AVS › AVS Forum › Audio › DIY Speakers and Subs › Twelve 10" NHT Subwoofer Build.
New Posts  All Forums:Forum Nav:

Twelve 10" NHT Subwoofer Build. - Page 3

post #61 of 274
9/8/08 at 7:42pm
Thread Starter 
Okay, I spent awhile doing the testing. I did use my SPL meter on a tripod, but no computer program to run sweeps.

I put the Denon sub level at '4', not +12 max and turned all other speakers down to -12. I didn't turn on the sub, and measured all the other speakers at volume -20 on my Denon. They didn't register high enough to mess with the subs levels.


Turned on the sub amp (100 watts each) to a bit under half way and played the same 1 minute portion of a decent bassy song and set the SPL meter to find min/max for that 1 minute. Then I moved the sub to another position and started the song over.
I checked the max SPL for each location and noted which sounded best. Obviously these numbers aren't max SPL. Each done twice. Lots of positions yeilding about the same, but I'll just post the main ones.



Directly next to each side of the center channel firing out: 107db
Almost directly under sides of the screen firing out: 109db
Backed into the corners firing towards the middle: 109db
Corner straight out 4" away from walls: 109db.
Corner shooting up, down, other: 109db


BUT...

I tested these next two 3 times each because I was kinda surprised. Got the same number with each test.

In the corners 4" away from walls, but firing directly towards each other across the room: 110db


And then oddly enough, same 4" position, but firing directly at the front wall that the screen is on, looking at the back of the box: 111db.




Definitely sounded more bassy in those last 2 positions. But it wasn't as clean sounding as having the subs pretty much directly under the sides of the screen facing me. I think I would be happy with any of it along that front wall.




Does any of that mean much of anything? Louder in the corners, but possibly a bit more punch under the screen.



Just went and tried another possible position. 4" away from corners, facing side walls: 110db.
Reply
Reply

AVS Top Picks

post #62 of 274
9/8/08 at 10:19pm
Wait till you start with the REW, it's usually an eye opener.
Reply
Reply
post #63 of 274
9/8/08 at 11:58pm
It is hard to measure anything worthwhile with a song and an SPL meter. The meter reads a composite of what is playing, and unless you had say pink noise, bandwidth limited, it is hard to say what is actually meaningful.

The corners will give you the most gain, typically +9db from anechoic measurement below 40 hz.

Interesting to note how dramatic room modes are, and how little bass there are in spots in the room due to destructive interference.

Cool to see that you have them up and running BTW..
Reply
Reply
post #64 of 274
9/9/08 at 6:36am
Thread Starter 
Quote:
Originally Posted by michael hurd View Post


Cool to see that you have them up and running BTW..



Oh sorry, I don't have them up and running yet. I was just using my 2 ED 12" subs for testing.

I haven't even started the build because I'm not sure which type of box to build yet.

I will take my laptop home and see what all this testing stuff is about.


I was hoping that my testing would help decide something one way or the other.
Reply
Reply
post #65 of 274
9/9/08 at 7:31am
Ah, my bad... I assumed that you had the boxes built and were trying them out. How about try stacking them both in one corner, one on top of each other, you should get a few more db this way.
Reply
Reply
post #66 of 274
9/9/08 at 12:36pm
Quote:
Originally Posted by NEO Dan View Post

Wait till you start with the REW, it's usually an eye opener.

That's an understatement. Usual reaction is "how can my response look that bad when it sounds pretty good to me?"
Reply
Reply
post #67 of 274
9/9/08 at 1:00pm
Thread Starter 
Quote:
Originally Posted by NEO Dan View Post

I don't have great feelings for what your about to do by putting your subs out into the room.

Have you measured the difference between the response from the box pointed out into the room at the proposed 21" verses facing sideways into the corner at ~4"?

Sorry guys, I guess there's so many ideas flowing through this thread, I can't get a handle on just one. Ported, sealed, ported and sealed. I'm very anxious to get this started.

Dan, I'm not sure I follow you here. I'm hoping to get some software for more accurate results tonight.

What do you mean by putting my subs out into the room at 21"?

