Top contenders in their class 15,18,21 inch drivers - Page 4

I was having the exact same problem. My internet supplier was out today and fixed the problem. It was a bad connection at the tapping box on the pole. They installed a new connector and now my internet is working great again. For several days it was slow as can be. So slow that sites would not load. It would time out. Very frustrating.

I posted parts of this on another thread and at different times. But in the context of this subwoofer I thought it really is interesting. I changed the math slighly to correct a problem on the last number that was missing in the first power progression.


The distortion in most subs is due to driver nonliterary pushing them so hard that they are no longer in control of the motor structure design. The ways to get around that are rather simple. Use more drivers. But to have any appreciable difference we have to remember that to drop an octave in bass we need four times the volume of air moved to keep the SPL equal. You need quite a few woofers. The second solution is to provide a better acoustical match between the driver and the air in the room. This is the method that is being discussed in this thread. I don't recall seeing a figure of actual box gain over raw data from Patrick's posts. But in the horn I'm fooling with the gain is between 8 and 10 db over the raw driver. That is an increase of efficiency of just over and around four times that of a driver on it's own mounted in a conventional enclosure. So the driver in a horn has to move one quarter the amount of air to get the same sound pressure level of that same driver in a vented enclosure.

Gain? Well the distortion is cut down dramatically. More than the 1/4 you would think. If you use a driver that has a very linear motor structure like the XBL type in the TRIO12 you are getting great control of the driver over a far longer stroke. A stable BL is very important in keeping all forms of distortion low. The BL is the number of turns of wire in the voice coil that are saturated by the magnet and motor structure. It is the best measurement of how in control the cone remains at any position. It is also a curve that most manufacturers will not provide.

No matter how much power you throw at a driver it is only a repulsive force from the existing motor structure. So the inherent linearity of the driver itself is very important. The other very interesting thing about drivers and actual speaker boxes is that the more power we push into the driver the more they heat up. Then the voice coil parameters change drastically. If we can create an enclosure where the driver is better coupled to the room we are in the required power is greatly diminished and the resulting sound from the driver/enclosure system is cleaner. There is no warmed up voice coil that is now two to three tikes the resistance of one at room temperature. The amplifier does not have to try and push gobs of power into a voice coil that is literally compressing the power instead of tracking it. Nope we can get a driver that will track the music signal and produce what we are after. The idea that we can take a driver and push hundreds of watts into it and get a louder response is more a myth than reality.

Here is the power versus SPL progression

Base line 90 db one watt.

1-2-4-8-16-32-64-128-256-512-1024-2048 watts

90-93-96-99-102-105-108-111-114-117-120 -123 in db

These are all increments which increase the sound level by 3 db.

This is a power output progression showing the power required to raise a signal 33 db. So from about 90db to about 120 db.

There are no drivers that can sustain this amount of power input. Not even pr sound drivers. Those with 4 and even 6 inch voice coils can take a steady 1000 watts and peaks of 1500 to 1800 watts. So the usual answer when you want to get louder is to use more drivers. Every doubling of drivers will take you back a notch in power requirement. So 2 drivers could produce 123 db with 1024 watts. 4 drivers 512 watts. Etc.

The greatest magic in getting things louder happens below 128 watts. Between 128 watts and 2048 watts there is a 12 db rise in sound pressure level. A little over twice as loud acoustically.
From 1 to 128 watts there is a 21 db increase from a 1 watt level in sound pressure.

Now lets examine the same requirement in getting this loud with the horn subwoofer.
If your bass line power is say 96 db then you get this funky thing happening

96-99-102-105-108-111-114-117-120- 123 in db
1 - 2 - 4 - 8 - 16 - 32- 64-128 -256 -512 in watts to get to that level of db.

You get very loud with not so much power when you drive a woofer in a horn. The not so much power enables greater ability to generate peaks without there being compression electrically in the driver itself or mechanically from there being no more stroke for the cone.

Trade off?

Big box. The one in the picture is the 23 foot horn it is 16.5 " x 32" x 92". Only addicts may apply! But once you hear one of these brutes all other sub-woofers sound.....neutered is the best term.


Ricci posted about back EMF.

And that is the source of the large peak in the impedance. But not all of the source.

The coolest part about the driver at or close to resonance is that it is very efficient near that point. That's why with so little power input they still make great bass.

The Bill Fitzmaurice horn is a great example of a well optimised horn. It has quite a short path length yet still provides great low end. The closed box behind the woofer acts as a spring for the driver cone correctly sized to a driver and horn as a system the sealed back can really improve the low end of the horn.



Mark

I have to disagree, I think 500w-ish (each) is the magic number for subs hopefully in multiples of two subs and 12" in size or greater, rather than 129w

120w-ish is the magic number for satellites

"They are in the top 4 sub drivers I know of so far."

:-)

the other three?

Noah, thanks for filling in the gaps in my brain! I never thought to link thermal compression to rising Q, although that is clearly a measured phenomenon.

