Subwoofer feels underwhelming

[ZolaIII]

Uffff R3 Metas are very strange speakers pushed very hard with their tuning. That results in high compression and distortion when pushed harder in other words they fall apart. Send the SVS Mini back, get two 10" sealed box sub's instead for the price or even 12" one's if you like. Put the Metas on top of the sub's (with isolation between) plug the port's on them and cross them at 120 Hz each with their own sub and make sure you kill the output under the high pass slope as much as possible with additional filters. You will need a rather potent four chanel DSP in order to make that work and you have to iron out peaks and smooth the crossover transition area (I use Butterwort and do the slope PEQ-ing also) to do it proper. That way you get a lot of small gains along the process (better clarity index in both mids and lows, more dynamics, better scaling with equal loudness, better separation, time domain...) which in the end all together do sound much better. If you don't want to do all of this just buy bigger and better main speakers or keep the one's you already have for up to moderate loud SPL levels. You won't get all if you don't do all and great majority of people won't do it.

 

[er|κzvio1in]

Uffff R3 Metas are very strange speakers pushed very hard with their tuning. That results in high compression and distortion when pushed harder in other words they fall apart. Send the SVS Mini back, get two 10" sealed box sub's instead for the price or even 12" one's if you like. Put the Metas on top of the sub's (with isolation between) plug the port's on them and cross them at 120 Hz each with their own sub and make sure you kill the output under the high pass slope as much as possible with additional filters. You will need a rather potent four chanel DSP in order to make that work and you have to iron out peaks and smooth the crossover transition area (I use Butterwort and do the slope PEQ-ing also) to do it proper. That way you get a lot of small gains along the process (better clarity index in both mids and lows, more dynamics, better scaling with equal loudness, better separation, time domain...) which in the end all together do sound much better. If you don't want to do all of this just buy bigger and better main speakers or keep the one's you already have for up to moderate loud SPL levels. You won't get all if you don't do all and great majority of people won't do it.

Interesting take. Some comments and questions:
1. At what SPL and distance do they start falling apart?
2. I assume the SVS mini has been bought because of aesthetic reasons, so bluntly commanding the thread starter to switch it for two bulky monsters isn't nice without knowing his buying decision
3. The meta's don't need to be on top of the subs, if you're going to use DSP anyway you can put the subs in the corners and place the meta's where they need to be. If your worried about bass localisation, then just set the crossover at a lower value because 120 hz seems very conservative. This has the potential to solve SBIR issues the meta's may have without crossover
4. What makes a DSP rather potent and why is this potency needed? A simple & old MiniDSP Nanodigi can handle this and will have 4 spare output channels
5. Too many (high q) filters on the crossover area will mess with phase and could be a pain in the ass to get it right: depending on placement, this can also be a frequency range with a lot of SBIR

 

[ZolaIII]

@er|κzvio1in
1. Near field of course:

https://dl.dropboxusercontent.com/s/dokttpxk9nvgfl6/Kef%20R3%20Meta%20%28Ported%29_Compression.png?dl=0

2. 10" sealed box won't really be much larger aesthetics I won't argue about.
3. See 2 and in purpose of saving space among other things.
4. Trough output, scalability, efficiency, adoption for specific math calculations formats, LUFS, software, documentation... but not in audio. In audio that it supports FIR and enough PEQ's with all filters (and quality implementation of them land that Q is also properly implemented so that you can do manipulation as supposed and of course that it's paird with good enough ADC & DAC ends and enough chenels of course and that software end is mature enough, easy to use and problem free.
5. You avoid using many, instead you use for example self filter and one or two PEQ's, do crossovers more shelow (Butterwort) and so on.

 

[er|κzvio1in]

@er|κzvio1in
1. Near field of course:

https://dl.dropboxusercontent.com/s/dokttpxk9nvgfl6/Kef%20R3%20Meta%20%28Ported%29_Compression.png?dl=0

2. 10" sealed box won't really be much larger aesthetics I won't argue about.
3. See 2 and in purpose of saving space among other things.
4. Trough output, scalability, efficiency, adoption for specific math calculations formats, LUFS, software, documentation... but not in audio. In audio that it supports FIR and enough PEQ's with all filters (and quality implementation of them land that Q is also properly implemented so that you can do manipulation as supposed and of course that it's paird with good enough ADC & DAC ends and enough chenels of course and that software end is mature enough, easy to use and problem free.
5. You avoid using many, instead you use for example self filter and one or two PEQ's, do crossovers more shelow (Butterwort) and so on.

