Wow that's a higher coefficient than I thought!
I don't think it's as big of a deal as you think though.
I've gone through it before, I'll go through some figures with my ATC SCM20 Pro PSL Mk2 speakers:
View attachment 291900
View attachment 291901
(although the woofer has a similar appearance to the SCM20s reviewed here. Since then there have been so many changes, that the woofer's similar appearance is the only thing similar that the 25 year old (likely somewhat broken) example has in common with my most recent ATC SCM20 Pro PSL Mk2
They call it a 150mm (6") driver, but the diameter of the cone + 1/3rd of the surround is 135mm, or 5.31"
It's a beast of a woofer - it's underhung, has an 8mm voice coil in a 20mm gap (travels very close to distortion-free when within its 12mm peak to peak excursion ability, and the magnet, is massive.
Sure it's small, but this could very well be a 10 or 12 inch high fidelity woofer as well - it's equipped, except for the frame (the frame would be bigger and would dissipate heat even better.
OK. So when I bought these speakers, I lived in an apartment. Unfortunately the woofers didn't care that people were around - I still needed to loosen them up. Fortunately I had an idea: play infrasonic tones, and increase the time you play them.
I did.
I put 23Hz through them during the day 40 watts of it. I didn't measure the woofer's impedance at that frequency, so I used 6.8 or something. 40 watts is an estimate, but it's probably within 5 watts or so. Anyway, these woofers, cast basket/frame, are quite heavy. They weigh about 20 pounds.
It took 40 watts and many many (6?) hours for the frame of this woofer (which includes that 5/8" thick part where the screws go through at the front) for the temperature to increase and then taper to 53 degrees Celsius
That's 40 watts RMS.
Music with:
RMS voltage: 0.987
Peak to Peak voltage: 8.12
Into 6.8 ohms:
RMS: 0.143W
Peak to Peak: 1.21W
1.21/0.143 * 40 = 338
It would take music with peaks (kicks) at 340 watts to do this same amount of heating.
The first figures I used after "Music with:" came from a screenshot of my oscilloscope. I was using my Topping G5 to drive the speakers above directly with ZZ Top - I Got the Six. I have them sitting about 5 feet apart, and I was about 4.5 feet away from them. They were playing more than loud enough to enjoy the music. With the door shut and someone in the next room watching TV, they could identify the song over the TV.
Now, let's increase this by 4 times so that RMS is ~0.5W and peak to peak is ~4W
That means we'll get 1/80th of the heat (not exactly, but good enough for my purpose here)
1/80 of 35 degrees is 0.5 degrees (35 degrees comes from the coil being about 5 degrees hotter than the chassis: very likely because of physics, and room temp being 23 degrees: very likely because I keep things at 23)
0.4% of 0.5 is 0.2%
You can only really buy 1% components, so even 2.5WRMS and 20W peak to peak would still be undetectable.
The SCM20 Pro PSL Mk2 speakers use 1% caps and hand wound coils which are
exact.
Also, the speaker discussed here is a sealed design, so all of the heat was kept inside the enclosure - a port would've let a lot of air in and out for cooling. And a 10 or 12 inch woofer (more typical candidates for 3" voice coils) would have a lot more surface area for cooling.
My take on it is this: coil heating doesn't have a meaningful effect until the speakers are underpowered or you're playing them too loud.
When you play them too loud, parts do go out of spec, and so do your ears. Don't listen too loud!
lol