Revel 328be and a sub/distortion

[Will2campb]

The measurements from the review of the 328be indicate that its distortion, including bass distortion, is very low. A subwoofer’s distortion (for example, the JL audio e-sub 110) is significantly higher than the bass distortion of the 328be. First, how audible is low frequency distortion? Second, would there be any benefit to adding two subwoofers to the 328be? Or would it actually be detrimental because of the increased distortion?

 

[pjug]

I am also wondering how does the 328be bass distortion compare to servo Rythmiks? In the @amirm Rythmik L12 review it looks like distortion is similar, about 30dB down at the bottom end, but it isn't clear to me what SPL level was used for that measurement (was it ~100dB SPL?)

 

[pierre]

Genelec subwoofer claims:
7370a: 3% 2nd harmonic distorsion between 30hz and 85hz
7380a: 1% same as above

they do not tell at which SPL.

below 80hz, I do not hear most distorsion below 10%.

i think that adding sub would help for movie and for electronic music. My system is flat to 30hz without a sub but on some music the sub makes a difference, plus also flattening the bass response.

 

[richard12511]

The measurements from the review of the 328be indicate that its distortion, including bass distortion, is very low. A subwoofer’s distortion (for example, the JL audio e-sub 110) is significantly higher than the bass distortion of the 328be. First, how audible is low frequency distortion? Second, would there be any benefit to adding two subwoofers to the 328be? Or would it actually be detrimental because of the increased distortion?

I guess it depends on the sub. 328 shows a little less than 3.0% at 30Hz and at 96dB. By comparison, the JTR Captivator 4000($3,700) shows ~1.50% distortion at 30Hz all the way up to 115dB. So, while the 328 does have impressive bass distortion for a tower speaker, the best subwoofers still have it beat.

Another example of a sub with lower 30Hz distortion would be the Rythmik F25HP($2,600). It shows ~2.0% distortion by the same metric, and it's not until 110dB that it jumps above 3%. The Rythmik FV18($1,800) has the same distortion(~3.0%) at 110dB as the Revel does at 96dB. The FV18 is actually under 1.0% at 95dB . The SVS PB13($1,800), at 2.0% also has the Revel beat.

Not trying to downplay the distortion performance of the 328Be, which is truly exceptional(makes me wonder how great the Salon2 is), but just pointing out that there are subwoofers out there that can match or exceed it at reasonable price points for someone purchasing a $16,000 speaker. A tower speaker competing with legitimate subwoofers for bass distortion performance is super impressive, though. There's no doubt about that.

Looking at the databass page for the JL E110($1,700), I see ~6.4% at 30Hz and at 95dB. That is indeed a bit worse than the 328, but I think it's important to point out that the E110 is a relatively small subwoofer(much smaller than the F328Be). The E112($2,300) is a bit larger, and shows comparable distortion to the 328, at around 3.5%.

All that said, I'm not sure how audible all of this even is. I've heard some say that anything below 10% is inaudible.

 

[amirm]

It would be quite hard to compare measurements of distortion performed using different fixtures/setups. So unless we test some subs ourselves, we won't be able to answer that.

 

[pozz]

I've heard some say that anything below 10% is inaudible.

It depends on the spectrum of the distortion. There's this study which was quoted by Audioholics: https://www.audioholics.com/loudspe...tortion-at-bass/total-harmonic-distortion-thd

For H2 to H5 the thresholds are different, and by the time you get to 100Hz they sink (or rise, depending on whether you're thinking in dB or %) rapidly. IMD is also unforgiving in that region, when measured, as is Doppler.

 

[Kal Rubinson]

That is indeed a bit worse than the 328, but I think it's important to point out that the E110 is a relatively small subwoofer(much smaller than the F328Be). The E112($2,300) is a bit larger, and shows comparable distortion to the 328, at around 3.5%.

And multiples of them can be placed in effective locations.

 

[onion]

Second, would there be any benefit to adding two subwoofers to the 328be? Or would it actually be detrimental because of the increased distortion?

I thought the main purpose of multiple subs is to even out the bass and mitigate room modes/ standing waves and nulls etc. This is much more audible than low frequency distortion IMO.

 

[richard12511]

It would be quite hard to compare measurements of distortion performed using different fixtures/setups. So unless we test some subs ourselves, we won't be able to answer that.

Comparing your distortion measurements of the L12, it's below 1% until 50Hz, and (probably?)less than 3% at 30Hz. Judging by that, I would say your distortion measurements are at least somewhat comparable to data-basses distortion measurements. They may not be exactly the same, but your L12 distortion measurements are pretty much what I would expect data-bass to measure, judging by the bigger Rythmiks they've measured. Also, I think that as long as it's within the same ballpark, small differences shouldn't matter all that much(given how insensitive we are to distortion at 30Hz).

You're right, though. It's not exactly apples to apples. To get a better idea of how comparable your measurements are to data-bass, you'd have to measure a sub that they've measured. Or, they would need to measure the L12.

Just going by what you've measured, here's the L12 @ ~102dB(?) side by side with the F328Be @96dB

Looks pretty comparable, which is very impressive for the F328Be. I've been looking back through old reviews to try and find something with less distortion, but I haven't found anything. I think the 328 may be the lowest distortion loudspeaker you've measured yet.

 

[Will2campb]

The measurements from the review of the 328be indicate that its distortion, including bass distortion, is very low. A subwoofer’s distortion (for example, the JL audio e-sub 110) is significantly higher than the bass distortion of the 328be. First, how audible is low frequency distortion? Second, would there be any benefit to adding two subwoofers to the 328be? Or would it actually be detrimental because of the increased distortion?

