Best 15inch/38cm woofer units?

[voodooless]

so I'll use 3 separate circuits and 3 amps for drive each (HPF-BPF / BPF / LPF)

Crossover between amps will be active then?

 

[thin bLue]

Crossover between amps will be active then?

purifi 400/700/700 amps with DSPs

 

[FrantzM]

I have to reiterate this. So far, we have no idea what the actual goal of this speaker system would be. We have no idea of what problem you are trying to solve here.
For example, I am working on a compact, high-efficiency monitor design which incorporates a horn tweeter. That is designed to solve the problem I have of having small listening spaces and wanting a speaker that can fit there, but also play loudly in larger rooms.
My specs for that design were: Point source (or very close to that), high output capacity above 2-400hz, and relatively compact size.

Here, I have no idea what you are aiming for. I don't know if you are trying to build some kind of open-baffle system, or a surround sound system, or whatever else. All I know is that you are apparently asking whether some random 15" driver will "match" some other (seemingly randomly picked as I don't know your intention) drivers.

I can have a terrible woofer, and match it with my KEFs because I know how to integrate it. That's what's important. Getting the integration right matters. There are a million good quality 15" woofers out there. You just need to find the one that fits what you want to do. Do you need bonkers output? Then get a PA woofer that can handle 2000w. Do you want super deep bass? Then look for a woofer with a low Fs and high xmax.

Basically what I'm saying is that there are tons of good drivers out there and the actual implementation matters more than the components you select. The design of the crossover, the geometry of the enclosure, etc. All that plays into how the speaker sounds and makes a much larger difference than the difference between a good B&C 15" and a good JBL 15".

So with that being said: What are your design specs? What is the "problem" you are trying to solve with this speaker? (Where / how are you going to use it, etc)
If you define the design specs, then we can help point you in the right direction.


--------------------------
This reminds me of a time I gave a joke answer to someone on a different forum. They asked for recommendations for speakers for a large festival. Naturally, I answered with a technically correct answer that would have been a really bad idea. (Use a bunch of Danley Jhericos [the ones with basically no vertical dispersion] hung from some helicopters. Then add sufficient MatterHorns for LF duties...)
Here, I am actually trying to help because this forum is quite nice

@thin bLue

Not able to tell you much except for these:

This post is what @sarumbear has been trying to tell you in his posts: You must present your goal for any recommendation to make sense. The best drivers in the best box won't make a great speaker, if you don't know what you are aiming at ...aka "Design Goals"
Yet, you went crude on him.
Not cool.

Out.

Peace

 

[fieldcar]

I've got a couple questions for the DIY'ers now that we are fired up and ready.

• Can I assume that the largest flaw in the passive 4 way design for this case is the large gap in the sensitivity between the 8" woofer ([email protected]/1M) and the rest of the drivers (~[email protected]/1M)? Would you have to attenuate the tweeter, mid, and especially the 15" woofer to match the poor efficiency of the 8" woofer. I imagine a complex network of high watt resistors just for the 15" woofer. This would generate higher than average losses in the amplifier, correct?
• I get skiddish with driving a tweeter directly to an amp fed by an active DSP crossover. I've only tinkered with this in software DSP like PeaceEQ from a computer, and this is probably why I'm so skiddish about it. I guess this is just my traditional style of thinking that passive crossovers essentially protect drivers. Though, you can't argue with the lack of phase shift and the 5-20% (or potentially more in this case) power loss of a passive crossover network. I guess a dedicated miniDSP would quell those fears.
• Any recommendations for a JBL 2216Nd or 2216Nd-1 alternative for a 2-way design with a selenium compression driver? I've found a few sites with TS parameters for the JBL. This is the best I've come up with. I'm sure that I'm missing a few things though, since it doesn't quite work in WinISD.
  
  
  
  

 

[voodooless]

I've got a couple questions for the DIY'ers now that we are fired up and ready.

• Can I assume that the largest flaw in the passive 4 way design for this case is the large gap in the sensitivity between the 8" woofer ([email protected]/1M) and the rest of the drivers (~[email protected]/1M)? Would you have to attenuate the tweeter, mid, and especially the 15" woofer to match the poor efficiency of the 8" woofer. I imagine a complex network of high watt resistors just for the 15" woofer. This would generate higher than average losses in the amplifier, correct?

Doing this passively is madness. Active is the way to go. Sensitivity of the woofer is largely irrelevant then.

• I get skiddish with driving a tweeter directly to an amp fed by an active DSP crossover. I've only tinkered with this in software DSP like PeaceEQ from a computer, and this is probably why I'm so skiddish about it. I guess this is just my traditional style of thinking that passive crossovers essentially protect drivers. Though, you can't argue with the lack of phase shift and the 5-20% (or potentially more in this case) power loss of a passive crossover network. I guess a dedicated miniDSP would quell those fears.

Just add a cap as protection and all should be fine.

• Any recommendations for a JBL 2216Nd or 2216Nd-1 alternative for a 2-way design with a selenium compression driver?

