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

I don't even have a TV anymore, nor cable or dish. Been like that for close to 10 years.

Projectors have come down in price, with 1080HD sometimes cheaper than the LCD TV counterpart. I couldn't come back to watching movies on a tv screen anymore. I love my 110" diagonal screen!

We mostly watch movies, and the occasional TV show that are available through online.

And yes, arrays for movies are just great and immerse you with what's happening on the screen, feels like watching as if you were down there:

An externally hosted image should be here but it was not working when we last tested it.
 
Last edited:
Better point of view!

This was (seems to be "is" now as it has been rescued from destruction) the best movie experiences in my life. I believe it was one of the first (or THE first) to get THX in Canada.

481609d1430987183-making-two-towers-25-driver-full-range-line-array-imperial.jpg
 

Attachments

  • imperial.jpg
    imperial.jpg
    70.9 KB · Views: 925
bringing the theatre home

But sometimes I wish I had a bigger TV screen :D.

I couldn't come back to watching movies on a tv screen anymore. I love my 110" diagonal screen!

An acquaintance of mine owns a TV and appliance store where I live, he is tempting me to buy a 4K monitor. I have seen those as big as 110", but big $$$$$ :eek: Sure be nice for Graphic Arts.

I have not paid for "TV shows" in over 20 years. People at work talk about them all the time and I have no clue. I enjoy movies when I get a chance, but mostly, I use my "TV" as a monitor on my computer. My daughter enjoys it for streaming Anime and playing games. I enjoy it for Graphic Arts. It even helps for typing in these small boxes diyaudio uses, at least I can see it without reading glasses :)

Perceval, that is a beautiful theatre, I am really glad they did not tear down such a well crafted piece of historical theatrical art. I would be curious to see how they set up the main front speakers to accommodate the listeners in the balcony. I see the surround speakers on both levels. We have a much smaller, but equally well crafted theatre in Ironwood, MI.

full-view-from-balcony.jpg


It was converted to a cinema in the movie era, but has been completely restored back to a stage theatre for plays and concerts. I actually had the honor to set-up and run the sound for a classic rock band in this theatre. It sounded great, but there are a lot of acoustical challenges we needed to overcome (note the concave ceiling). I am really glad the theatre sound man was there to give me a few pointers (like the standing wave frequencies to tame). Unfortunately, the closest thing to THX I will ever hear in Ironwood, will be what I build in my living room. ;) With a population of about 5000 and 75% of those being over 65, there is not a big market for audio fanatics...

Bringing this back to the topic of arrays and perceived soundstage, I can say that the video recordings of my daughter's school band recitals in this theatre has a more true to life scale when played back through the arrays then what is does though the Avebury.

I tried EQing the Avebury with a 3 dB 1.5kHz centered bell curve with a Q of .25 (swallow slope). I enjoyed this more then the surround effect, as it did extract more of the spacious, lively sound of the recording without taking away from the direct, focused sound of this system. Yet, I still found myself not convinced. It just did not engage me...

The conclusion I have with these recent discoveries, especially regarding soundstage, is that there is a physical scaling element at play. How can one expect a singular 3" to 8" full-range or mid-range recreate a true to life soundstage? In fact, taken farther, can listening in the far field re-create a properly scaled soundstage? Far field listening sounds distant to me. How can it be an encompassing sound field if it sounds far away? The sound is "over there". Only option here is surround sound. But I still think even with 11.2 point source speakers in a Dolby Atmos system, there will still be a "detachment" or a distancing to what we are hearing with the far field sources. Much like we need to move a certain volume of air to feel 20 Hz bass, I believe we need to move a certain "area" of air to produce a realistic soundstage. We need to be listening near field. Maybe other people's minds do not need as much convincing as my mind, so maybe a far field source is enough for them. As for me, I am glad I have made this discovery, as I know for sure my next build will be a line source. I am itching to know what a 4 channel surround system sounds like all arrays. :D

You could say, I am completely sold on the line source/near field concept, no going back now...
 
That looks like a sweet little theater! Must have been fun (and challenging) running the sound there!

The Imperial (that's the one I posted) also has a stage and gets either shows or movies. Had to look into it. It was built in 1913. 7 years after the very first palace theater in Canada (also in Montreal). I knew the guy calibrating the system there, so he pointed to me the best seats in the house to get the best sound possible. :) I saw the first Star Wars there in 1977, and the last Star Trek with the original crew there also.

