Low Sensitivity Speaker Thoughts?

[DS-21]

This is from what, page one or two?

And I stand by it as correct.

There are a very few speakers under ~90dB/W/m that are useful tools for serious music listening.

In fact, I can only think of two loudspeakers under ~90dB/W/m that are worth a damn: KEF Reference 201/2 and Revel Ultima Salon2.

There's quite a few "good" speakers that don't fit into your narrow, jaundiced view of good.

My "narrow, jaundiced view of good" boils down to four things, more or less:

-flat frequency response
-consistent and even midrange polar response
-minimal dynamic compression (i.e. sufficient cone area, reasonable efficiency, good thermal management)
-low diffraction cabinetry
-appropriate directivity for the room size/room reflectivity/listening distance
-well-damped mechanical resonances

It is true that anything failing to get all six of those things right does not interest me.

It is also true that there are a number of speakers that others like but that clearly fail on 1, 2, 3, 4, 5, or often enough all 6 of those aspects. So what? Should I lower my standards just to make people with lower standards feel better about themselves?

You're just peeved that Maggies single handedly make your pronouncement of dynamics wrong and made a liar out of you in ons swell foop, so you simply fall back to "well, I don't like them so they don't count".

Again, please learn to read. I accepted your precious crappy-sounding sunshades as a possible exception to my comments. (Whether they actually are or not I don't know, because they otherwise sound awful to me so I've never. And I frankly prefer quiet-awful to loud-awful.)

However, an etched mylar diaphragm exposed to lots of air on both sides is going to have very different thermal properties from voicecoils surrounded by metal motor parts. Which is what most speakers, and just about all good ones except for Quads, use. (Electrostats have arc'ing issues with too much power, so they can't generally get that loud, either. But a set of ESL-63's with Gradient's dipole sub-stands under them are really good for most music, if not Mahler 8 or Shostakovich 10.)

So, please, keep your biases to yourself and not pass judgment on things you're obviously incapable of impartially judging and obsessed with trying to make your point.

Um, where did I claim myself "impartially judging" anything? You just made that up.

I did not. I am in fact highly biased towards high-fidelity reproduction, and highly biased against colored reproduction. And I've never claimed otherwise.

Don't worry. I'll be here to call you a sphincter and knock you off that high horse you rode in on when you ask for it.

Oh wow, threats over the internet. How manly of you. Don't know why you're too immature to merely address the content in front of you. I mean, seriously, are you a Sasha Baron Cohen character or something?

 

[GranteedEV]

The grails of many folk are in the ~84db range.

Don't think of it in those terms. People don't "prefer" 84db/2.83v/m speakers. What they prefer, is deep bass extension without a subwoofer (damned if I know why though. Get multiple subs and be done with it. If I have to read one more "I had to ask if the subs were on" i'm gonna snap )

I'm struggling to understand how this level of sensitivity would work in most homes for both HT and various genre's of music.

Well.. Some of our posters actually listen at 70db. I think Pyrrho is one

Anyways I guess people don't really pay attention to high efficiency, or are even biased against it under the assumption that you will surely lose low level resolution.

Are these speakers designed to play classical music at low volumes as background noise?

Classical music has wide dynamic range. Contemporary music, where 80db sounds loud all the time, is probably more closer to what you're talking about.

But I'd say it like this:

Most people just don't care about dynamics... or much in general. They just listen for the system with the most "details" in the audition room.

...and some of us also consider one system specifically for balls-to-the-walls and one system of a different persuasion.

Will these speakers clip or distort with the given 2500 watts necessary for moderate listening?

They will undergo thermal modulation / power compression, yes. If a 2500 watt amp is clipping, something's definitely wrong. Personally, I can't get a 100 watt to clip audibly on most content I listen to at levels I listen at, although movies will do it. I think my speakers are about 88db/w/m or a shade under 90db/2.83v/m. I do still use a more powerful amp, although I'm not sure that's the way to do it. I believe more efficiency is the optimal route, though not always an option I suppose.

 

[AcuDefTechGuy]

Well, I never took Engineering Physics II since I changed major.

But when I auditioned Magnepan, I felt they were not my kind of speakers. They didn't sound as crystal clear to me.

I don't know if the low sensitivity of Magnepan speakers cause it. I just know I did not enjoy the sound as much as some folks. Could be just personal preference.

 

[somesnapper]

Thanks for taking the time to write this. So much to absorb!

You have asked a difficult question, that opens up a huge set of issues.

It goes to the heart of the matter, as to why speakers should be very large and the very high cost of making high performance speakers.