Why specifically Selenium? Probably get something with at least 1.4/5” exit, lowest possible extension. Horn wil dictate the rest.

• I've found a few sites with TS parameters for the JBL. This is the best I've come up with. I'm sure that I'm missing a few things though, since it doesn't quite work in WinISD.
  View attachment 210397

The trick is to not enter all values. I always enter Qes, Qms, fs,Vas, mms, re, le, xmax and Pe. It will do the rest for you.

 

[thin bLue]

@thin bLue

Not able to tell you much except for these:

This post is what @sarumbear has been trying to tell you in his posts: You must present your goal for any recommendation to make sense. The best drivers in the best box won't make a great speaker, if you don't know what you are aiming at ...aka "Design Goals"
Yet, you went crude on him.
Not cool.

Out.

Peace

Thanks. You are right.

 

[fieldcar]

Just as a cap as protection and all should be fine.

Oh, right! As long as the tuning frequency is below the active range, it's just protection.

Why specifically Selenium? Probably get something with at least 1.4/5” exit, lowest possible extension. Horn wil dictate the rest.

Mainly to keep it "somewhat JBL", but I'm not tied to any specific brand. Maybe I'll start a thread if I get serious about a build.

Another reason why I was looking at selenium drivers.

The trick is to not enter all values. I always enter Qes, Qms, fs,Vas, mms, re, le, xmax and Pe. It will do the rest for you.

Thanks! I'll give that a shot. Now the hard part. To convince the wife to let me do the build in the first place.

 

[thin bLue]

I've got a couple questions for the DIY'ers now that we are fired up and ready.

• Can I assume that the largest flaw in the passive 4 way design for this case is the large gap in the sensitivity between the 8" woofer ([email protected]/1M) and the rest of the drivers (~[email protected]/1M)? Would you have to the attenuate tweeter, mid, and especially the 15" woofer to match the poor efficiency of the 8" woofer. I imagine a complex network of high watt resistors just for the 15" woofer. This would generate higher than average losses in the amplifier, correct?
• I get skiddish with driving a tweeter directly to an amp fed by an active DSP crossover. I've only tinkered with this in software DSP like PeaceEQ from a computer, and this is probably why I'm so skiddish about it. I guess this is just my traditional style of thinking that passive crossovers essentially protect drivers. Though, you can't argue with the lack of phase shift and the 5-20% (or potentially more in this case) power loss of a passive crossover network. I guess a dedicated miniDSP would quell those fears.
• Any recommendations for a JBL 2216Nd or 2216Nd-1 alternative for a 2-way design with a selenium compression driver? I've found a few sites with TS parameters for the JBL. This is the best I've come up with. I'm sure that I'm missing a few things though, since it doesn't quite work in WinISD.
  View attachment 210397

1.Make fully passive attenuation with 3 drivers make massive heat, difficulties of design, lack of efficiency. match them by Amp gain is much much reasonable and realistic solution for that.

2. For me, protective passive circuits seems necessary. Mini dsp and every dsp has possibilities to fail. (Sometimes it just reseted by accident, that case without caps, LF will immediately roast tweeters.

3.Well, I don't know. sorry. someone who knows will tell you.

 

[DWPress]

I used a miniDSP 4x10 for years to do my active XO and from experience will say you definitely need a protection cap on that tweeter no matter how your system is set up. Even with active XO you still want to match the sensitivity of the drivers and voltage gain of the amps as close as you can to minimize problems, no free lunch.

I will once again remind you that 6 cubic feet is larger than most European refrigerators. If I were to do it all over again I would go smaller, better to pass the LF to subs anyway in most rooms and there are plenty of even 10" drivers out there that will reach 30Hz with room gain.

 

[sarumbear]

I will once again remind you that 6 cubic feet is larger than most European refrigerators. If I were to do it all over again I would go smaller, better to pass the LF to subs anyway in most rooms and there are plenty of even 10" drivers out there that will reach 30Hz with room gain.

Good point. I cannot see any reason for a 15” driver on a modern speaker. Amplifier power is cheap. Many manufacturers demonstrated that multiple 10” drivers are perfectly capable for bass in normal size room. You can reach over 110dBSPL at 30Hz with a pair of 10” drivers (Xmax: 20mm).

 

[TulseLuper]

I cannot see any reason for a 15” driver on a modern speaker.

In the words of OP, "big towers with big units make sexy-look aren't they?"

Can't argue there.

 

[DWPress]

M

In the words of OP, "big towers with big units make sexy-look aren't they?"

Can't argue there.

Me neither but would still build smaller now if I were to start over:

 

[dualazmak]

Hello OP @thin bLue,

I hope my multichannel multi-driver multi-way multi-amplifier stereo project thread would be of your reference.

You may find the latest system configuration and various related info as of May 30, 2022 in my post here (#540) on the project thread.

And, please find here (on the project thread) and here (remote independent thread post) the Hyperlink Index of my project as well as some of my related posts on remote threads.

 

[OWC]

I would also not recommend going passive at all.

