Speaker upgrades or keep LS50 Metas with Subs

[Heinrich]

Everything is a compromise.

Listening at 96dB for long, repeated periods certainly is going to impact someone's "lifestyle" since it'll result in hearing loss.

Sure, but take home message (tl;dr): standard "subs" won't do substancial good to achievable sound pressure levels if "home theater" is the aim AND a minimum of quality is of consideration. But, what is which and how is debatable.

At least I corrected my suggestive error.

I had the LS50 and had to hand it back in a minute after adding "a sub". Puh, that remaining weakness I didn't expect. And I'm decidedly not into headbanging.

Reiterated, if you don't mind, the Ls50 taken as is for the purpose it was designed for is one of the better choices.

 

[terryforsythe]

To get there, said 10dB attenuation is to be achieved. It would need a crossover second order at 200Hz.

The OP has a MiniDSP 2x4HD. With that he can use a higher order crossover slope with a lower crossover frequency to accomplish at least the same amount of attentuation down below 100 Hz. This may work better since his subwoofers are asymetrically placed.

For context, I have spent a LOT of time experimenting with different crossover frequencies and slopes for my standmount speakers and subwoofer, both with and without Dirac Live, measuring frequency response and harmonic distortion, and listening. Every room is different, and my system is different, but what worked best in my system to achieve both good frequency response in the crossover region and good harmonic distortion was using LR eighth order slopes and the speakers' high pass filters around 100Hz.

When using Dirac Live, a suggestion I recently read from miniDSP was to specify the subwoofer's crossover frequency to be a bit higher than the speakers', and let Dirac Live work its magic to tune it (supposedly it uses both FIR and IIR filters, so I assume it implements all-pass FIR filters where appropriate to address phasing issues that otherwise would cause). That seemed to work well for me.

 

[NTK]

The JBL 4722 is not the Revel F226Be.

I concur with Heinrich's assessment of the LS50's distortion at higher SPL. Also, take a look at the below response linearity vs SPL Erin measured. Much depends on how loud the OP is listening and the high pass crossover frequency and topology being used on the LS50s.

It doesn't take a genius to infer that speakers that will require a forklift to unload from the delivery truck and a team of movers to get into your room may be better at producing high SPL than the LS50. The question is how much SPL you need/want in your room. Otherwise the bigger speaker is just better at giving you permanent hearing damage.

This is the (original) LS50 Meta distortion graph from Heinrich's post, and below are the distortion criteria for peak SPL from ANSI/CTA-2034-B. For band 2 (63 - 200 Hz), the limit for the perceptually benign H2 is -24 dB = 6.3%; and -30 dB = 3.2% for H3. Therefore, according to the technical committee that wrote CTA-2034, the LS50 is perfectly usable to 96 dBSPL when crossed to subs at ~120 Hz. Remember that at the crossover frequency, for correlated signals, the LS50 and the sub(s) are each outputting 6 dB less than the total.

Therefore, the question is how much and how often you need to exceed that level (which obviously also depends on the room size as the numbers are for 1 m free space). The answer may be less often than you think.

 

[terryforsythe]

The question is how much SPL you need/want in your room.

Like I said:

Much depends on how loud the OP is listening and the high pass crossover frequency and topology being used on the LS50s.

Also, it is not a bad idea to have headroom for dynamics, especially if listening to uncompressed music. I ran a test and was getting peaks (LZpeak) up to 19dB over my listening volume (LAFmax): https://audiosciencereview.com/foru...ping-b100-amplifier-review.57036/post-2087328

Here are Sokel's results from the same type of test: https://audiosciencereview.com/foru...ping-b100-amplifier-review.57036/post-2087242

 

[terryforsythe]

Funny, the OP has not posted in this thread since it was started, and probably not even following it.

 

[Heinrich]

This is the (original) LS50 Meta distortion graph from Heinrich's post, ...

No, it is not. I decidedly posted my more elaborate post later in order to show that *not* the percentage graphs are relevant, but the absolute distorion figures in dB, not percentage. My bad, the first post.

One could fiddle with frequency and slopes, but there are much more powerful speakers available in roughly the same class of size. Even cheaper ones.
Reiterated relentlessly, the LS50 is a nice package alltogether, but not fpr "home cinema", whatever that means. (Think of intermodulation.) A sub wont help. As said, I virtually only know people that would be fine and happy with the LS50 until the end of time. Most, 99+% do not need that spl that I'm addressing, sure. But, these nice guys won't think of an upgrade either. And the upgrade was the original topic.

Not the least, I've got in person experience with the LS50 that fuels my caveat.

 

[Sokel]

The particular speakers can greatly benefit from subs, specially 2 stereo ones but under conditions:

That the x-over point is high (more than 200Hz) and that inevitably the subs are tight to them or directly under them at such high x-over point (in fact at every x-over point more than the "normal" 80Hz as it's the area we start going well int voices) .
Even better from bass bins at the Genelec W371A fashion, it's exactly what it's made for.

And no, it's not only about SPL or dynamic range (for sure it's a factor though) looking at it from a sweep point.
It's that at dynamic music it's the combined energy that matters and it happens that the main bulk of this energy is right at the weak spot of combined satellites with distant subs.

Yes, a combination like that nicely made (x-over at 80-ish Hz) can play loud, can maybe rumble but if mains are not up to the task cannot kick. That's where the bigger boys take over. Now, combine all that with the benefits of AE described by @Thomas Lund and you have your answer.