[mjbuoni]

Hi all, I am a hobbyist speaker designer and wanted to ask for any comments, links or advice you can provide around a design I am considering.  The midrange / tweeter section consists of an array of nine 4" Tangband W4-1337SDF drivers (wired as 3 series x 3 parallel, acting as a single driver) flanked around an Aurum Cantum G1 ribbon as shown in the attached picture.
.  Crossover will be in the 2.4-3kHz range, resulting in a fairly narrow vertical listening window near the crossover point.  Also, the G1 ribbon is nearly 6" long, so it beams vertically especially in the top octave.  I've seen several designs that augment the G1 ribbon with a rear-firing ambiance "super tweeter" to fill in the top octave or so spatially.  And I've also read that it is important not to cross the ambiance tweeter over too low, as the imaging will lose focus.  All that makes sense, and I will experiment to find what works best.

My question to you Danny, or anyone with relevant experience, is what seems to work best for such a rear ambiance tweeter?  In terms of: directivity pattern, crossover frequency and slope, relative level to front firing tweeter, or any other considerations that you may be aware of.  I found an interesting article describing a similar design with the ambiance tweeter turned sideways: http://www.northcreekmusic.com/Manifest/ManifestInfo.htm.  Another interesting design is by Von Schweikert: https://www.vonschweikert.com/vr-9se-mkii. 

Thanks in advance for any feedback!

Matt

[Oscillate]

Hi Matt ...welcome to the forum.

I can't really answer your question about the placement of a ambiance tweeter.
But if I may share some other things ... I believe your choice and placement of
the 9 Tangbands around the AC ribbon driver will yield a 'lobing effect'. Also the
extra width it requires of the front baffle will mitigate the speaker's ability to 'image'.

I know in the past there had been a brief discussion of a slightly different design
where a tweeter in the center of a front baffle was encircled / ringed by numerous
midrange drivers. I forget Danny's take / opinion of that, but I thought that was an
interesting design.

[mjbuoni]

Hi Matt ...welcome to the forum.

I can't really answer your question about the placement of a ambiance tweeter.
But if I may share some other things ... I believe your choice and placement of
the 9 Tangbands around the AC ribbon driver will yield a 'lobing effect'. Also the
extra width it requires of the front baffle will mitigate the speaker's ability to 'image'.

I know in the past there had been a brief discussion of a slightly different design
where a tweeter in the center of a front baffle was encircled / ringed by numerous
midrange drivers. I forget Danny's take / opinion of that, but I thought that was an
interesting design.

Thanks for the feedback.  The issues you raise are valid concerns, and I'll start by saying that I've implemented this same basic configuration with four drivers in a C-shape around a dome tweeter on a 12" wide baffle (above drawing is 14" wide baffle), crossed over at 2.8-3kHz.  Sounds fabulous, and images like some of the best designs that I've ever heard.  With this previous C-shape four driver design, I experimented with adding felt strips around the tweeter to reduce diffracted energy due to the relatively wide baffle (which I agree can smear imaging).  Better though I found is that the asymmetrical arrangement of drivers acts to "dither" out diffraction effects, and a relatively high crossover point and selecting a tweeter with a shallow waveguide reduced much of the omni-directional tweeter energy that interacts with the baffle.  To the issue of lobing... Briefly, let me share that I have a physics and computer science background, and so I developed a piece of software that simulates the polar radiation pattern of my proposed designs (not shooting in the dark completely, here).  This nine mid-driver config will have some mild lobing near and just above the crossover point (lobe peaks down 7-8 dB from main lobe).  This is considered in the design as a tradeoff for the KILLER dynamics and very low distortion that it will surely have.  But all that said, this is also an experiment, and I don't know exactly how it will go until I implement, measure, listen, tune, ...

But back to the topic of my post: the ambiance tweeter is what I am most interested in hearing about, if anyone has some perspectives on this ...

[Danny Richie]

Before you move forward with such an experiment, I can hand you proven designs that will get you much greater performance and probably with a lower driver cost as well.

As for a rear ambient tweeter usually a good dome tweeter works pretty well as it will give you a nice even dispersion pattern. Yes, you will not want for it to play down to low or it will start effecting the on axis response. And you really don't want to use another tall ribbon as your vertical dispersion is already very limited by the tall one up front. Typically I only like the shorter versions so that you can have a more even in room response. I would save the taller ones for use in line arrays.

I haven't worked too much with rear ambiance tweeters in a while. At least not since we started using the planar magnetic drivers in an open baffle. That took things to a whole new level.

 

[mjbuoni]

Before you move forward with such an experiment, I can hand you proven designs that will get you much greater performance and probably with a lower driver cost as well.

