in each chamber and not have the tweeter added into one of the chambers to get consistent volume (and just add one more brace) or does it really matter?

To me, tweeters sound best when aimed assuming you have quality on-axis
tweeters so my plan of attack would be to align the tweeter height with ear level
for best SQ. Will you be sitting down or standing up doing critical listening?
Find out and optimize for that position.... or if you want the array to look symetrical and looking cool, center it. Don't worry about consistent box volume.

So every three NSB's are in their own seperate chamber, with some loose polyfill (so no air can pass between any of the chambers)?

yes

With a 1" dowl rod horizontally through the chamber to add extra support?


The rod is inside each chamber, three chambers, three rods. Center the rod
horizontally in each chamber, think bird cage - hehe

Ok, how much Airspace is needed for the NSB's?

Depends on what you want. Some ideas.

Every 3 NSB chamber;

if 1.5 cu. ft. tuned to 75hz - about 3dB peak in upper midbass. if you want a
6dB peak, tune to 90hz.

if 1.0 cu. ft. tuned to 100hz mimicks the box I made where I used four drivers
in a 1.5 cu. ft. box. 6dB peak in response, i like it actually.

Those are ball part numbers.

(and yes I consider myself pretty skilled in a woodshop, just uneducated about box building, I've only built one sub box)

Cool, the wood skills are harder than the speaker design *if* you don't
do passive crossover networks - harr harr.

And do I need to chamfer the inside side of the hole for the NSB so I don't choke the airflow (I think that is what I read, am I right?)

3/4" thick wood baffle with a hole cut out for an NSB almost blocks airflow, it's
unsweet. The chamfering is way cool if you can do it. I think i used 4 3/8"
center to center space with a 3/8"?? chamfer, i don't remember, but this weakens the front baffle. I placed wood strips on the weak spots for strength. You can
increase center to center spacing if you don't mind making the box a wee - bit
taller ... This is your call ...

Ok, how do I know what the tweeter will sound good at, or is this just a trial and error sort of thing?

I have a closet full of sample drivers because I prefer to listen to drivers
and then pass judgement. There is really no other way to judge speakers,
you have to listen to them. I usually make a list of candidates and buy one
of each, do my listening trials and form conclusions. It svcks to spend money
but it's an R&D expense if you like this hobby.

Would this crossover work also? http://www.partsexpress.com/pe/showdetl.cfm?&DID=7&Partnumber=260-174
(one of the cheaper ones at 3khz)

Not good. incomplete.

Something like this;
http://www.partsexpress.com/pe/showdetl.cfm?&DID=7&Partnumber=260-150

but you don't need the woofer section.

Or would it be better to make my own? I can solder, I just don't know anything about making a crossover.

I think it would be better.

read this and tell me what you think.
http://www.caraudioforum.com/vbb3/showthread.php?t=227180

the passive crossover design will make or break that system so
you need to choose the tweeter, ideally find one with response graphs
to determine best crossover region to use then plan a strategy.

Linkwitz Riley slopes are just better for SQ, i'd use those, but
the LR 24dB/octave is more complicated to build and cost more
because it has more parts.

thylantyr
10-28-2005, 12:35 AM
Since you are good with wood skills and you feel like the Iron Chef,
you might want to explore someting exotic, perhaps a Variable port, sealed,
open baffle 'SQ like' box.. WTF is that?

Check something similar on my array.

Make your chamber for three NSB's 1.25 cu. ft. net. Use 3/4" wood for the
rear baffle. Cut a 5.25" hole in the back of the chamber and this is your
tuned port ~100hz tuning with a ~6dB hump near 125hz. The port length = ~3/4"
which just happens to be the thickness of the wood. /hehe

Install polyfill pillows on the walls instead of installing loose polyfill.. To make
these all you need is poly batting [Walmart] + polyfill [Walmart] and make pillows
maybe 2-3 inches thick that cover each internal chamber wall using 3M spray glue.
This will sound condition the standing waves, leave the chamber open so the port
is 'free' and any rear standing waves will see less of a back wall to reduce reflections and sorta have open baffle performance on SQ because you are directing those waves out the hole.