And I'm not sure I follow what you mean by facing sideways into the corner at 4". I guess I'm just getting too much info in this thread. I keep changing box designs and demensions but nothing is really giving me the correct volume needed for ported.

I realize I must be missing something, but your original thought was to put 4 in a tall box in the corners. Where does that 4" idea come in at? And won't that box put the drivers out 20" into the room?
Reply
Reply
post #68 of 274
9/9/08 at 1:11pm
Thread Starter 
Quote:
Originally Posted by Ricci View Post

That's an understatement. Usual reaction is "how can my response look that bad when it sounds pretty good to me?"



Yes, I definitely understand that. But how can I test the area for the proper future box design if the box I curently have isn't going to be anything like the build? Won't the test be flawed if the subs I have are tuned in the low 30's? And the future ones will be sealed or tuned in the upper teens?
Reply
Reply
post #69 of 274
9/9/08 at 4:26pm
There is no advantage to taking the woofers and putting some of them in vented boxes and others in sealed boxes, and run them in the same system. Since the phase responses of the two system types will be different at low frequencies, you will get destructive interference and reduced output. Not much, but some. Combine this with the fact that each system will need a different high pass filter tuning to get optimal flat response without overloading the ported system below Fp and there is no reason to do this.

Since enclosure size doesn't seem to be a consideration, you may want to try this.

Build the systems with a vented design. Use an optimally designed active HP filter with it. You can then plug the vents and run the system with an optimal active HP filter for a sealed system.

The bottom line is the sealed system is going to have better transient response, with lower distortion down to its cutoff frequency. The vented system will have more bass extension and lower distortion around Fp.
Reply
Reply
post #70 of 274
9/9/08 at 5:16pm
Thread Starter 
Thanks Jack. I listened to you subs today and I was really impressed.

I had 2 new 10" ported boxes sitting around that are just like the 2 holding my 12" ED's.

The 10" boxes are only 1.5 cu ft and tuning is high, I think around 35hz.


I went ahead and put 2 of your 10's in and disconnected the ED's. I must say, they sound very nice even in those boxes. This is the first time I have actually listened to them. Honestly, I believe 12 of these will be serious overkill! The sound I was getting from these 2 for music was great. I think 4 of them ported for my music would be perfect. Obviously those boxes aren't going to give me the lows I need for HT.

I stuffed some very dense foam into the ports and listened for a while. I liked the ported better. Though to be fair, 12 of them sealed might be just fine.





So, I'm down to the final days before I just give up on all my ideas and build some boxes. After todays listen, I honestly think 8 of these will blow me out of the room, let alone 12. I will be getting an EQ and the HP filter.




My final plea for help: If you have the time, could you let me know the smallest cu ft that you would recommend putting these in for a ported box? I used Unibox and seem to have come up with 3 drivers in a 6.75 cuft heavy filled box tuned around 16hz. But that seems kinda small compared to what you and Neo Dan mentioned. Would 7 or 8 cu ft be the absolute smallest to get a decent response with 3 in a ported box?

Smallest ported box for 3?
Optimum ported box for 3?
Reply
Reply
post #71 of 274
9/9/08 at 6:40pm
chopping the box down for three drivers make the port longer unless you step down to a 4", I wouldn't do that.

OTOH you do definitely have the room for any configuration, I checked:
Reply
Reply
post #72 of 274
9/9/08 at 6:48pm
those are over 6'³ sealed, so 10'³ towers are an easy fit.
Reply
Reply
post #73 of 274
9/9/08 at 9:39pm
Thread Starter 
Dan, the boxes of 3 are exactly what I was hoping for. But I'm assuming I can't port these with only 2 cf per driver? It would sure be a nice fit if I could. Because as soon as I go any larger than 2 cuft per, they no longer fit without the boxes being different depths.

The only way that I have found to get about 2.5 cuft for each driver would be like this:




I don't like the look of this as much, plus I will be finding a new place for the amp.

But this makes the bottom boxes 7.74 ft
side boxes are 7.5 each

So I could get them pretty close.


Not sure where the ports would go. Maybe I could do two 4" ports on each box, one between each sub.

