An idea for studio monitoring

[Consul]

Off-axis response isn't as big a deal for me as it might be for other people. These will be soffit-mounted, and the soffits will be built at angles so that the speakers point right to the listening position, forming an equilateral triangle. All listening will be right on-axis. Here's a Sketchup picture to show the basic idea:


(Just click on the thumbnail for the full-sized pic.)

I'm thinking about the BG planar with that Dayton woofer I posted earlier. Since I plan to bi-amp with an active crossover, I can play with different crossover points more easily.

 

[usekgb]

That should work pretty well, but don't expect a ton of SPL's if you're crossed over below 1k. Take a second look at the power handling of the BG planer, and you will realize that it can't handle a ton of power at lower frequencies. I'm not saying it's a bad idea, but you will be limited by power handling.

Cheers,
Zach

 

[Consul]

Well, I could get one of these:

http://www.madisound.com/cgi-bin/index.cgi?cart_id=9132322.16025&pid=1037

And cross it at around 1.8khz.

Of course, by then, the advantages of the planar tweeter are mostly gone. I'm thinking I could get the Seas 27TDFC and waveguide-load it, like Zaph describes here:

http://www.zaphaudio.com/Waveguidetmm.html

I'm also thinking that I don't really need high volumes for a small room like this.

 

[Consul]

Hrm...

Although Zaph got a lower total harmonic distortion loading the tweeter thus, the main advantage is in allowing a simpler crossover in order to get a nice second order slope. That won't be an issue with an active crossover. The only other advantage I can think of with the waveguide is time alignment with the woofer.

 

[mhelin]

[quote author="Consul"]Well, I could get one of these:

http://www.madisound.com/cgi-bin/index.cgi?cart_id=9132322.16025&pid=1037

And cross it at around 1.8khz.

[/quote]

It would be fine, though you better make a notch for the resonance of the L18 too. Even though you used LR4 (24 db/oct) it's still there. You can also solder the required notch with C (4.7 uF) and L (0.10 mH) in series directly to the driver making it a passive filter and still use the active xover.

Another popular metal (alu) cone woofer is the Dayton RS180 - see htguide.com forums, they have some discussion on it like in this thread: http://www.htguide.com/forum/showthread.php4?t=13154&highlight=rs180
Many also built the RS180 with the Seas 27TBFCG tweeter.

Dayton RS180 at PE:
http://www.partsexpress.com/pe/showdetl.cfm?Partnumber=295-364

These are cheap <$40, actually $30 if you buy 4 or more. In MTM it might be fine, though in MT I'd use the L18RNX/P (in fact I've got Amphion Argon2 speakers at home with a L18 driver and they are excellent!).

I've tried unsuccesfully use large (2-way with 8" woofer and 1.1" tweeter in waveguide crossed over @ 1.7kHz) speakers in small room. They don't work because there are so many nulls in room response if you place the speakers like the nearfields. Best solutions for small rooms are either the soffit mount (large or small) speakers or small nearfields with a subwoofer augmenting the bass.

 

[mhelin]

Here's discussion on the MTM design using RS180 and 27TDFC (which can be replaced by 27TBFCG with minor changes in crossover):

http://www.htguide.com/forum/showthread.php4?t=11321

The tweeter position on front panel is meant to minimize diffractions, in soffit mount design you may place the tweeter in the middle between the woofers.

E: Another budget MTM with DA175 and 27TBFCG:

http://www.zaphaudio.com/BAMTM.html

DA175 sells for $16 at PE:
http://www.partsexpress.com/pe/pshowdetl.cfm?&DID=7&Partnumber=295-335

 

[Consul]

Okay, so today, I got, via Inter-Library Loan, the 4th edition of Dickason's Loudspeaker Design Cookbook.

A chapter and a half in, and my brain's already overloaded.

Nevertheless, I will keep at it as much as I can, and see what I can come up with.

 

[Viitalahde]

Darren, by now you've probably discovered that the directions you can go to are endless. :grin:

I am knee deep in it.. I *might* have decided with the AOS kit, but on the other hand..

http://zaphaudio.com/ZD5.html

 

[Consul]

The ZD5s are a little ways out of my price range, unfortunately.

The BG planars are still on the list to investigate. I want to get further into the book before contacting a few people with questions.

Otherwise, waveguide-loaded 27TDFC along with the L18 (H1224) would probably be an unbeatable combo for the price.