As for a rear ambient tweeter usually a good dome tweeter works pretty well as it will give you a nice even dispersion pattern. Yes, you will not want for it to play down to low or it will start effecting the on axis response. And you really don't want to use another tall ribbon as your vertical dispersion is already very limited by the tall one up front. Typically I only like the shorter versions so that you can have a more even in room response. I would save the taller ones for use in line arrays.

I haven't worked too much with rear ambiance tweeters in a while. At least not since we started using the planar magnetic drivers in an open baffle. That took things to a whole new level.

I appreciate the feedback, Danny.  So I already have 30 of these drivers, because I purchased them on sale for about $20/each after being very impressed with how clean they sound and measure.  I want to use them to build L / C / R speakers for my home theater, as a mid in multiples for a 3-way system.  I already have the woofer section built in a separate enclosure.  An alternate idea I had was a short line array (again nine mids, but in a line, and possibly open baffle), but couldn't decide on the right line of tweeters to complement the mids.  I have owned the LS-6 from AV123 previously, and while very impressive in many ways, I always missed that last bit of air and extension on the top end, so the Neo8 isn't for me.

Danny, if you were constrained to use these drivers in multiples as a midrange down to around 250 Hz, what design and tweeter(s) would you use with them?  And if you weren't constrained to use these drivers, what proven design / drivers would you recommend that provide much greater performance?

[Danny Richie]

Okay, let me give you some options that will make what you are trying to do a lot easier and get you much better results. I am going to come back to this in a little while. I am covered up with e-mails and messages this morning.

[Danny Richie]

Okay, I have worked with all of those ribbon drivers and offered kits using a couple of them in the past. The new GR Neo 3's have quite a few performance advantages and sound better overall.

So you might consider using them like we did in the Wedgie, NX-Otica, and NX-Treme models.

For the Wedgie we crossed them to four 3" LGK drivers. The voice coil offset on the 3" drivers was 3/4" In other words 3/4" from the surface. So we designed a waveguide for the tweeter that was 3/4" deep to physically align it with the voice coils of the LGK's. That way the crossover design would keep the drivers in phase in front and in back as this was a full open baffle design.

See some Wedgie builds: http://www.audiocircle.com/index.php?topic=135561.0

http://www.audiocircle.com/index.php?topic=151056.msg1616048#msg1616048

Initial testing and development: http://www.audiocircle.com/index.php?topic=126112.0

Putting the Neo tweeter in an open baffle really takes it to another level. And no rear for rear ambiance add on's.

with the NX-Otica and NX-Treme models the offset was 1.5" to the voice coil of the larger M165NQ drivers. So a 1.5" deep waveguide was used.

You could do the same thing with your 4" drivers. Just cross to the center two drivers though and allow the additional drivers to cross to them in the 200Hz to 250Hz region where they handle lower ranges and fill in any baffle step loss. This keeps sensitivity high too. And it is ideal to minimize baffle width.

You could even drop in a Wedgie center section complete and let the larger TB drivers handle from 200Hz to 250Hz on down to the mid 80Hz range or so. You could use them in an open baffle too.

Oh, yeah and the GR Neo 3's can be crossed much lower as well. We cross them in the 1,300Hz range in the large waveguide of the NX-Otica and NX-Treme models. So the response all across the vertical plan is almost as flat as the on axis response.

 

[mjbuoni]

Danny, thank you for the suggestions!  This gives me several ideas to toss around.  I really, really like the idea of a 2-way line array of your new GR Neo-3's and the Tang Bangs, open baffle on the Neo-3's for sure.  What do you think of open baffle on just the tweeters down to around 2 kHz?  Compared to an open baffle mid + tweeter line array, how much "openness" is achieved by just making the tweeters open?

[Danny Richie]

Danny, thank you for the suggestions!  This gives me several ideas to toss around.  I really, really like the idea of a 2-way line array of your new GR Neo-3's and the Tang Bangs, open baffle on the Neo-3's for sure.  What do you think of open baffle on just the tweeters down to around 2 kHz?  Compared to an open baffle mid + tweeter line array, how much "openness" is achieved by just making the tweeters open?

If I were going open baffle then I'd go open baffle for everything. You will need something else to handle the lower region anyway with those drivers.

And I have tons of our new GR Neo 3 and will make you a deal on them if you wanted to run about 16 or more of them per side.

[Captainhemo]

Danny,
If  one was to  do this all OB,  wouldn't the  tweeters have to be in,on the backside of a long wave guide to  time align both the front and rear sides  of the   woofers/tweeters ?  Thought last time  we talked about the long wave  guide it was  a no go ... 

jay