I am also aching to start a new line source myself. I have a set of sweet speakers (TB W8-1772s), and they sound nice, but nowhere engaging as my 16-driver arrays were.

I was in contact with the VIFA factory in China last year... they were out of drivers but were waiting 6 to 8 months before doing another run. I contacted them lately, and it seems they don't not have much inventory now either.

Getting them from Parts-express and shipped here is prohibitive, so I am looking at local drivers. There's one that isn't bad. I sent an example to X for his second round of 3" driver shootout. It's not as flat as the TC9, with a bump at 2KHz, but it has some nice audio qualities. I'm tempted to use them in my next build. They would be an upgrade from my first shot at arrays, when I used cheap local $3 drivers... and it still sounded really good!

I've been looking to find a CNC store that would cut the front baffle for me. Once I do... I don't know how long I can refrain from starting another build!
 
I don't even have a TV anymore, nor cable or dish. Been like that for close to 10 years.

Projectors have come down in price, with 1080HD sometimes cheaper than the LCD TV counterpart. I couldn't come back to watching movies on a tv screen anymore. I love my 110" diagonal screen!

We mostly watch movies, and the occasional TV show that are available through online.

And yes, arrays for movies are just great and immerse you with what's happening on the screen, feels like watching as if you were down there:

An externally hosted image should be here but it was not working when we last tested it.

Beautiful theatre! I just love old buildings like these... Must have been quite impressive in real life.

If I were living on my own, the choice to own a projector would be easy. About 18 years ago I regularly had to set up the company projector on location whenever it was needed for a conference at location. That would mean theatre time for me throughout the following weekend! At that time the box containing the projector sitting in my passenger seat was more expensive than the car it was in. And it wasn't a really cheap car... it was this one:
An externally hosted image should be here but it was not working when we last tested it.

(it also explains why the box was in the passenger seat ;))

I haven't found a good way yet to bring up this subject again with my girlfriend... Ever since those weekend movie runs long ago I wanted one for myself. Mainly to see concert footage with the occasional movie thrown in.
 
Last edited:
Array project starting soon...

So, I am preparing to design my next array.

What would you do different with your current array designs? (I know someone mentioned that they would have each driver in their own isolated compartment.)

Also, if money was no object, what would you do?

I just started a new thread, but I am posting here also. I figured I would pick all of your brains before branching off, as there are many happy, experienced array builders here. You can post your options here:

http://www.diyaudio.com/forums/full...ing-full-range-array-project.html#post4318765

Right now, I am lending towards the Fostex FF85wk for the array down to 80 Hz, then bringing in a sub-woofer.

My second choice would be the Celeston AN3510, as I really like the driver design for line array applications. This would also need a sub-woofer at 80 Hz.

Crazy expensive, would be the Alpair 7, but I could run that full-range. I would need to cut the bezels down to get closer spacing :eek:, Mistakes would be quite costly.

The Foundex FE88EX can be run full-range, but listening impressions have lead me to believe that this unit has a rougher high compared to the Fostex and Alpair options.

It seems the compromise is if You want full-range, You must sacrifice the highs a little. If You want the highs, You will not have the low of the lows.

Thoughts anyone?
 
Last edited:
One thing I would do different is the baffle and maybe even the tower. As can be found in this thread I posted an impedance graph taken from the array. It never entered my mind that the resonance frequency of the array and baffle length (and the actual length of the array itself) were that close to each other. In hindsight that wasn't the smartest choice.
The baffle length in centimetres is 220,7 cm. That corresponds to a wavelength of 155.4 Hz. The 25x TC9 all have a resonance frequency close to this, depending on actual damping in the enclosure. It resulted in a (albeit small) double peak in the measured impedance curve of my towers:
impedance%20right.jpg


I managed to reduce the effect somewhat but in hindsight I'd cut the (back) baffle into 5 separate pieces. Not as easy to do afterwards for me, as You'd need to still have it seal up the back chambers.

Lucky for me, a lot of the choices I made I can still support. I spend a lot of time on the damping of the back chamber/wave and the above impedance plot shows I was successful to damp the back wave as this becomes evident in an impedance plot. I spend a week on tests with that little "one speaker" test box trying different combinations. I never completely disclosed what my found recipe was. The actual damping materials I used in my tower results in 5 different ones to get the best effect(!).