1. 96 dB down to 60 Hz is where it crosses the 1 dB deviation, not too shabby at all

 

[ZolaIII]

1. 96 dB down to 60 Hz is where it crosses the 1 dB deviation, not too shabby at all

It needs high pass at 48 Hz when going solo as that's where Fb is which you can see in impedance/phase measurements.

https://dl.dropboxusercontent.com/s/76xyvywqujq7lpz/Kef%20R3%20Meta%20%28Ported%29%20Impedance.png?dl=0

That would lower compression/distortion to a extent (on the under it output but would reflect on over it).
Graph also shows in conjunction to plugged port measurements why to cross it at 120 Hz where Fs is (actually that's typical for a woofer of that size) to stress it out which would improve things further for eight distortion or SPL output. Of course then you need one sub per each speaker and they do sum on their own (which is multiple beneficial in the end).

 

[wardzin]

The graphs look like you are doing nothing wrong. So if you are happy with the sound without the sub, I would advise returning it and saving the money.

Yeah, I think I'm probably gonna do that. The subwoofer is an upgrade, but nowhere near worth the $1500 price tag (to my ears, at least).

What crossover do you have currently? If you crossover at 100hz, a graph like the one in red below seems realistic by employing a bit of EQ.

Crossover is 80 Hz. I tried a bunch of other crossovers, but they all messed up the in-room response, so 80 Hz really is ideal.

This sounds like you are not high-passing the R3s? You should definitely do that at around 80 Hz. Pull down the peak at 90 Hz as well. There also seem to be differences above 100 Hz. Why is that? Is your subwoofer low pass too high?

The speakers are being high-passed with the WiiM Amp, I confirmed this with measurements:

Though, now that you mention it, is it interesting that the sub roll-off extends past 200 Hz. And that the speakers are also not fully being muted below 80 Hz, just brought down to 60 dB. This is my first experience dealing with subs and crossovers though, so maybe this is normal?

I'd be more worried about the dip 100-200hz.

I did already set-up an EQ for the dip, but these are the "raw" measurements without EQ just to take that out of the equation. The EQ does help a bit with music overall, but not necessarily with helping me feel more impressed with the sub performance.

With regards to speaker placement - they're in a rather small room (3m x 3m, probably why the bass extends so deep). The only way to tame the 100-200 Hz hole without EQ boosting is to bring the speakers out about 1 metre out, which isn't really practical. I originally had them about 20 cm away from the back wall, but this amazing guide from Genelec suggested that if I can't bring the speakers out, it may actually be better to place them right against the wall, with only about 5 cm for the back port to breathe. This did slightly help, but it wasn't a huge difference.

 

[wardzin]

EDIT: Also, see what it sounds like with the KEF ports plugged when you are using the subwoofer. On my Elac speakers this helped tighten up the bass in the mid-bass region, but some speakers may not work so well with the ports plugged. With the ports plugged I use a 100 Hz crossover frequency.

I did actually try this before I wrote this post! These are the results. Seemed to inexplicably boost the 40-60 Hz region, and well as slightly tame the 225 Hz peak. Not sure what to make of it, or which graph is even "better".

 

[terryforsythe]

I did actually try this before I wrote this post! These are the results. Seemed to inexplicably boost the 40-60 Hz region, and well as slightly tame the 225 Hz peak. Not sure what to make of it, or which graph is even "better".

View attachment 357780

The 225 Hz difference could mean you have some energy at that frequency coming out of the port. This is not uncommon. Try putting the microphone close to the port and measure the port response if you are curious.

Which is better is up to your ears. Listen both ways to find what you like better.

 

[levimax]

I added 2 SVS 3000 to my system and carefully integrated them in with my 2 other close to full speakers. Unless I am playing loud I don't really notice the subs much which is as expected due to the Fletcher- Munsen curve. When playing loud especially with music that has some LF content the subs make a big difference both LF and mid range, for normal listening levels not so much. The measurements you get look OK, you may need to adjust your expectations as to what subs actually bring to the table.

 

[Daverz]

I added 2 SVS 3000 to my system and carefully integrated them in with my 2 other close to full speakers. Unless I am playing loud I don't really notice the subs much which is as expected due to the Fletcher- Munsen curve. When playing loud especially with music that has some LF content the subs make a big difference both LF and mid range, for normal listening levels not so much. The measurements you get look OK, you may need to adjust your expectations as to what subs actually bring to the table.

I'm using a single SVS SB-1000 Pro, crossed over at 120 Hz with steep second-order Neville-Thiele filters (generated by Acourate) and with the mains suitably delayed (6 ms for the sub DSP delay + extra time of flight from the sub). The mains are Buchardt S400s.