If you have the space for 15 inch, it doesn't really matter that much.
Most of any (very) descent 15 inch will perform extremely well, especially compared to any 8 or 10 inch subwoofer.
Make sure the Fs is low enough (< 30-35Hz), Qt is low and they have enough cone excursion (> 6mm or so)
4 ohm is nice to have because of the voltage the amplifier needs to spit out.

The way I would set it up, is just make a nice 2-way system with the tweeter and woofer, ideally with a waveguide, otherwise you will never be able to get a good directivity out of it.
Second is ideally have 4 subwoofers, otherwise 3. If you can set up the two front ones in stereo, that would be ideal, best even in the corners of the room.
Because of this "stereo" approach of the subwoofers, we can move our crossover point all the way up to even 200-300Hz if we really need.
This is one of the best approaches to get low distortion, good directivity (with the waveguide) as well as being able to get rid of those very audible and horrible room modes.

 

[thin bLue]

In the words of OP, "big towers with big units make sexy-look aren't they?"

Can't argue there.

 

[thin bLue]

I used a miniDSP 4x10 for years to do my active XO and from experience will say you definitely need a protection cap on that tweeter no matter how your system is set up. Even with active XO you still want to match the sensitivity of the drivers and voltage gain of the amps as close as you can to minimize problems, no free lunch.

I will once again remind you that 6 cubic feet is larger than most European refrigerators. If I were to do it all over again I would go smaller, better to pass the LF to subs anyway in most rooms and there are plenty of even 10" drivers out there that will reach 30Hz with room gain.

Of course, that's I'm already aware of.
But ,Thanks for reminding me!

M


Me neither but would still build smaller now if I were to start over:

View attachment 210437

WOW

BTW, presses look so nice

 

[thin bLue]

Hello OP @thin bLue,

I hope my multichannel multi-driver multi-way multi-amplifier stereo project thread would be of your reference.

You may find the latest system configuration and various related info as of May 30, 2022 in my post here (#540) on the project thread.

And, please find here (on the project thread) and here (remote independent thread post) the Hyperlink Index of my project as well as some of my related posts on remote threads.

Thanks for info!

Your system feels very cool
Look at that array of VU meters!!!

 

[thin bLue]

I would also not recommend going passive at all.

If you have the space for 15 inch, it doesn't really matter that much.
Most of any (very) descent 15 inch will perform extremely well, especially compared to any 8 or 10 inch subwoofer.
Make sure the Fs is low enough (< 30-35Hz), Qt is low and they have enough cone excursion (> 6mm or so)
4 ohm is nice to have because of the voltage the amplifier needs to spit out.

The way I would set it up, is just make a nice 2-way system with the tweeter and woofer, ideally with a waveguide, otherwise you will never be able to get a good directivity out of it.
Second is ideally have 4 subwoofers, otherwise 3. If you can set up the two front ones in stereo, that would be ideal, best even in the corners of the room.
Because of this "stereo" approach of the subwoofers, we can move our crossover point all the way up to even 200-300Hz if we really need.
This is one of the best approaches to get low distortion, good directivity (with the waveguide) as well as being able to get rid of those very audible and horrible room modes.

As subs Fully agree with that.

Thanks for great advices for DIY

Of course, I'll use multiple subwoofers to cure in room responses.(Esp. below Schroeder Frequency.)

But big towers with big units make sexy-look aren't they? Acoustical gain maybe marginal, but psychologically.

How about 2-way vs 3-way?

2-ways are great!
3-ways can do that as 2-ways do, even with lower THD, but way harder to design.

So I prefer well designed 3-way over well designed 2-way.

 

[OWC]

As subs Fully agree with that.



How about 2-way vs 3-way?

2-ways are great!
3-ways can do that as 2-ways do, even with lower THD, but way harder to design.

So I prefer well designed 3-way over well designed 2-way.

You can go for 2-way with 15 inch, something like Geddes did.
Disadvantage is that the beaming will be very high, but it will get you a very nice directivity when combined with a 12-15 inch horn.

 

[gy-k]

With the right driver, a 3-way can be built with a very small back volume for the mid cone driver. This is interesting because small is easier to build, and box resonances of a small a mid chamber are going to be high in frequency, even up to 1kHz, which is easier to mitigate probably. Very low Vas, low Qts drives that'd work for this best are going to be higher sensitivity, so better match for something like a 15" woofer. Also such drivers have high Fs, in theory this might be good for excursion limiting effect from the sealed response, higher resonance and roll-of also works better for a passive crossover. (With the right sealed response could be even just one cap, and a notch filter to flatten the impedance.) On a similar note, one 15" that has been mentioned has Vas of 400l, that's some difference compared to 110l of something like a B&C driver, I mean smaller is easer to build.
If you have the mid and tweeter time aligned (by using a waveguide for example) and you're able to get nice curves from a passive crossover, using DSP crossover would be more convoluted in comparison. You can have passive LPF on the low, and passive HPF on the mid and high and use separate amp channels for them. Sensitivity could be matched either way with simple line-level attenuation before the amp, and DSP could be used upstream for additional tweaking.