I've about run out of ideas to keep these in ported sets of 3, and that's my biggest goal. If I can't do that, then I'll just run only 4-5 per corner solely for movies. And run a whole seperate amp for music and 2 other subs.. But that's last resort.
Reply
Reply
post #74 of 274
9/10/08 at 1:38am
Erich,

I think you must not be using Unibox correctly. The smallest vented enclosure I would put the 083-x woofer in is 4.4 cu ft, per driver. The port tuning should be about 17-17.5Hz. If your modeling with Unibox says to put the driver in a 2.25 cu ft enclosure, something is really wrong. Can you post or e-mail your driver file/enclosure design from Unibox?

In a sealed enclosure, the volume should be 2.25 cu ft, per driver.
Reply
Reply
post #75 of 274
9/10/08 at 6:45am
Quote:
Originally Posted by Erich H View Post

...I don't like the look of this as much, plus I will be finding a new place for the amp....

Would you consider building a shallow cabinet the entire length between the two upright enclosures? It would be an 18" deep "shelf" at the bottom of your screen, and could be subdivided with verticals to house your center channel and, in another division, your amp. It would give you the option to later add another amp, an equalizer, or other equipment. It would tie the whole arrangement together, and integrate the entire construction as one "built-in".
Reply
Reply
post #76 of 274
9/10/08 at 7:46am
Thread Starter 
Quote:
Originally Posted by Jack Hidley View Post

Erich,

I think you must not be using Unibox correctly. The smallest vented enclosure I would put the 083-x woofer in is 4.4 cu ft, per driver. The port tuning should be about 17-17.5Hz. If your modeling with Unibox says to put the driver in a 2.25 cu ft enclosure, something is really wrong. Can you post or e-mail your driver file/enclosure design from Unibox?

In a sealed enclosure, the volume should be 2.25 cu ft, per driver.

Wow, I didn't think they needed such a large box for ported. I don't think I'll be able to use them all up front by any stretch.

I guess I was trying to keep the cone excursion from going over it's limits when I used Unibox. Obviously I did something wrong.








I guess I thought these could be used in smaller enclosures as I saw some people saying they were doing 2.5 cuft ported. I think I need to step back from this project for a few days and reevaluate those box sizes in this room. I kinda feel a bit frustrated that I'm wasting people's time right now by just throwing out ideas every day without any building going on.
Reply
Reply
post #77 of 274
9/10/08 at 9:00am
Erich,

A couple of comments on the simulation. You have selected Heavy fill for enclosure damping. This isn't possible/desirable with a vented enclosure. You need to select Walls covered. The calculated f3 under Design by Vb, Fb and Q is wrong. It should be around 16Hz. The only way this value makes sense is if Unibox is doing a large signal analysis under this section. Can you post a screen capture of the top half of the spreadsheet with all of the woofer parameters?

Note that the volume for the entire enclosure is 561.4l (liters). This is listed under Standard design Vb. The enclosure volume per driver is Physical Vb = 192.0l (liters), listed under Design by Vb, Fb and Q. I think your main problem is this. You've converted from l (liters) to cu ft incorrectly. To convert, divide the volume in l by 28.3 to get cu ft. That makes the enclosure volume 6.78cu ft per driver.

I wouldn't recommend this alignment (combination of Vb and Fb) with the woofer at all. The box is too large and it is tuned too low. I would use the 125l (4.4cu ft) tuned to about 17.5Hz. It will sound better and play louder.
Reply
Reply
post #78 of 274
9/10/08 at 9:24am
Quote:
Originally Posted by Jack Hidley View Post

Erich,

A couple of comments on the simulation. You have selected Heavy fill for enclosure damping. This isn't possible/desirable with a vented enclosure. You need to select Walls covered. The calculated f3 under Design by Vb, Fb and Q is wrong. It should be around 16Hz. The only way this value makes sense is if Unibox is doing a large signal analysis under this section. Can you post a screen capture of the top half of the spreadsheet with all of the woofer parameters?