For a driver I'd look at the drivers with the cleanest waterfall plot in the high frequencies, but also the best off axis plots 0-30 degree. The cheap Vifa actually does very well in both, but perhaps better can be had.
The Scan Speak 10F is very clean in waterfall CSD plots. It's off axis behaviour does look a bit worse (on factory plots) than the TC9 surprisingly. But I wouldn't pick any driver with ringing at high frequencies that can clearly be seen in the CSD plots.
Think about it, for one driver it wouldn't be a great deal and one can even correct it some with DSP power. But for an array, the spread of the high frequencies becomes larger due to the difference of the drivers to your ear. The ringing would also last longer as a consequence. No DSP could fix that for a 25 driver array.

This should get you started. A lot of my thinking is right here in this thread. I still think I did rather well but I had to after such a long period of thinking about it. I stand behind the concept I made, my driver choice and my crazy way of building the towers. Including the aluminium baffle material and the beauty front plate with the filleted holes. The impedance thingy with the baffle length is the most important factor I've thought about where I would make another choice.
 
Last edited:
Opel GT! :) Never thought I'd see that in this thread.

I have a thing for sports cars from that time period. I owned 2 Opel GT's at that time, this blue one and a yellow GT with the black bumpers (though without the junior stripes).
In the morning I'd choose which one would take me to work :D. They were my daily drivers. Always been funny like that, taking a 30+ year old car as my means of transport.
My current car is a 1982 Porsche 911, has been my daily driver for 9 and a half years. That one has a very good stereo on board which led to me doing this project: 1982 Porsche 911, first build. simple system - Car Audio | DiyMobileAudio.com | Car Stereo Forum
I also own a VW Karmann Ghia, which is still more of a project (I'll probably never finish ;)).
 
Ooh, a Karmann Ghia!! One of the nicest ones ever! I would love to see it finished with your skills :)

That one was supposed to be equipped with a Type IV engine (VW Transporter) with cylinder heads from another project of a friend and myself... see www.apfelbeck.nl
230 HP with a 2 litre engine would be possible.

So many projects and so little time... or money, always short on one or the other...
 
Well to be honest, in my current situation I don't see how I could finish the Karmann, nor the engine. The engine project is currently on hold (has been for a few years) but hopefully we will be able to pick it up again. We started the engine project when both of us didn't have kids, now that we both have kids it has been difficult to get something done on that project.

I kind of lost a bit of interest in finishing the Karmann as with all my plans for it, I was sort of building a Porsche, and I already had one of those :). I still love the look of the Karmann Ghia though. Maybe even more than the 911. But the 911 is a great car that I enjoy very much. I can only afford that one by doing all maintenance and repairs myself.
I need to sell the Karmann project, but am very reluctant to do it.

But even the 911 isn't in mint condition after 9 years of all year drive in the Netherlands. But I'd hate to let it go. Still I have to find a new source of income, which has been hard. One would think with this many hobby's I should be able to make some money on one of them :D.

In other words, not much reason to envy me. ;) Just struggling to keep it all together like most.
 
Last edited:
Founder of XSA-Labs
Joined 2012
Paid Member
One thing I would do different is the baffle and maybe even the tower. As can be found in this thread I posted an impedance graph taken from the array. It never entered my mind that the resonance frequency of the array and baffle length (and the actual length of the array itself) were that close to each other. In hindsight that wasn't the smartest choice.
The baffle length in centimetres is 220,7 cm. That corresponds to a wavelength of 155.4 Hz. The 25x TC9 all have a resonance frequency close to this, depending on actual damping in the enclosure. It resulted in a (albeit small) double peak in the measured impedance curve of my towers:
impedance%20right.jpg


I managed to reduce the effect somewhat but in hindsight I'd cut the (back) baffle into 5 separate pieces. Not as easy to do afterwards for me, as You'd need to still have it seal up the back chambers.

Lucky for me, a lot of the choices I made I can still support. I spend a lot of time on the damping of the back chamber/wave and the above impedance plot shows I was successful to damp the back wave as this becomes evident in an impedance plot. I spend a week on tests with that little "one speaker" test box trying different combinations. I never completely disclosed what my found recipe was. The actual damping materials I used in my tower results in 5 different ones to get the best effect(!).