As for the "LT" part, I'm not sure what you mean. For the listening sessions with the W15GTi in a progressively smaller cabinet volumes, the subs were EQ'ed to the same target curve. And it sounded pretty much the same until the box was so small that (modeled) Qtc was higher than an LS3/5a's.

Just wanted to chime in and say:

Great Thread! Thanks!

One question - we have power to weight (mms), what about power normalized to cone area? Does that have any bearing?

Seriously?

As Scott alluded to, the performance is dominated by the horn on the front.

I thought you said something to the effect that the W15GTi made LT (which in my mind = EQ) unnecessary.

Once it is normalized to cone area you are basically working your way back to efficiency.


You do realise that if you listened to the two power levels you would notice not that much difference in loudness. It would sound louder but after a minute or so you would hardly notice it. From 128 to 512 watts is only an increase in 6 db. You will find most people discern an increase when there is a change by 3 db.

So the 128 watts make sense now Noah?

It is not an arbitrary number but the point in power versus loudness that gives the biggest bang for the buck.

MArk

i'd put the elbow in the bang-for-the-buck power curve at somewhere between 2kw and 4kw +/- given current amplifier pricing. even an el cheapo ep4000 has some pretty big caps.

a better measure would seem to be bang-for-the-buck spl, as that would also consider the efficiency of the speaker/sub in combination with the amplifier. but then we have hoffman to deal with all over again. :-)

128 watts, without qualification, seems a bit arbitrary in my mind.

Ricci

I am having this same issue for the last week or so. Sucks(though seems better tonight)

Going by efficiency of drivers that are legitimate subs.
TC 5200 18
21nlw9600
21sw152
AE TD18H+ (gets in there with a crazy light moving assembly and ultra low inductance coupled with a strong motor.)
There are a whole bunch of drivers that are just a bit less.

I hear you man. I've got some posts I'd like to make with attachments and I I can't. The connection is lost before I can finish composing things. Lame.

ricci, i like your list. but just to argue in the spirit of learning, you have disputed the value of the efficiency equation in the past, so why not the titanic 400c-4, the mtx 5400, the sdx 15, or the mach 5 18.2.2.? these seem to have good bang for the buck more or less. the reason that i press the argument is simply to glean insights from you and/or others. all in good faith. :-)

fyi, no probs with the avs site during the past several days. something else is going on.

That's a list of most efficient, not necessarily best for an application. I don't dispute the efficiency thing. It can have too much made of it sometimes though. Sometimes you need to look at other aspects or where the efficiency is greatest. Also I've learned a lot and changed some of my views in the last year or 2. The efficiency equation which Im still not sure exactly what it needs to be is derived from at least 5 basic parameters. I've been thinking about it since I got more interested in horns. None of the drivers above have that special something that set them apart imho.

interesting ricci, so give us some more...what are you thinking...about all this...

set horns aside, as they seem to be an oddball. weak motor gives a choppy response, but so does an overly strong motor...

And - I'm thinking I'm wandering down that same trail, but I am a good bit behind you. Bear with me - I said I was wandering.... I'm sort of thinking out loud here.

This is why I asked what I did earlier. Thanks for the comment Mark, that is sort of what I am getting at, but I think there might be a little more to it than that. Certainly the "power to weight ratio" is a great analogy, but it can't be the whole picture. It is an incredibly useful way to classify drivers though. That power has to be "transferred" to the air in the room (or within the enclosure) in order for the speaker to perform its best. I guess this would be sort of like a "payload capacity", in other words, how much "load" can the motor "pull" with that much cone area.

It seems to me that one can get too much motor for a given cone area, making that woofer less useful in some types of enclosures, yet well suited for other varieties where the acoustic load on the cone can be increased. This ratio may only be useful to suggest the type of enclosure (horn, tapped horn, bandpass, ported, sealed, etc) that the driver might be happiest in.

While the opposite is also true (too little power for cone area), that sort of driver is not what this thread is about. I have lots of those drivers.....

The question that Ricci has answered to degree and I'll flesh out a bit more.

The Qts or the total system Q is a great parameter to look at for these reasons.

Higher Qts usually means smaller box
Lower Qts usually means higher efficieny and a larger box.

There are always exceptions that is why I put in usually.

The ways that the low Qts can be created relate to the motor strength and the moving mass. Sort of!

One of the largest contributors to the Qts after moving mass is BL. Now seperate out the B or flux density from the L or winding length.

The closer you can get to a magnet the stronger the coercive force. The corresponding "closer" within a loudspeaker motor is the distance around the outside of the coil in relation to the top plate gap. Thge donut ri nthe coil passes through. A tight gap will give you a high BL. That will lower the Qts. That will raise the Fs and increase the size of box to get a decent low end. What the small gap will also do is create a problem with the coil rubbing the top plate. Tough nut to crack. You can pour in magnet size or strength as in coercivity like a smaller Neo structure. But all things must be balaced out to get a combination of parameters that will allow a decent stroke with a decent efficiency.

Qts will give you a ball park in how the driver designer did this.

