The making of: The Two Towers (a 25 driver Full Range line array)

I know, I'm merely bringing it up as an example how you can avoid such problems with a bit of engineering. :)

I briefly tried with the missus, nope, no spheres for me ;).

What do they hit and at what level, I'm genuinely interested here. Also about seeing the impedance graph. I'm scouting a lot of sub builds to see what does what right now. Main interest is small sealed though, for various reasons.
 
Last edited:

TNT

Member
Joined 2003
Paid Member
It is virtually impossible to make a non round thing to not vibrate when you put a speaker in it. I'm sure you did a fine job minimise it.

I will try to measure. I have had problems with my impedance jig for a long time now.

I'll get back...

Small and sealed.

//
 

Attachments

  • bboll1.jpg
    bboll1.jpg
    107.7 KB · Views: 266
It is virtually impossible to make a non round thing to not vibrate when you put a speaker in it. I'm sure you did a fine job minimise it.

I will try to measure. I have had problems with my impedance jig for a long time now.

I'll get back...

Small and sealed.

//

I don't think it is. With the right bracing scheme.

As long as I'm typing I might as well add, the arrays go loud and clear, partly due to their design with multiple drivers sharing the load, partly due to the ideas that went into the enclosure. The subs will need a completely different approach.

Whatever I will build will have to live up to that performance of the arrays. I'm not willing to sacrifice that performance advantage with these added subs. They will be quite different animals which is why I currently look at the materials to use.

Building enclosures that do the whole spectrum justice isn't that easy. As each frequency spectrum has it's own quirks. Subwoofers are only playing a part of that spectrum and can be purposely build for that limited spectrum.
 

TNT

Member
Joined 2003
Paid Member
Found this. I believe this was the initial verifying measurements after build. One for each.

One was found to be leaking :)

Still, I will redo them. Drivers are Seas H.1208

There is no stuffing at all.

//
 

Attachments

  • bboll 1.png
    bboll 1.png
    651.4 KB · Views: 254
  • bboll 2.png
    bboll 2.png
    633 KB · Views: 245
Last edited:
Found this. I believe this was the initial verifying measurements after build. One for each.

One was found to be leaking :)

Still, I will redo them. Drivers are Seas H.1208

There is no stuffing at all.

//

It's obvious to see which one didn't behave like planned. Do you apply a boost below the peak, like a Linkwitz transform?
Any specific reason not to use stuffing?

Ronald, what were you looking for in the impedance plot when you were experimenting with stuffing?

Reducing peak Z at resonance? Or making the plot as smooth as possible, no blips? Thanks

The last one, smooth as possible impedance curve and avoidance of little bumps, blips. These usually translate into frequency curve problems. While it's relatively easy to check, not many look real hard at these graphs. Compare it to the driver in free air. It's worth the trouble.

Reducing the peak (and getting it to move to a lower frequency) helps for the power needed on the bottom end but wasn't the main reason for using impedance checks.
 
The reason for the sphere has noting to do with diffraction. It is the fact that when you pressurise a sphere, the forces on the cabinet will not make it bulge like if it was a box. This is the advantage. If I touch my subs while playing loud, I cant feel any vibration what so ever. To me, this means that what I'm hearing comes form the drivers and not the box. Magnets are glued together.

Do you not think that it is more to do with the dual opposed setup cancelling vibrations than the sphere?
 

TNT

Member
Joined 2003
Paid Member
It's obvious to see which one didn't behave like planned. Do you apply a boost below the peak, like a Linkwitz transform?
Any specific reason not to use stuffing?

Yes they are EQed. Had the idea to run them just the drivers as is. Noting added, nothing subtracted - well except for the (to small) volume of air as the load. Measurement is in listening position after eq.

//
 

Attachments

  • bb15.jpg
    bb15.jpg
    58.7 KB · Views: 248
  • bb6.jpg
    bb6.jpg
    36.3 KB · Views: 254
Last edited:
The last one, smooth as possible impedance curve and avoidance of little bumps, blips. These usually translate into frequency curve problems. While it's relatively easy to check, not many look real hard at these graphs. Compare it to the driver in free air. It's worth the trouble.

Reducing the peak (and getting it to move to a lower frequency) helps for the power needed on the bottom end but wasn't the main reason for using impedance checks.

Thanks!
The blips are probably from internal reflections hitting the back of the driver, so they should be at wavelengths similar to interior dimensions, and therefore pretty easy to damp.

I found that exercising (gently stretching) the suspension lowers Fs. I do it on all my woofers. I'll try it on one of these to see if it responds favorably.
 
Wesayso
Do you intend to use 2 stereo dacs and compensate the subs digitally?

I have 8 channels available on my Asus Xonar Essence ST, which sends it's first two channels trough SP/dif to my DAC. The subs will be fed of the Xonar Essence ST card for the time being.
So I have one stereo DAC and one 8 channel DAC with the 8 channel DAC responsible for timing. That's why I got the ST and not the later model STX.

Actually I already use 4 channels (with 6 channels of processing) right now.
The ambient channels are also fed by the Xonar Essence ST.

The subs will be corrected, yes. With FIR filters to align them with the arrays at the listening spot. It's going to be a puzzle, just the way I like it. :p
 
Thanks!
The blips are probably from internal reflections hitting the back of the driver, so they should be at wavelengths similar to interior dimensions, and therefore pretty easy to damp.

I found that exercising (gently stretching) the suspension lowers Fs. I do it on all my woofers. I'll try it on one of these to see if it responds favorably.

Yes, blips like that often come from internal reflections and can play a large role in the frequency response curve. I had a wavy internal shape but these full range drivers are sensitive to the slightest internal waves hitting it.
It's worth it to put some time into these kind of measurements. Especially if you want or expect top notch performance.

I bet playing all the way down like I do will naturally stretch the suspension over time :D. I still can't believe the bass they put out. My girlfriend was listening to a couple of songs yesterday evening and said: you want to add subwoofers to this?! That's insane! I promised her it wouldn't become louder, just more precise :devily:.