Sealed, Active, 2-Way or 3-Way Monitor Build | Advice on Drivers, Enclosure & DSP Welcome

Jump to Latest
#23
Not unlikely. I think budget is probably the main reason for my hesitation. Additional driver + bigger amplifier. I'm sure with some help I would manage the technical side of things. It's also challenging to estimate outcomes performance-wise. I suppose ultimately it's a question of which would bring better overall value...
 
#24
A 3-way shrinks each drivers' passband. Which makes less expensive drivers more viable.
An FA123 should be plenty of amp to do what you want and is only a bit more than the 122 you were planning on.

Here's a couple simple generic suggestions:
-Pick a DIY design that you like and convert it to active. This gives you a known output, and you focus on learning measurements and DSP design.
-Pick a 2-way bookshelf speaker pairing and add a bass section to make it a 3-way. The easy way to do this is attach a small sealed subwoofer.

If you want to implement a DSP design, you have some learning to do no matter what you do. If you're worried about having a satisfactory outcome at the end, the above two approaches should mitigate that concern.
 
  • Like
Reactions: svp and AlmaAtaKZ
#25
Good points! I might have to give serious consideration to your suggestions.

Here's my chicken-or-egg challenge:
Most DIY-designs are ported, but I'm firm on a sealed configuration as I really don't want to risk all work to end up with same bloat-issues that I've had before.

To give an example, this might work as a reference > http://www.troelsgravesen.dk/SBAcoustics-3WC.htm
  • How do I go about modifying that design into an optimal sealed cabinet?
  • As the sealed configuration limits bass extension of that woofer, some might argue going with a larger, maybe 10", woofer might work better, no?
  • Do I then calculate a sealed cabinet for the 10" woofer and consider it done or are there other parameters to consider?
I'm essentially trying to figure out the right approach and marching order.

I really appreciate the help and patience.
 
#26
You should download WinISD. It's good for basic enclosure modeling.
You'll eventually need to learn VituixCAD and REW, but start with WinISD for now.

Basically, you'll need to model a new sealed enclosure, with your planned DSP boost, and see if you can get enough SPL / bass extension with that combo.
 
  • Thank You
Reactions: super-panda
#27
super-panda said:
To give an example, this might work as a reference > http://www.troelsgravesen.dk/SBAcoustics-3WC.htm
  • How do I go about modifying that design into an optimal sealed cabinet?
  • As the sealed configuration limits bass extension of that woofer, some might argue going with a larger, maybe 10", woofer might work better, no?
  • Do I then calculate a sealed cabinet for the 10" woofer and consider it done or are there other parameters to consider?
Click to expand...
Sad reality is there are hardly any current woofers around with Qts optimized for sealed. In fact, probably the only one is the Eton 8-412, or the expensive Wavecor 10" I mentioned. With the Eton, it would be pretty straightforward to take many cues from Troels' 3WC format.
 
#29
I expected, at some point, to start second guessing my concept and here we are 😅
But since we're only just brainstorming, here's a bit of a left-field idea:
So basically going back to the coaxial 3-way idea. These would enable:
  • XOs at <200 hz and 1.6 kHz
  • Relatively controlled directivity
  • Decent bass extension, decent low-mid body
  • More than enough headroom, low distortion, good linearity
  • Monitor+ sized cabinet @ ~50-60l
  • Original budget
I think the Dayton sub is one of the more suitable options for this application. The coax arrangement is intriguing since the treble should have lower IMD than a traditional coax with the waveguide / horn?

What possible drawbacks am I missing? Treble response issues?
 
#30
I use a coax 3-way as my main speakers. I use the two drivers below, the subwoofer is in a very small enclosure to keep the whole thing stand mount sized.
Dayton Sub
SEAS Coax

The thing with coaxials is you need to be prepared to make compromises. In my case, I shaped the response for good off axis and predicted in-room response, and ignored the on-axis response (its rough) They are pointed straight out into the room, so I'm listening about 20 degrees off axis. Basically the same approach KEF took with the original LS50.

I don't know what you can get the compression driver sica coax for, but they also make one with a built in silk dome that will almost certainly be less expensive and more suitable for hifi use. Sica Coax

The polars for the Sica coax I posted look pretty good, but I would still expect a need for on-axis compromise. There is a flare in dispersion right around the crossover frequency that you'd have to figure out. Either ignored, or allow the response to dip a bit at crossover on-axis to smooth the off-axis.

Another budget option is to but a pair of KEF Q150s when they're "on sale" for ~$350 and pull the drivers out of them. Otherwise, Seas and Satori have coaxials.

This is what I meant when I said coaxials get weird. I like the way they image. But there are trade offs for sure.
 
  • Like
Reactions: super-panda
#31
The Sicas are interesting for sure and they look more promising than many other coaxes I've come across.
The 8" I linked in my previous post has that waveguide which I assume might remedy some coax-inherent issues but I can't find any measurements. I think the Antelope Atlas has a Sica coax and it's well reviewed, though that's not necessarily saying much. But I think I'm ultimately leaning toward a 3-way, coax or no. It just makes more sense to have something to fill the band between a 1" and an 8". And even more so a 10".
 
#36
super-panda said:
The 8" I linked in my previous post has that waveguide which I assume might remedy some coax-inherent issues but I can't find any measurements.
Click to expand...
I would not assume that.
There are a bunch more coaxials like that in the professional driver world, but they tend to be more expensive, or have very rough frequency responses. Not saying you couldn't find a gem in there, but I didn't. I also didn't look that hard.
 
  • Like
Reactions: super-panda
#39
Busdriver02 said:
I would not assume that.
There are a bunch more coaxials like that in the professional driver world, but they tend to be more expensive, or have very rough frequency responses. Not saying you couldn't find a gem in there, but I didn't. I also didn't look that hard.
Click to expand...
I hear you. Not assuming. I see the theoretical merit but practice is, of course a different horse.

I suppose it would be "safe" to use the https://sica.it/prodotto/5-5-c-15-cp/ as in the Antelope Atlas and pair that with a sealed-appropriate woofer. At least that's a pretty well received commercial monitor, so it shows some potential promise 🤷‍♂️
Does the rule of, generally trying to avoid crossovers in the 2 - 5k range, apply to coaxials? I assume since their phase behaviour is different, to separate drivers, this might not be as critical?
 
#40
That driver looks like it would work. I would think you'd be good with that and a reasonable woofer choice.

I'm not really sure what the "rule" is trying to address. I guess the thought is avoiding potential driver summing/coherence problems in the upper midrange / lower treble? If you go with coaxial, there is no summing/lobing/coherence issues since they're point source.

In any event, the tweeter in that sica has a resonant freq of 1500hz, which is likely why they recommend a 3khz cross. You can almost certainly bring that down a bit if you use steeper high pass slopes.

With an FA123, you have a DSP processor capable of FIR filters. Which means your crossover between mid to tweeter can be completely linear phase, and 8th order slopes if you want. Looking at the polar chart, it looks like bringing the cross down a bit would be useful off axis.
 
  • Like
Reactions: super-panda
You must log in or register to reply here.
Top Bottom