 

[Consul]

I think I'm going to stick with the Zaph-inspired Seas L18 (H1224) + 27TDFC in a waveguide-loaded configuration, bi-amped with 4th-order active crossovers at 1.8khz. The only thing I might need to deal with is the peak at 7khz, with a notch filter which I can do at the crossover.

I'm still iffy on whether I need the waveguide on the tweeter. According to Zaph, it does help reduce distortion down in the lower operating range, so I guess it couldn't hurt. I'm also thinking that setting the tweeter back the distance required for waveguide-loading would help phase-align the drivers.

Of course, I could still be pissing in the wind.

I'm also thinking I should include the bass reflex port, like in Zaph's design.

I've got a copy of Vance Dickason's Loudspeaker Design Cookbook. I'm giving it a read, but it's not easy.

EDIT: There's still a part of me, though, that wants to experiment with a tri-amped three-way system using something like the CSS FR125S as an extended mid-range. http://www.creativesound.ca/details.php?model=FR125S

 

[Consul]

You know, I really don't know why I'm focusing so much on this. I can't do anything about constructing soffits and speakers until the Spring, anyway. I should be worrying about my modular synth, instead. :wink:

 

[mhelin]

[quote author="Consul"]I think I'm going to stick with the Zaph-inspired Seas L18 (H1224) + 27TDFC in a waveguide-loaded configuration, bi-amped with 4th-order active crossovers at 1.8khz. The only thing I might need to deal with is the peak at 7khz, with a notch filter which I can do at the crossover.
[/quote]

The waveguide loaded 27TDFC in series of the cap has already 2nd order response so you need only 2nd order active crossover for the tweeter. The cap is useful as it protects the tweeter in case of DC. Alternatively build 3rd order filter for the tweeter and 4th order for the woofer. I would use 2nd order for both like in original but build active crossover just for the woofer.

I'm still iffy on whether I need the waveguide on the tweeter. According to Zaph, it does help reduce distortion down in the lower operating range, so I guess it couldn't hurt. I'm also thinking that setting the tweeter back the distance required for waveguide-loading would help phase-align the drivers.

You are right. If you are not going to use the waveguide then you may have to use the 4th order crossovers for both elements.

I'm also thinking I should include the bass reflex port, like in Zaph's design.

It's fine as long as you tune the port low enough not to cause large group delay at bass frequencies. I'd say 30-35 Hz is fine, there's not very much these frequencies in musical material at these frequencies. Use WinISD to see the tuning effects (group delay, max. SPL etc.). If you need high SPL use smallish closed boxed so it will protect the woofer though will increase the Q which may be what you don't want - there's always the Linkwitz transform circuit which can exceed the bass response and take care of the raised Q.

Even though you are not going to build them now you could first build some cheaper boxes like the Zaph MTM's using the ready made PE boxes (saves time), Dayton $16 woofers and Seas tweeters. So you'll be more experienced later when you're going to build the final soffit mounted speakers. Also IMHO you should give a listen to the 27TBFC's (the aluminium ones), they are much more accurate compared to 27TDFC which are kind of "easy" tweeters. Maybe use both and switch on the pair of the tweeters on(/off) depending on what kind of sound you want (soft, easy to listen or more critical).

 

[Consul]

Reading through Dickason's book, it's amazing to me just how much of speaker design really is just an art in determining the right compromises to make. At least the physics of the whole thing are well-understood enough to give me some good starting points.

One thing I wasn't sure about is what the result of waveguide-loading the 27TBCF aluminum tweeter might be. I suppose that's one place I can experiment in the coming months.

For my first speaker, I'm thinking about doing something simple with a full-range driver. This one looks pretty good:

http://www.vikash.info/audio/FR125S/index.asp

Those are pricey drivers, though, at $70 apiece. Well, relatively pricey, anyway.

Thank you for your help, mhelin. I now know enough to know that I do not know enough. :wink:

 

[Kev]

I'm not ignoring this thread and I have read and been following all of it.

The basic pretence of the wide range MID speaker and then fill the tops and bottoms is an approach that has been done before.

I have always had a soft spot for this theory/method.
ATC - control rooms
Turbo Sound (TMS series) - PA systems

it has been done before.

The only advice I have at this stage is that picking the MID driver is so very important as the character of this drive will leave a very large foot print on the overall sound.

Be conscious of it's sound IN the range that you will use it AND look to it's OFF AXIS response. In other words don't just pick it for it's sound in FULL range mode while looking directly down it's throat.