Just an FYI, the reason the f3 isn't matching and the size looks way too large is that the model he's attached above is for 3 drive units wired in parallel in the 192L box.
Reply
Reply
post #79 of 274
9/10/08 at 10:22am
Thread Starter 
Okay, here is the top page from Unibox:




For what it's worth, I didn't convert liters to cu/ft incorrectly, I just thought that because I selected '3 drivers in parallel' that the 192 liters was for the entire box of 3, not per sub at that point.
Reply
Reply
post #80 of 274
9/10/08 at 11:10am
The 192l is definitely per woofer.

If you change the number of drivers under Drive Unit Configuration, the Vb under Standard Design will always be equal to the Vb under Design by Vb, Fb and Q multiplied by the number of drivers selected under Drive Unit Configuration.

Sealed is your friend.
Reply
Reply
post #81 of 274
9/10/08 at 12:15pm
Quote:
Originally Posted by Jack Hidley View Post

The 192l is definitely per woofer.

If you change the number of drivers under Drive Unit Configuration, the Vb under Standard Design will always be equal to the Vb under Design by Vb, Fb and Q multiplied by the number of drivers selected under Drive Unit Configuration.

Sealed is your friend.

I think you're mistaken on that. The "Standard Design" and "Design by Vb, Fb and Q" are independent - the graphs displayed are based soley on the "Design by Vb, Fb and Q" variables. I believe the "Standard Design" is just what Unibox is recommending for the number of drivers and power you have selected. For example, if I leave it on 3 wired in parallel and change Vb to 700L the "Standard Design" Vb stays at 561.4l rather than being calculated to 2100l
LL
Reply
Reply
post #82 of 274
9/10/08 at 1:39pm
Lennon,

I see what you are saying now. I think I must have looked at the closed box numbers at one point, without realizing it, when I was trying to figure out what assumptions Unibox was making. It's frustrating having a spreadsheet and not being able to look at the source code to see what is really going on.

64l per woofer tuned to 16Hz really isn't going to work well. This is the type of alignment that gives bass reflex systems a bad name. Compare the systems with a port tuning frequency of 16Hz in a 64l enclosure and a 200l enclosure. Compare the impedance curves. See how much higher in frequency the upper impedance peak is when the enclosure is only 64l? This will make the system sound very boomy.

Compare one of the woofers in the 64l enclosure tuned to 16Hz versus a sealed 64l enclosure. The vented system has only a 2dB advantage between 25 and 45Hz. Now compare the impedance curves. They look almost identical.

Now compare the step response simulations. The sealed box woofer is much better. Because of the enclosure being too small, the port is barely adding any output to the system, but is making the transient response much worse.
Reply
Reply
post #83 of 274
9/10/08 at 1:54pm
Frequency response simulation of one woofer in 64l sealed versus 64l with a port tuning of 16Hz.
LL
Reply
Reply
post #84 of 274
9/10/08 at 2:02pm
Quote:
Originally Posted by Jack Hidley View Post

Lennon,

I see what you are saying now. I think I must have looked at the closed box numbers at one point, without realizing it, when I was trying to figure out what assumptions Unibox was making. It's frustrating having a spreadsheet and not being able to look at the source code to see what is really going on.

64l per woofer tuned to 16Hz really isn't going to work well. This is the type of alignment that gives bass reflex systems a bad name. Compare the systems with a port tuning frequency of 16Hz in a 64l enclosure and a 200l enclosure. Compare the impedance curves. See how much higher in frequency the upper impedance peak is when the enclosure is only 64l? This will make the system sound very boomy.

Compare one of the woofers in the 64l enclosure tuned to 16Hz versus a sealed 64l enclosure. The vented system has only a 2dB advantage between 25 and 45Hz. Now compare the impedance curves. They look almost identical.

Now compare the step response simulations. The sealed box woofer is much better. Because of the enclosure being too small, the port is barely adding any output to the system, but is making the transient response much worse.

I figured it was just a simple mistake (I almost hated to call you on it - but I wanted to clear up any confusion).

Thanks for the insight on using the Step Response graph - I've always wondered how to use that. So basically the shorter amount of time it takes the step response to even out the better?