The sensitivity/efficiency/power handling equation is one that always calls for compromise to meet design goals.

On the one hand is the desire for speakers of practical size, even before you get to WAF.

So what makes a highly sensitive speakers.

1). First and foremost is the flux density in the gap. So you achieve this by having a powerful, magnet, with a voice coil entirely within the gap. (Under slung). Also you need to get the coil in very close proximity to the pole piece. This calls for expensive precision engineering. This also helps greatly in heat transfer to the pole piece and significantly reduces thermal compression and voice coil burn out. This also increases power handling greatly. The poles may or may not be vented and require copper shorting rings to focus the magnetic flux. This all adds up to a very expensive motor.

2). Next is moving mass, especially the weight of the cone being the major contributor. A heavy cone, lowers Fs and therefore how low the speaker can play. However a light cone will be more sensitive.

3). The stiffness of the suspension. This obviously is a factor. If you make a light cone, with high flux density and a very compliant suspension. The speaker will be sensitive and efficient, but have limited power handling, as the motor system will easily be driven beyond its mechanical limits..

Now lets look at how all these issues work against each other.

Now every speaker has a resonance, which is a combination of the electrical resonance (Qes) and mechanical resonance (Qms)

The relationship between the total Q Qts of a speaker and Qes and Qms is: - 1/Qts = 1/Qes + 1/Qms.

Now a moving coil driver does not exist in isolation. There is the relationship of the speakers and its loading. The total system Q, Qtlc, can never be lower then the Qts of the driver and it is always a bit higher.

Now Q really is a measure of damping. (acoustic over hang if you like).

The higher the number, the more uncontrolled and sloppy the bass.

It has been traditional to believe that system Qs of 0.7 are optimal. Many systems are higher than this. However as tastes become more refined, there is a marked shift to 0.5. I personally believe that total system Q needs to be 0.5 or even a little less. I am convinced now, that accurate bass reproduction is impossible, with total system Q greater than 0.5.

So how does this fit together.

As you increase gap flux density and sensitivity, you increase electrical damping and therefore lower Qes and therefore Qts.

If the suspension is loose, the driver will notp handle much power. So the mechanical damping has to increase, so we have a driver with low Qes and Qms and therefore low Qts.

At first sight this seems like just what we want. However, when you try to align the driver with a low Qts, you find a bass cut off well above f3.

So you have poor bass, unless you make the cone very heavy to lower Fs. Then you loose sensitivity.

The only way round this is a large horn enclosure. That way you can get deep bass and very high sensitivity and efficiency.

If you load a low Qts driver in a sealed enclosure, the F3 is very high and it takes far too much power to equalize it.

So you end up having a lot of trade offs. In the end you end up with a driver in with a Qts in the 0.3 to 0.35 and porting the enclosure. The result is you get an speaker with sensitivity in the +/- 90 db range, and resonant reproduction with Qtc too high and blame the room for the bass bloom, which is sometimes true, but by no means always.

If you make a sealed alignment, then you end up selecting a higher Qts driver with a soft (poorly damped) suspension. F3 will then be in the range where Eq is possible, as long as the motor design can handle the heat and there is sufficient linear travel (xmax) allowed for. However you can still end up with a design with a tight bass and Qtc in the 0.5 range, as long as you select a driver with a Qts not far north of 0.4. Very high Qts drivers, quite common in even expensive subs always sound lousy.

As far as a TL is concerned again the sweet spot is a driver Qts 0.3 to O.35, with a low Fs. Because of the nature of the loading sensitivities of 93 db I watt 1 meter even down to 20 Hz are achievable, and with a Qtc no higher than 0.5. So you can have deep bass without resonance or bloom with relatively high sensitivity down to 20 Hz. However the enclosure will be large, but not as large as a horn.

So in summary low efficiency and sensitivity, in addition to wasting energy and amp power is not a good thing. It generally implies poor electrical and or mechanical damping. It also can imply high order passive crossover networks, which can aid in producing a speaker with excellent frequency response, but never results in a first class overall result, mainly due to horrendous time and phase shifts, which really do preclude excellent results.

The exception to the above is very small sealed enclosures where driver parameters can be set to achieve an F3 in the 80 to 90 Hz range.

Sensitivity then ends up being in the 84 db 1 watt 1 meter range. In order to get high spl, extraordinary design and manufacture techniques have to be brought to bear on motor design. You end up with an expensive small package. The classic example is the little ATC SCM 7 at $1000 per pair and worth every penny, where WAF is an overriding issue.

I hope this has explained the close cluster in in the spread is speaker sensitivities and bass extensions.