At moderate levels the difference when playing with the sub volume control is very subtle with my typical diet of symphonic music. There's an extra solidity, perhaps. However, crossing the sub/mains this way also smooths out the response between 100-200 Hz, even before applying any room correction (I do use the SVS PEQ filters to knock down some modes below 100 Hz).

The biggest gain, though, is in dynamics. With very dynamic music the climaxes can be startling even at moderate levels. The KEFs are larger than the Buchardts, so the effect may not be as pronounced.

 

[wardzin]

The 225 Hz difference could mean you have some energy at that frequency coming out of the port. This is not uncommon. Try close putting the microphone close to the port and measure the port response if you are curious.

Which is better is up to your ears. Listen both ways to find what you like better.

There is no way I'm even close to an experienced enough listener to be able to tell apart such a marginal difference in the frequency response, so I've generally been guiding my placement and EQ decisions by measuring and matching the Harman curve as closely as I can.

When playing loud especially with music that has some LF content the subs make a big difference both LF and mid range, for normal listening levels not so much. The measurements you get look OK, you may need to adjust your expectations as to what subs actually bring to the table.

That's what I noticed as well, I can only really tell the subs are there once I crank the volume. Which, I rarely do because I live in an apartment. To be fair, the soundproofing is excellent, and I haven't gotten any complaints, but pushing the volume still stresses me out. So, I guess my situation isn't even a great use-case for a subwoofer.

 

[boxerfan88]

Try this experiment (if your system allows).
Play your favourite music that has reasonable bass content at average levels (eg. 75 SPL).

Part way through the track, cut off both the mains, leaving just the sub playing. Now you can hear very clearly what the subwoofer is contributing to the overall performance. To me the subs provide quite a bit of “body” to the music. It’s quite subtle when all speakers are playing.

I did the above experiment a few times, and it helped “convince” or “brainwash” me that having a subwoofer for stereo is really good. My current thinking is that I don’t think I’ll go back to a stereo system without subs.

 

[HiMu]

Seemed to inexplicably boost the 40-60 Hz region

Plugging the ports reduced the bass from speakers that might've caused interference with the subwoofers frequencies? Something is not right with the high pass/ crossover settings based on your post #26, looks like it acts like shelf filter instead. If you compare the measurements from Erin's review in the image below https://www.erinsaudiocorner.com/electronics/wiim_amp/

Punch is higher than subs range.

But also a component of rhythm/timing.

 

[AaronJ]

Return the sub and split your budget on two subs. Two “lesser” subs will outperform one more expensive sub when properly integrated, if SPL isn’t an issue. I use a pair of B&W ASW608 and they sound much better than a single B&W ASW10CM which was 25% more expensive than the other two combined.

 

[er|κzvio1in]

I'm using a single SVS SB-1000 Pro, crossed over at 120 Hz with steep second-order Neville-Thiele filters (generated by Acourate) and with the mains suitably delayed (6 ms for the sub DSP delay + extra time of flight from the sub). The mains are Buchardt S400s.

At moderate levels the difference when playing with the sub volume control is very subtle with my typical diet of symphonic music. There's an extra solidity, perhaps. However, crossing the sub/mains this way also smooths out the response between 100-200 Hz, even before applying any room correction (I do use the SVS PEQ filters to knock down some modes below 100 Hz).

The biggest gain, though, is in dynamics. With very dynamic music the climaxes can be startling even at moderate levels. The KEFs are larger than the Buchardts, so the effect may not be as pronounced.

120 hz steep crossover @ 120 hz is crazy high, I looked up some measurements of the S400 and noticed high distortion up to that frequency so I guess that's the reason for it

 

[ExPerfectionist]

Though, now that you mention it, is it interesting that the sub roll-off extends past 200 Hz. And that the speakers are also not fully being muted below 80 Hz, just brought down to 60 dB. This is my first experience dealing with subs and crossovers though, so maybe this is normal?

Yes that's how crossovers work. There are different slopes, like 12dB or 24dB per octave. The reduced volume between the two sides, on a slope, is supposed to combine and blend to be the "correct" volume.

Per octave means that a speaker crossed to a sub at say 100Hz, the speakers' level is reduced by the slope (12/24dB) by 50Hz, and continues. The sub's signal is reduced by the slope amount by the time it gets to 200Hz, and so on. Same/similar for the crossovers within speakers to blend tweeters, mids/bass drivers.

