I am utterly befuddled by all this...
I understand that Josh Ricci is using a beast of an amp to test the absolute limits of each sub, and that they are controlled experiments that therefore don't result in the same rapid transients and combinations of wavelengths that a driver playing music would have to deal with, so the max outputs achieved and measured by Josh are effectively unachievable in 'real life' uses.
But... going from the post above and the previous posts on this, are we effectively saying that any 'normal' amp in 'normal' use is unable to control voltages enough to ensure that THD distortion levels are kept in check at certain frequencies in the driver's range, specifically focused at and around the peak impedance of the driver where only a minimal voltage / power is required for the driver output to reach the same level as those levels across the wider bandwidth that require a lot more power to achieve due to lower impedances?
Are we saying that, in this example, a 95v input across the bandwidth range is going to wreck the driver in the bottom of the range due to excessive THD levels, so we must therefore apply a limiter across its entire bandwidth so the driver does not 'see' more than the voltages required at the peak impedance point (53v), effectively limiting output?
Is such a limiter possible? And how are the various systems that are giving drivers around double their rated power (popalock's and carp's, for example) actually surviving this onslaught if the THD distortion rises so massively and they need so little power at the peak impedance point? Are the amps reducing the power at this point? Can they do this when they are playing multiple wavelengths that include these peak impedance points as well as other points further away from this? Or does the amp just react to the driver rather than the amp dictate what the driver is given?
I'm sure all the above are dumb questions but I need to get them clear in my head - having bought a CV5000 to run four SI18s @ 2Ohm (1250wRMS each at max power) I'm wondering if I've gone completely overkill, even though the WinISD modelling showing it's safe excursion-wise even under full power...
We have Ilkka's and Josh's tests of the LMS Ultra, both in pretty much the same volume enclosure , so, with pretty identical naked responses.
The amplifier is the primary difference. Yeah, I'm aware that they were different production examples tested in different parts of the world and years apart, so spare me that irrelevance and look at the differences in the results:
If run naked, the top sine sweep is the maximum the sub can take, as opposed to the posted noise burst results, which allow a much higher power input up top than at the bottom. Without limiting the power input at the bottom in real life use, there's no way you could increase the input power level to the top end results without blowing the driver out of the box.
Take the 115dB sweep from Ilkka's test, IF you have a CE4K fed by a 220V/16A line as your ceiling. Subtract a dB if the same amp is on a 120V-20A home run. Add 4-5dB if you own the $7,000.00 K10 and feed it with 22V-50A.
Forget the 123dB top end max burst numbers because without being able to limit the bottom end the driver will be destroyed at that level.
Then, look at what happens when you apply a +10dB L/T Boost (as I normally do). The top end is further restricted by that same 10dB that's being used to boost the low end to change the FR of the sub.
This is why the multiple sub builds are not nearly as hair-brained or crazy as people make them to be. If your boost sub has a max top end of 105dB, then you need 8 of them to see 120+dB peaks at the seats without losing your frequency response to compression.
Sounds like commercial subs have the advantage because they employ those limiters, but the price is a radically and infinitely changing frequency response with dynamic source, which is what the device is built for.
So, you painstakingly work your in-room FR to your liking at 'X' dB sweep and when you play a movie you have no idea what FR you're actually listening to. It may not seem like much on a graph, but a FR with F3 of 30 Hz vs an F3 of 14 Hz is pretty huge and you never know what the response is while playing the movie because it purely depends on the content and playback level.