The hole happens to be a tuned port, a bonus, and it's tuned to 100hz
with a nice peak. This peak helps offset some diffraction loss.
http://www.trueaudio.com/st_diff1.htm

So, you have a ported box with a big azz hole in the back of each chamber,
rear waves exit the hole minimizing internal cabinet reflections, any residual
reflections absorb/diffused by the poly pillows and the natural box peak
of 6dB rise near 125hz and tailoring off near 500hz sorta offset that diffraction
loss.

Short hand -> my black sheep NSB recipe that owns joo...

variable port tuning is nothing more than making a door in the back of
the box to cover the hole and use a friction hinge. Door closed = sealed
box performance. Door open = 100hz tuned port/6dB peak. Door somewhere
in between = tuning drops lower than 100hz and the upper midbass response
attenuated. This is optional, you can EQ the peak if you want.

You can have ported box performance, sealed box too, and some open baffle SQ qualities if
that rear wave exit the hole, less coloration and the cheat to offset diffraction loss is neat,
kill many birds with one stone.

ballstothewall
10-28-2005, 02:12 AM
Thylantyr, You are The f'ing man! Ya know.

I though this was gonna be somewhat simple, throw a bunch of drivers into a box and then slap a crossover in it and throw power at it. **** was I wrong.
My mind is blown from reading all this stuff, but ya know, this is actually the most I have enjoyed myself in awhile, this stuff is Frickin cool!

Ok, so I just build the chamber for the 3 nsb's put a hole in the back that is 5.25", and since the back is 3/4", that is the port itself?

On the "pillows", I spray glue polyfill 2-3" thick on all walls?

Since you said leave the chamber open so the port is 'free'....Does this mean that I don't put the dowel rod (bird cage effect) in there, or keep in in the plans?

Then put a friction door on the back that covers the port, so I can do many things with a simple design?

I Really like the sound of this...

I will finish reading up on everthing and continue to fry my mind later tommorow (well, its today by now). I will try to draw up the plans for everthing this weekend when I have a little more time to devote to this.

Once again, THANKS Thylantyr

-Ryan

thylantyr
10-28-2005, 03:33 AM
I though this was gonna be somewhat simple, throw a bunch of drivers into a box and then slap a crossover in it and throw power at it. [b]

It can be simple and good, but also could be boring.. depends on your
expectations. If you are skilled in wood, try for a bigger design. Here is my
NSB/PT2 array;
http://home.pacbell.net/lordpk/robarray/

I made port doors -> 8" x 8" plywood, 3/4" thick with veneered edges, stained.
But i didn't install them because i can EQ the midbass peak instead of the
mechanical solution. also, i liked the sound of the ported design over sealed.

[b]Ok, so I just build the chamber for the 3 nsb's put a hole in the back that is 5.25", and since the back is 3/4", that is the port itself?

yes

On the "pillows", I spray glue polyfill 2-3" thick on all walls?
Polyfill will fall off so you make pillows using poly batting. the batting is the
skin to hold the polyfill inside. Glue the pillow on the wall. Thickness? thicker the better but you don't want to clog the airspace, judgement call.

do this, make a chamber, cut port hole and stick your head in there and yell
in a short loud burst. You may hear echo. Echo is not good. With pillows installed,
no echo-> good.

Bigger chambers are more obvious for echo sorta like an empty room in your house, major echo. You start adding 'stuff' to the room and echo is minimized.

Since you said leave the chamber open so the port is 'free'....Does this mean that I don't put the dowel rod (bird cage effect) in there, or keep in in the plans?

The reason I said dowel is because it's round, aerodynamic, placed in the center
of the chamber to brace the side walls {left and right side, not front/back},
this is good to lower cabinet resonance and you will see the rod when looking
at the speaker box through the port hole. Wrap some poly batting around the rod
too.. Looks esoteric.