 

[FrantzM]

There is no way I'm even close to an experienced enough listener to be able to tell apart such a marginal difference in the frequency response, so I've generally been guiding my placement and EQ decisions by measuring and matching the Harman curve as closely as I can.


That's what I noticed as well, I can only really tell the subs are there once I crank the volume. Which, I rarely do because I live in an apartment. To be fair, the soundproofing is excellent, and I haven't gotten any complaints, but pushing the volume still stresses me out. So, I guess my situation isn't even a great use-case for a subwoofer.

Hi

I was about to reply to your original post.. I've had different issues to deal with , these past month ...

I have taken the liberty to highlight this part. I believe this is an issue much more pervasive than many realize. Many system sound .. underwhelming at low volume. Some take it as a badge of golden-eared audiophilia"... Telling you how good the midrange was, and that bass is for "bassheads" , HT or for hip-hop..
The reality is that our hearing follows roughly a curve that Harvey Fletcher and Wilden A. Munson set out to answer in the early 1930s, and in 1933 they published their results in a paper entitled, “Loudness, its definition, measurement and calculation.”
here is the curve of equal loudness:

TLDR from all of these is: Let's suppose a measured SPL of 60 dB at 1000 Hz. For most humans to perceive the same sensation of loudness at 40 Hz requires an actual SPL at 40 Hz of measured SPL of about 87 dB...27 dB higher in actual SPL... That means in order for many to perceive a balanced sound reproduction at low average and general SPL, the bass (and somewhat the treble ) must be raised some.. How much is an issue that for odd reasons, is only addressed or implemented by only a few companies and this with various degree of success... I know that Denon through Audyssey has that and they call it DynEQ, to me a game changer: It varies the amount of boost of lows and highs according to volume position/level. I also believe RME has it and so does the MonoPrice HTP-1 PrePro.. Not sure DIRAC has it.. Variable loudness according to volume, is to me essential. We don't listen at a fixed level. It varies according to the situation thus the balance of sound... I tend to think that is what is happening based on the highlighted observation from you.

if you have access to a Denon AVR, try to use DynEQ , after calibration with REW, I am almost certain you will hear a better, more balanced reproduction at low volumes...

Peace.

 

[Daverz]

120 hz steep crossover @ 120 hz is crazy high, I looked up some measurements of the S400 and noticed high distortion up to that frequency so I guess that's the reason for it

Yes, but it was based on study that showed a subwoofer could not be localized up to 120 Hz with a 24 dB/octave crossover. I can't find the URL for that right now. I know the conventional wisdom is 80 Hz (with what crossover slope, though?) I use a very steep crossover filter, steeper even than 48 dB/octave. I do not hear the sub as a separate source of sound.

The reason for going so high is not really distortion, but to try to get a smoother frequency response in the region over 100 Hz with the speakers well out into the room and the subwoofer against the front wall. I do get better dynamics with this arrangement, though.

 

[er|κzvio1in]

Yes, but it was based on study that showed a subwoofer could not be localized up to 120 Hz with a 24 dB/octave crossover. I can't find the URL for that right now. I know the conventional wisdom is 80 Hz (with what crossover slope, though?) I use a very steep crossover filter, steeper even than 48 dB/octave. I do not hear the sub as a separate source of sound.

The reason for going so high is not really distortion, but to try to get a smoother frequency response in the region over 100 Hz with the speakers well out into the room and the subwoofer against the front wall. I do get better dynamics with this arrangement, though.

Yes I understand: by pulling the speakers forward (and away from side walls) and crossing them over high, while putting subs in the corner/against the wall, you should have eliminated all SBIR which is great!

 

[Sokel]

Yes, but it was based on study that showed a subwoofer could not be localized up to 120 Hz with a 24 dB/octave crossover. I can't find the URL for that right now. I know the conventional wisdom is 80 Hz (with what crossover slope, though?) I use a very steep crossover filter, steeper even than 48 dB/octave. I do not hear the sub as a separate source of sound.

The reason for going so high is not really distortion, but to try to get a smoother frequency response in the region over 100 Hz with the speakers well out into the room and the subwoofer against the front wall. I do get better dynamics with this arrangement, though.

Nope,120hz is stratosphere for subs at a distance from mains.
Have a look at the studies here:

Bass and subwoofers

This is the second in a series of follow-up posts to https://www.audiosciencereview.com/...-of-lokki-bech-toole-et-al.27540/#post-950580, this time focussing on bass perception with a brief digression on subwoofers, since there's relevance to (reproduction of) concert hall acoustics. I’ve...

www.audiosciencereview.com