We should reject hard-to-drive speakers more often

Good topic, Erik.

A difficult speaker load is created foremost by its passive crossover, more so really than (a purer load) low impedance and low sensitivity as a function of the drivers sans passive XO. Combining higher sensitivity, higher driver impedance (i.e.: 8 ohms minimum, as pointed out by @atmasphere ) with an active approach would be ideal in making the most of both amps and cables, and in effect the sound coming from the speakers with the better, direct amp-driver coupling.

What I don’t get is seeing these complex passive filters in many a high-end segment speaker that really only nurtures the forced existence of crazy expensive amps, and which comes down to the need for them to be more or less impervious to load in the face of steep phase angles and the occasional ultra-low impedance dips in the lower frequencies. All the more reason to go active and start throwing those mega amps under the bus, because actively they wouldn’t be needed any longer.

If indeed there’s merit to the claim of people consciously looking for speakers that are known to be a difficult passive load, for reasons apparently that it’s somehow deemed a desirable trait in a speaker, it would seem a fool’s errand; to my ears difficult-to-drive speakers usually can’t shake off an inherent sensation of them actually sounding like a difficult load, irrespective mostly of any beast of an amp thrown at them, so why bother with such speakers in the first place?

@jon_5912 wrote:

High power class D amps more than make up for the difference in dynamic contrast between low and high efficiency speakers. 

And how do they make up for that? There's only so much heat that can be dissipated in a given voice coil, not least a smaller one through typical low sensitivity. Power is power, and the less efficient receiver, unless extremely capable in power handling (which could have other, potentially detrimental effects), always ends up storing more heat, with all that entails. Thermal compression as in actually overheating the VC and causing heavy compression or sending the VC up in smoke is hardly the only, if even the main consequence following here, but rather what happens way earlier as something that has actual, audible effect. The degree to which this is pronounced, and at the (early) juncture this occurs and starts becoming a problem (referencing not least to a higher eff. scenario in which it isn't) would seem to be the more important aspect to investigate here. 

High efficiency speakers have way worse problems than thermal compression. 

I don't see how they do when properly implemented. 

@atmasphere wrote:

The signal itself is the source of dynamic contrast. Loudspeakers only take away from that; if you value dynamic contrast, using a speaker that has the least thermal compression will bring you closer to your goal. In this light, ESLs have the least thermal compression owing to no voice coil at all; a close runner up is higher efficiency loudspeakers [...]

Which only gets you so far being ESL's have the lesser macro-dynamic range compared to horns/high efficiency designs, unless extremely (unrealistically?) large and rid of bass signals. Before thermal compression would ever become an issue with horns, other potential factors like horn material resonance and truncated horn size in the midbass horn in particular would be the primary concerns. While ESL's don't compress per se - and as such some of their implementations can have an advantage over low eff. direct radiating, dynamic driver designs - it's well known that when they limit out they do so abruptly, not least when applied full-range. 

"Some compression occurs in planar drivers, including ESLs, because the membranes are stretched across a frame. The mechanical impedance is not linear."

True for all drivers, except perhaps the massive fan subwoofers. I think the difference in measurement/thinking about thermal compression vs. mechanical is that thermal compression changes the behavior of the speaker in time, sometimes within milliseconds, while mechanical compression is always there, until you blow the driver. :)

As a mechanical compression artefact this is hardly true for all drivers. Dynamic drivers have suspensions, and the "stretch" of a membrane here wouldn't occur in a way comparable unless the suspension iself is (getting close to being) mechanically edged out. 

I do think it’s odd audiophiles have fixated on thermal compression, specifically, as being the only one that matters, though I do agree that higher efficiency drivers seem to be at an advantage here.

Question is when, and perhaps not least how thermal compression starts becoming prevalent and an actual audible effect. Thermal 'modulation' may be a better term to explain or correlate what happens sonically; thermal compression impacts SPL envelope and ultimately driver failure, but it also appears to dull transient cleanliness and snap at a much earlier juncture as a very dynamic phenomena. 

There’s a reason JBL professional drivers are so expensive, and one of the main reasons that is that they are built specifically to avoid thermal compression even at constant power levels that would make most audiophile systems weep.

Other pro brands would do equally well, and at a cheaper price, but yes pro drivers are simply on another level here. 

@erik_squires wrote:

Then I’d really like to know what you think happens when your reach the maximum excursion of a driver. Either they have limited excursion, and therefore compression, or they have infinite excursion and no compression.

That’s indeed what I’m talking about in regards to mechanical compression, but how often do we reach, let alone exceed X-mech? My point is, before that happens the driver’s suspension has a range of motion within which compression per se as a mechanically induced phenomenon isn’t relevant. What’s arguably more relevant mechanically is hysteresis (or magnetically, hysteresis distortion) as that which happens through the range of cone movement, but that’s hardly a compression issue, no?

Although: smaller (<10"), low eff. woofers certainly makes their effort in reproducing the range <40Hz at elevated, but hardly prodigious levels be known, and when you’re used to a pair of corner loaded, high eff. tapped horns fitted with 15" pro woofers in 20cf. enclosures where the cones barely more than a few mm’s at bonkers SPL’s, it’s all the more obvious. With smaller hifi woofer cones working hard it’s likely a combo of mechanical noise and -compression, as well as the onset of thermally induced compression.

I wish I could find them but I remember seeing tone burst tests showing that thermal compression could happen in a tweeter in less than half a second. You could see the first tone burst perform perfectly, and then half way through the second compression sets in.

Interesting; I take it what could also be referred to as thermal modulation as a more dynamic phenomenon. It’s a shame this area isn’t more well documented, leaving us with the more general ’thermal compression’ description and correlation as heat build-up in the voice coils over longer time.

@erik_squires wrote:

At some point significantly below Xmax I believe there begins to appear evidence of compression, both within the FR and distortion.

In addition to thermal issues this would only stress the importance of headroom, and also as a phenomenon somewhat equivalent to thermal modulation as an effect that has sonic implications well before the typically assessed limits, both thermally and mechanically. These "grey areas" definitely need more attention. 

@ditusa wrote:

In my opinion, a high powered amplifier that can drive lower impedances will never be a proxy for a speakers lack of true efficiency.

I fully agree, Mike.

@mrdecibel wrote:

The times I used tubes for mids / highs, and ss for bass, I always was aware of the " transition " between the two.

Even via my horn hybrid, actively configured main speakers using different SS amps for the top horn section (~600Hz on up) and bass bin it doesn’t go unnoticed that they "speak" slightly differently when compared to using two identical amps here, not least when the identical amps are bi-amped vertically. The difference is likely somewhat more subtle vs. the scenario of tubes + SS you’re describing, but it goes to show there are points in coherency to gain with extra attention invested here.

The SS amp chosen for the subs below ~85Hz is less critical wrt. ultimate coherency, but to my ears what is critical in this region is choosing an amp that doesn’t go totally bonkers with a damping factor in the thousands as a means in itself, but this is obviously also a matter of proper subs-amp-room matching and what sounds the most musically full and natural here, or whatever one finds appealing. Overly damped (i.e.: "tight") bass isn’t natural bass to my ears, nor is the inverse scenario - it’s certainly about finding the right balance in relation to one’s (p)reference and context.

@ditusa --

Very good article by Floyd E. Toole - thank you :)