Then put a friction door on the back that covers the port, so I can do many things with a simple design?

What i did was make a test box first. i made it ported, tried different tuning, etc.
You can do the same if unsure, just make a scaled down version, maybe a 3 NSB
chamber only using cheap wood. Point is, when you make your box or final product you can just plug the hole to hear what the box sounds like 'sealed',
then determine if you like it ported or sealed soundwise. if you like both, add the port door or make a port plug. if the door is on a hinge and swings open/closed,
tuning changes.. Wide open = 100hz tuning, as you close it the tuning drops in frequency and the peaky box response mellows.

It doesn't have to swing open/close, the door can slide up/down on the surface
of the rear. do a test first on what you like.

/////

An upgrade to the NSB/PT2 array would be a Dayton RS 6" / PT2 array shown
here;
http://www.caraudioforum.com/vbb3/showthread.php?t=224281&highlight=array

There are many recipes, it all comes down to skill and cash. If you want to punish
people bad, save up for the big array using the $25 tweeters, go active, use
some proamps and you can dish out some death. My NSB array knocks
stuff off the TV when I play certain upper bass notes, about 126dB peak,
not bad for 49 cent speakers. I have no subwoofer installed either. Add a sub
to this and it can be even more cooler.

The lacquer cone treatment improves SQ too... cost $5 to do and some patience.
All other mods are neat but you probably won't need them.


The 16 line NSB array can dish out some punishment in a good way but
the design can be scaled higher like this;

http://home.pacbell.net/lordpk/misc/Feandil.GIF

Matrix array using a bigger horn in the center.

if you are really the skilled woodcrafter, a Death Star array came to mind.
Make a giant ball with NSB's all around and hang the globe with string to
the ceiling, more of a novelty but i bet someone would pay premimum if
your woodcraft is done well. It would be chaotic but who cares. /heh
This would be cool to torture neighbors, place it outside and do 360 degree
torture playing slayer.

ballstothewall
10-28-2005, 12:12 PM
Thanks for all the help, now I think i know enough about the cabinet to acutally build it correctly. I would love to do something compeltely off the wall and crazy (death star) but this is a budget project right now, I'm selling some other audio stuff to finance this project, but later it will have a sub on it, but that has to wait till I get money to buy a plate amp. I really can't wait to get started on this project.

Somehow I have a feeling that I am going to get a noise complain in my dorm room sometime after I get these done:up2somet: ? Any bets on how long thats gonna take??

I will be sure to take lots of pictures and post them when I get back home to build this monster...

Thanks again

-Ryan

PV Audio
10-28-2005, 02:15 PM
it will be finished when it will be finished. take it from me. I learn something new about cabinet construction EVERY time i do it. the first ones i made were the goldwoods, and those took literally eh 3 hrs a side. this is cutting, screwing and soldering (no glue was used :crap: ). I could whip out a pair in a few days. The time when I really started to take my time was with the HIVI lacquered cabinets. those SHOULD have taken about a month more because i couldn't lacquer for my life honestly. They just don't look good, but they perform like a **** star on red bull. The next after that was using my dayton 6.5" cones with a tang band full range. These took about a month or so, but they finally turned out sounding very nice regardless of the electrical problems. The new dayton subwoofer cabinet didn't take too long actually because I had my saws using new blades and i could just cut it straight down, probably took a day or so to do all the cutting. physical construction took about a week, as well for the finishing. All in all, it didn't take that long, probably 3 weeks. To give you an idea, over on diyaudio.com, a member named shinobiwan is doing a project called Percieve 2.0. these are projected to be completed by march 2006. When you no longer care how long it takes, versus getting the sound quickly and thus jeopardizing what you are dreaming about, that is when you can build the best to your abilities. Truly takes time.

ballstothewall
10-28-2005, 02:49 PM
They just don't look good, but they perform like a **** star on red bull
:laugh: :laugh: :laugh:
****, them things must really perform well...

PV Audio
10-28-2005, 02:53 PM
hell yes!