The answer is something like a miniDSP SHD and two SVS SB2000 or SB3000s crossed over at 150 hz, with Maiky 76s eq applied, and a 24db per octave LW High/Low pass, and then using Dirac Live 3 over the top to clean things up.
Yeah ok, but the point that was made is that the LS50s roll off well before 100 Hz which does not appear to be the case.
Besides, how audible is THD at 80-150 Hz? How much of an issue is this, really?
Note that the resonant frequency of the woofer is ~90Hz (see impedance plot?).
The frequency plot clearly begins to roll off starting at ~120Hz.
Still, that is an anechoic measurement: in-room will extend somewhat deeper.
From that THD+N plot, distortion is already 10% (-20dB) at 100Hz at only 90dB input.
I can't speak to the audibility... But, maybe?
That is certainly showing strain for not a lot of input...
The point of audibility is a key aspect though - and IMHO it should always be.
I am inclined to think that sending spatially identifiable frequencies to a sub (or two) on the floor is more of an audible issue than dealing with a bit of distortion at 100 Hz, especially in smallish rooms, which is where I would think one would place the LS50s. I do however own the LS50 (meta), two SVS SB2000 and a MinidspSHD (currently crossing over at 80Hz), so I will give it a go.
Tried crossing ver the Metas at 120 Hz to the two SVS subs and I have no localization issues. Distortion is quite good too (ignore the peak at 30 Hz which is my suspended floor rattling a bit I think. The small peak at 160 Hz is Dirac trying to fill a dip due to distance of speakers from side walls). Thanks for the tip.