For a driver I'd look at the drivers with the cleanest waterfall plot in the high frequencies, but also the best off axis plots 0-30 degree. The cheap Vifa actually does very well in both, but perhaps better can be had.
The Scan Speak 10F is very clean in waterfall CSD plots. It's off axis behaviour does look a bit worse (on factory plots) than the TC9 surprisingly. But I wouldn't pick any driver with ringing at high frequencies that can clearly be seen in the CSD plots.
Think about it, for one driver it wouldn't be a great deal and one can even correct it some with DSP power. But for an array, the spread of the high frequencies becomes larger due to the difference of the drivers to your ear. The ringing would also last longer as a consequence. No DSP could fix that for a 25 driver array.

This should get you started. A lot of my thinking is right here in this thread. I still think I did rather well but I had to after such a long period of thinking about it. I stand behind the concept I made, my driver choice and my crazy way of building the towers. Including the aluminium baffle material and the beauty front plate with the filleted holes. The impedance thingy with the baffle length is the most important factor I've thought about where I would make another choice.

Regarding the rear chamber of a tower, I am convinced that my 3-sided pyramid Dagger is perhaps one of the cleanest low coloration rear chambers you can find and it is compact and easy to make. From the data and sound clips in the 10F/RS225 thread you see and hear it is very low coloration. I think 25 of those daggers pointing out the back of a baffle (or covered in a casing) would serve very well in a line array.
 
An impedance test of the 10F in the pyramid Dagger would demonstrate how clean it is.
I'm surprised not more people use that simple test to see what's really happening.

Take an impedance sweep of the bare 10F and compare it to the sweep in the enclosure. Trust the data, not the ears ;).

I was very pleased with my end result, but that won't mean you can get the same or similar results with a less challenging back chamber.
But the volume is also very important if you want the most out of the low side of the driver. I got lower with the resonance peak than my
WinISD predictions predicted.

Your Dagger somewhat resembles the strange back chamber of the Scaena (if you fold it). But it would be a bit harder to get the volume
up using a 3 sided pyramid.
 
Last edited:
Founder of XSA-Labs
Joined 2012
Paid Member
An impedance test of the 10F in the pyramid Dagger would demonstrate how clean it is.
I'm surprised not more people use that simple test to see what's really happening.

Take an impedance sweep of the bare 10F and compare it to the sweep in the enclosure. Trust the data, not the ears ;).

I was very pleased with my end result, but that won't mean you can get the same or similar results with a less challenging back chamber.
But the volume is also very important if you want the most out of the low side of the driver. I got lower with the resonance peak than my
WinISD predictions predicted.

Your Dagger somewhat resembles the strange back chamber of the Scaena (if you fold it). But it would be a bit harder to get the volume
up using a 3 sided pyramid.

Your wish is but my command... :)

481800d1431080087-making-two-towers-25-driver-full-range-line-array-dagger-10f-8424-compare-impedance.png


The Dagger (blue) actually smooths out the bare driver impedance (purple) ripples somewhat, it also raised the resonant frequency, and lowered the overall impedance. If it were a bigger chamber, it would approach Q of 0.71, for this volume it is at 0.61.

What do you think?
 

Attachments

  • Dagger-10F-8424-Compare-Impedance.PNG
    Dagger-10F-8424-Compare-Impedance.PNG
    103.3 KB · Views: 4,525
Last edited:
That's the kind of proof we need! Very good! A slightly bigger chamber would be best. This driver really intrigues me. It has really clean CSD, good X-max, good distortion figures... I'd only try to see if off axis plots can be made a little cleaner out to 30 degrees.
Cleaner and more extended in that they follow on axis plots a little better. The filleted hole in front of the driver helped the TC9, it might help this one too. 0 to 30 degree is all I ask.
Mounting the 3 sided pyramids in an alternate fashion could still guaranty the closest possible centre to centre distance. Might have to shave the round bezel a bit to get them really close. What's the outer dimension, 97.5mm? My heart to heart is 85 mm. Mainly chosen to have some meat between drivers to seal the back chambers.
I'd still pack all those Daggers in a diffraction reducing outer shape at least from the baffle back with a rounded corner 1" or greater. But you could even make it pretty.
Make the baffle in 5 pieces to not have to worry about it's own resonant behaviour.

Any takers? :D
 
Founder of XSA-Labs
Joined 2012
Paid Member
Everyone should have a DATS or WT3 from PE. One of the best, and most useful instruments a diy'er can get for $100. It works in seconds, no fuss. I literally can make TS measurements in 30 seconds. I leave it hooked up to the PC and the software is always open and ready to take data. It has a nice LCR feature where you can measure any mystery capacitor, resistor, or inductor accurately. Great for making your own BSC or passive XO's where you aren't sure what the part values are, or for "rolling your own" homemade inductors or fancy boutique foil/paper-in-oil caps.