The Behringer Digital unit could be a very good way of checking combinations quickly.

Last point.
Be vary aware of the maximum volume you want to achieve. It's ALL about cone excursion.

.....
oh !!

and keep reading Vance Dickason's Cook Book
:thumb:

as you get closer to choosing the driver/s
your more direct questions can be answered by Thomas or myself.

 

[Consul]

Hi, Kev! Thank you for being around!

That CSS FR125S has a one-way Xmax of 6mm. That's quite a lot, from what I'm reading.

I saw a design on the web that used a pair of CSS WR125S speakers (a cheaper version of the FR125S with less output up top) combined with a tweeter per side. $50 apiece time four for the WR125S, plus cost of a pair of tweeters (the design I saw used a Fostex), in vented cabinets embedded in the wall...

Might be doable.

I remain amazed at the many, many places I can take this whole idea.

 

[Consul]

Oh, hey! The WR125S comes in a 16 ohm version. Two of those in parallel would do the trick!

http://www.creativesound.ca/details.php?model=WR125ST

Along with this Fostex Tweeter (Madisound has them at about $41 apiece):

http://www.fostexinternational.com/docs/speaker_comp/pdf/ft207d.pdf

I'm thinking I can cross them at about 4khz, using an active 4th order crossover. Then I could play around with tuning the port.

 

[Kev]

the wide range driver you are looking at, is typical of the modern stuff with the phase plug
Peerless have one and so do others

very nice to see the 16 ohm version if you are to try a dual approach.

remember that dual will produce an interference pattern with it's associated problems and lobbing etc.

dual will mean that you can ease off on the cone excursion for a given volume/power

... a one-way Xmax of 6mm

some manufacturers say +/- xx mm and some say total excursion
here we have one-way
...
double check what you are looking at.

A ported enclosure ? ... for the wide range drivers ? obviously not the tweeters.

If you are to use a true woofer then I'd look at a sealed enclosure for the Mid/Wide as it will simplify and make things more predictable.
That's what I think.
Even a rear port wouldn't be my choice.

An enclosure at critical volume or one larger than needed so you have room to adjust.
This is all assuming that this area is below where you want to use it.

FS of 65Hz is way below your 200Hz ... ish
even with a sealed box you don't have to come up more than 120Hz which is an octave and you are still almost an octave below your 200Hz.

juggle the numbers and the Mid Chamber can be quite non critical for volume.
Concentrate on internal standing waves that are in line with the cone dimensions and half dimensions ... can explain.


As for the tweeter ...
Fostex

err
Fostex

this is probably made by someone else

The PDF looks fine ...and 90dB/1w/1m is ok
but I might keep searching for a unit from the usual suspects
... and perhaps just little more grunt/strength
more robust ?

 

[Consul]

Since I'm setting these up to have a fixed listening position, I don't think I need to worry too much about vertical lobing.

The two WR125ST's on each side would, in theory anyway, fill in all the bass I need, probably with the help of a port. If I can get decent response down to 40hz, I'm happy. They just don't have much up top, hence the additional tweeter.

Of course, the "tried-and-true" course would be to stick with the L18 + 27TBFC combo. I might even be able to do 2 x L18 in an MTM configuration.

Again, thank you much for your help, Kev. :guinness:

 

[Consul]

Heh...

http://www.madisound.com/cgi-bin/index.cgi?cart_id=3747986.22877&pid=1960

I'm half-tempted to get a pair of these to try them out with active crossovers, if only to see how well they do. I could always save them for my car, otherwise...

 

[Consul]

Hey, Kev...

Why couldn't we use a computer as a digital crossover?

At first, when I heard of digital crossovers, I thought that was kind-of a corny idea, because it's an extra layer of conversion: signal out through D/A, then converted back to digital for processing, then back out to analog. Of course, it then suddenly occurred to me that you could just feed a digital crossover with a digital signal out from the DAW machine. Duh. :wink:

So, why not use a separate, dedicated computer for the task? It wouldn't have to be a screaming machine at all. Hell, an old 500Mhz something or other would still be more than enough. One could probably make a Pure Data patch to do the actual crossing over. Then you'd have all the wave-shaping options you could possibly want for notching out peaks caused by cone break-up and all that jazz. Part of the system, of course, would be in keeping the signal digital from the DAW to the crossover machine.

I know I'm not the first person to think of this, as I've already found crossover DSP software doing a Google search.

Whaddya think?