So will the size of the magnets.

How?

Take some of these monster car audio subs that have triple magnets 60 lbs motor structures. They also have very heavy cones and extremely stiff suspensions. Really low 1 watt SPL's to boot.

When I have a bit more time today I'll post something on those monsters to.

I think this is how I view this thread.

Give a man a fish you feed him for a day. Teach a man to fish and you feed him for life.

Learn what makes a good driver and all the flavour of the month reccomendations will be out the window.

Go fish!

Mark

Looking forward to your follow up post.

Precisely to my point... what happens in front of the box far outweighs what happens inside the box.

I've used dozens of over hung, under hung, inner hung, single, double, triple spider, low dollar, high dollar, small diameter-tall stacked magnets, large diameter-short stacked magnets, metal cone, reinforced paper cone, long throw, shorter throw, high efficiency, low efficiency, high Qts, low Qts, high Fs, low Fs, high Cms, low Cms, etc drivers, and there is a recurring and inescapable reality with all of them...

Performance is dominated by the alignment and it's adaptability to the environment.

The driver and it's magic motor/suspension/cone/basket are far less relevant than these discussions suggest.

I remember the discussions and subsequent conclusions surrounding Ilk's test of the LMS in 75L vs 200L:


Many pages of discussion and subsequent builds doubled in size for an average difference of 1.37dB from 10-80Hz for a tripling of box size.

A lesser driver may easily yield better actual performance results than either of those in any actual room with the addition of any number of optional exterior additions (such as putting a horn in front of it, Auto EQ, placement options, multiples for equal cost, better amplification, multiples for same size, configuration of the alignment, etc., etc.).

Again, I'm stating the obvious here, but given that it's obvious, the points of what to do about the supposed huge trade offs in the way of problems that arise from fighting iron laws and added heat, compression, distortion, loss of output, etc, they keep popping up even though they're not only not significant in the grand scheme, they seem to lead to better actual performance in the end.

Bosso

I have had this problem for awhile now.

The trick is to start the post, and place all images off site. Now you can compose the entire message at your leisure, copy it, then hop back in hit reply to thread, and paste it in really quickly, then submit.

Obviously, since 99% of our listening is <128 W.

Compared to your other interesting/informative postings, I'm just not getting what the insight is here.


Isn't Fs is a purely mechanical function of mass and compliance?

Me too. I have been doing this for awhile. Lately it's gotten so bad that I can't even do that. Site drops out waiting for the reply to the post screen.

That is pretty bad. I went in and changed a lot of my personal setting, and things are running a lot better right now.

Now I just get logged out if I can't type my post our in under 2min, and I stay logged in for ever if doing anything else?

"Again, I'm stating the obvious here, but given that it's obvious, the points of what to do about the supposed huge trade offs in the way of problems that arise from fighting iron laws and added heat, compression, distortion, loss of output, etc, they keep popping up even though they're not only not significant in the grand scheme, they seem to lead to better actual performance in the end."

what are you saying here bosso? you seem to have said that these things are obvious, not significant, but lead to better performance...

I agree with much of the above but I do disagree that the driver is not a big part of the equation. Why would anyone use an alignment that handicaps the better driver and allow a different much larger alignment for a cheaper driver? The better driver like an lms for example is usually more versatile in the alignments it's suited to. All of the drivers that I consider to be top class offer a very high amount of potential power density that the cheaper drivers don't. Again it might only be an average of 1.5 db for a whole lot more money, but it is still an improvement from the same enclosure size. Any build that I might look at the first thing I start with is the enclosure volume available. To my way of thinking the question in this thread is not what can a driver do in one enclosure versus another but what can each driver do in a certain enclosure size? How many drivers either alone or in multiples do you think are going to outperform an lms in overall performance in a 20" cube?

"How many drivers either alone or in multiples do you think are going to outperform an lms in overall performance in a 20" cube?"

wouldn't two av15h's give the lms a run for the money but at about 1/2 the cost in such an enclosure?

Hi noah katz


Nope.

Think of the flux strength as acting like a shock absorber. The lower the Qts the greater the in gap BL. Again this is a generality but pretty wide spread generality. The higher flux strength will dampen the spring system that is the surround and the spider. Note I say dampen. A dampened spring will have a higher resonant frequency. Therefore push up the Fs.

I just caught the reply Ricci made about the importance of alignments.

In a nutshell they make or break what even the best driver can do.

If you don't pay carefull attention to what a driver is best suited to as Ricci already pointed out, you end up with a compromised box.

Drivers can be tailored to cover a wide range of possible alignments. But there are only 6 real types of alignments that of any use. Sometimes the size diference between them is significant as is the performance.

Work with a program like WinISD for a while and you will be able to take the same driver and pop it into a bunch of box alignments. There is always one that shines above the others.

There is a reason why this field is still an art. Computers and programs are not intelligent makers of choices. They widen the available choices but it is the people who understand the choices that are the ones who have to display the intelligence.

Mark

I'd take four AV15h's over a single LMS 18 anyday.