DATS and a UMM-6 (and REW) have really changed my diy capabilities.

10F external bezel dia is 3-13/16 inch diameter. It does not have countersunk screw holes and I am using kind of big screws that protrude out a bit. The TC9/TG9 bezels are square for closer packing. I could have put C-C spacing of dagger and woofer closer but left room for internal bracing to prevent the main front-back "breathing" mode of vibration. The C-T-C spacing is 8in which corresponds to 500Hz 1/3-rd wavelength, so as long as I XO under 400Hz, I should be OK. I will take some polars in the next few days...
 
Last edited:
Autos, Audio, Aveburys and Art...

One thing I would do different is the baffle and maybe even the tower. As can be found in this thread I posted an impedance graph taken from the array. It never entered my mind that the resonance frequency of the array and baffle length (and the actual length of the array itself) were that close to each other. In hindsight that wasn't the smartest choice.

Wow, who would have thought! Amazing observation! Once I pick out the driver and start designing the cabinet, we will review this to double check.

Regarding the rear chamber of a tower, I am convinced that my 3-sided pyramid Dagger is perhaps one of the cleanest low coloration rear chambers you can find and it is compact and easy to make.

The dagger in this thread? http://www.diyaudio.com/forums/full-range/273524-10f-8424-rs225-8-fast-ref-monitor.html

Everyone should have a DATS or WT3 from PE. One of the best, and most useful instruments a diy'er can get for $100

Yes, I plan on getting that, just so I can run that impedance test. Like Wesayso, I will be building some mock up cabinets before committing to the final. I have heard that Aperiodic cabinets can eliminate that impedance spike altogether. Is that true, or is that "snake oil?"

I'd still pack all those Daggers in a diffraction reducing outer shape at least from the baffle back with a rounded corner 1" or greater. But you could even make it pretty.
Make the baffle in 5 pieces to not have to worry about it's own resonant behaviour.

Any takers? :D

In the midst of the amazing Theatre and Auto Art on this thread (I would have picked the Yellow Opal everyday, by the way! :) ), I will share a little of my own.

For my array cabinet, I am planning on about 4" radius corners flanking the baffle. This will make the cabinet about 12" wide, but because of the radius and the height, this cabinet will still seem quite "slim". Of course none of this is in stone, as I have not selected a driver yet. :) I am looking forward to doing one of my crazy labor intensive mosaic designs for the cabinets for the "pretty" aspect.

The first photo below are the diffusers I built for Avebury, the second is the solid oak and maple mosaic front I made over the Avebury BB plywood cabinets. Still lots of sanding to do on those. :cubist: The last one shows a fully finished box I make for my Mum. I have this odd thing for making patterns out of hard woods. :D

Do You think I am up for the challenge?

Thanks everyone for your input so far, it is all very valid and appreciated.

Those "daggers" might be a little easier then my diffusers ;)

Hang in there, Wesayso, it took me two years of grueling searching to get the career I have now. In a town with only 5000 people and long winters, not much need for Art. But none the less, my gifts are put to use in the job I have. It is not the most glamorous work, but my detail oriented, no cutting corner work ethic was greatly needed to clean, sanitize and prepare the rooms for medical procedures at our local hospital. There are lots of opportunities for advancement also, so I consider myself greatly blessed. Yours will be coming soon...

Allen
 

Attachments

  • diffuserpair3D.jpg
    diffuserpair3D.jpg
    250.6 KB · Views: 387
  • HT Right Avebury.jpg
    HT Right Avebury.jpg
    913.4 KB · Views: 384
  • MumsBoxDone.jpg
    MumsBoxDone.jpg
    187.4 KB · Views: 187
Founder of XSA-Labs
Joined 2012
Paid Member
Aperiodic boxes will reduce the impedance peak, as the dagger brought it down. It won't eliminate the natural resonance frequency of the driver though. What can completely flatten it is a impedance flattened circuit. I forget what they are called? Maybe a Zobel but at fs. Byrtt has shown me some pretty nice examples he did on his drivers.

Venting the tip of the Dagger will flatten the Impedance somewhat further but introduces an asymmetry to the peak. Not sure if I can hear it but I prefer the sealed for now.