You wouldn't want to say a bit about how to read the impedance graph while you're at it, would you ?
Reply
Reply
post #85 of 274
9/10/08 at 2:43pm
A step function is defined as:

t<0 amplitude = 0
t>=0 amplitude = 1

So basically it looks just like a step that goes infinitely far to the right. For any speaker to reproduce this, it must have flat response all the way to DC. The faster the output of the speaker drops from amplitude = 1 to amplitude = 0, the less bass extension the speaker has. How the curve returns along this path is a function of the transient response of the system. The more it wiggles up and down, crossing the 0 line, the worse the transient response is.

There is much more information about a speaker system in the impedance curve than there is in the frequency response curve. Unfortunately, it takes a fairly complex understanding of the physics involved to interpret the information. I would get a copy of the LDC (Loudspeaker Design Cookbook) and the Loudspeaker Anthologies I and II from the AES (Audio Engineering Society). You can find everything you need in those books. The LDC is not too difficult to read. The AES books will be quite challenging unless you have a heavy math background.
Reply
Reply
post #86 of 274
9/10/08 at 3:34pm
Quote:
Originally Posted by Jack Hidley View Post

A step function is defined as:

t<0 amplitude = 0
t>=0 amplitude = 1

So basically it looks just like a step that goes infinitely far to the right. For any speaker to reproduce this, it must have flat response all the way to DC. The faster the output of the speaker drops from amplitude = 1 to amplitude = 0, the less bass extension the speaker has. How the curve returns along this path is a function of the transient response of the system. The more it wiggles up and down, crossing the 0 line, the worse the transient response is.

There is much more information about a speaker system in the impedance curve than there is in the frequency response curve. Unfortunately, it takes a fairly complex understanding of the physics involved to interpret the information. I would get a copy of the LDC (Loudspeaker Design Cookbook) and the Loudspeaker Anthologies I and II from the AES (Audio Engineering Society). You can find everything you need in those books. The LDC is not too difficult to read. The AES books will be quite challenging unless you have a heavy math background.

Thanks again for the insight. I should have known better than to think there was a short answer to that question. Some day if my interest in DIY speakers doesn't fade I'll have to pick up a copy of that LDC book. It seems to be pretty commonly recommended. I'd like to think I have a "heavy math background" but I'm not willing to bet the price of the AES books on it
Reply
Reply
post #87 of 274
9/11/08 at 3:22pm
Thread Starter 
Well gentlemen, I've finally done it. This should bring a tear to the eyes of Neo Dan, Jack, lennon 68 and anyone else that got tired of the design ideas!


I was having a problem making 4 boxes exactly the same size that would fit in the area and still look like they were actually designed for the area. Oh, and not stick out too far into the room.


Feast your eyes on my final thoughts:





These are sealed boxes. 60" x 18" x 13" deep. 6.42 cu ft that should give me right around 2 cu ft per driver after all is said and done. They go from the floor all the way to the ceiling. They will need to be bolted to the wall because they are only 13" deep.


Front shot:







Anything I should tweak before I start on these????

Reply
Reply
post #88 of 274
9/12/08 at 12:04pm
no tears

Looks well distributed, should work nicely.
Reply
Reply
post #89 of 274
9/12/08 at 1:50pm
Thread Starter 
Tears of JOY Dan.....tears of joy!
Reply
Reply
post #90 of 274
9/12/08 at 3:44pm
I hope I'm not a killjoy but....
When I built my original 3-way speakers I added 2 Dayton Reference 12" subs to the top of each tower. Now I had 2 12's up high, 4 in total(sealed) and found the bass way to thin all the way up there. I ended up cutting them off, selling them and building my subs seperately. I was caught up on making things look pretty instead of real world listening.

Are you getting any oomph from those higher up subs.

I hate to see you spend all this time and effort on subs with that high of placement and not get the best out of them.

My apologies though as I have not read through this thread and don't know your restrictions.

Here is a pic of my original speakers with the 2 12's on top. They were powered by a Behringer ep2500.




Mike
Reply
Reply
New Posts  All Forums:Forum Nav:
  Return Home
  Back to Forum: DIY Speakers and Subs

AVS Top Picks

AVS › AVS Forum › Audio › DIY Speakers and Subs › Twelve 10" NHT Subwoofer Build.