Thanks for all the excellent responses!

 

[andreasmaaan]

I guess it depends on the sub. 328 shows a little less than 3.0% at 30Hz and at 96dB.

Keep in mind that, at “96dB”, the F328Be is only producing c. 82dB @ 30Hz.

If it were called on to actually produce 96dB @ 30Hz, you could expect distortion to be vastly higher than what you see on the ASR graphs (but still no doubt quite low compared to many other speakers).

 

[ernestcarl]

I am also wondering how does the 328be bass distortion compare to servo Rythmiks? In the @amirm Rythmik L12 review it looks like distortion is similar, about 30dB down at the bottom end, but it isn't clear to me what SPL level was used for that measurement (was it ~100dB SPL?)

If one can afford the Revel F328be, then perhaps one could get bigger & better subs to match.

I don't have the Rythmik L12, but I do have the older F12 version.

Depending on the usual listening volume, one might be able to get away with a smaller sub(s).

The ff. aren't anechoic measurements, but are already summed with my speakers and taken in-room at the main listening position. All kinds of harmonic and non-harmonic distortion + noise from all the drivers coupling with room acoustics is included.

With the settings above, I believe it took over an hour to finish.

Comparing the values to the table image pozz linked to:

Seems to be a typo in the 2nd harmonic value on that printed table from AES

When I increase the volume to near my interface's maximum (-3dBFS) so I get ~95dB SPL a calibrated volume level (actual magnitude of the bass frequencies is higher):

highlighted exceeds perceptual threshold limits (based on said research table)

Hmmmn... It looks like the 3rd harmonic distortion is particularly audible between 10-20Hz (maybe moreso the harmonics?) -- duh, of course!


Here's a different view at the same vol. level (1M length)

Not really sure how relevant and reflective of reality the values are if plotted this way, also taking into account the sub-speaker xo probably minimizing the effect higher up.

 

[richard12511]

Keep in mind that, at “96dB”, the F328Be is only producing c. 82dB @ 30Hz.

If it were called on to actually produce 96dB @ 30Hz, you could expect distortion to be vastly higher than what you see on the ASR graphs (but still no doubt quite low compared to many other speakers).

Interesting, I guess I never thought about that aspect. So really, speakers like this that are pretty bass light are at a big advantage when it comes to distortion performance?

 

[andreasmaaan]

Interesting, I guess I never thought about that aspect. So really, speakers like this that are pretty bass light are at a big advantage when it comes to distortion performance?

If you specify distortion at a specific voltage input rather than at a specific SPL output, yeh

EDIT: @richard12511, actually I can answer this question better:

The reason for the confusion here is that Amir measures distortion at a fixed input voltage. However, he determines this voltage based on the speaker's output in the midrange.

So we really have here a distortion measurement that is, technically speaking, specified at a fixed input voltage, but that's reported as though it's at a fixed output SPL.

 

[detlev24]

The measurements from the review of the 328be indicate that its distortion, including bass distortion, is very low. A subwoofer’s distortion (for example, the JL audio e-sub 110) is significantly higher than the bass distortion of the 328be. First, how audible is low frequency distortion? Second, would there be any benefit to adding two subwoofers to the 328be? Or would it actually be detrimental because of the increased distortion?

It is safe to assume that our hearing cannot perceive up to ~10% THD in the bass region; and most probably considerably higher amounts in the 'sub bass' region.

For example, "Sound & Recording" and other well-established magazines aiming at audio professionals have their tolerance at 10% THD in the bass region (3% THD above); based on recent psychoacoustic findings. You can see this in their measurements, e.g., of the superb Genelec 8361A [EN translation].

A while back, I did simple but reproducible measurements of my 'JBL M2' in-room and they outperform many subwoofers; in terms of (especially) frequency response and magnitude of THD. However, there are some reasons for adding one or more subwoofers - even to 'reference' loudspeakers (as I did with a pair of 'Rythmik Audio G25HP'): Higher max. SPL at lower distortion in the bass region [mainly of interest for movie explosions] and if crossed-over to the main loudspeakers + properly time-aligned, potentially less distortion in the upper bass and MF regions; which might lead to improved clarity and minimal ringing [differences most likely not audible with the 'JBL M2' in a "standard" listening environment]. Obviously, there would also be an extension of frequency response in case of non-full-range main loudspeakers.

In my specific case, the G25HP_s show a similar frequency response in-room to the 'JBL M2' aside. However, I would need to repeat the subwoofer measurements under the same conditions, 99.54 dB @ SPL 3.28 feet [= 1 meter], for comparable distortion values. I forgot I did - at least with a summed and room EQed LFE-channel. The differences might become audible only at much higher SPL, I guess, especially if using two or more 'reference' subwoofers like these, summed to a mono-channel.

So, yes, judging from amirm's measurements, a lot more "weight" could be given to the music by adding one or more high-fidelity subwoofers to a 'Revel F328Be' setup. If considering Rythmik Audio; I would suggest a pair of at least 'F12', to keep distortion low [and to allow for better "treatment" of room modes, if a wider listening area needs to be covered].

 

[Robh3606]

Or would it actually be detrimental because of the increased distortion?

On the flip side how much lower would the distortion be if you highpassed the 328Be @ 80Hz?? I don't think you have to worry all that much at sub frequencies and any rise there may be countered by a reduction above.

Rob

 

[ernestcarl]

On the flip side how much lower would the distortion be if you highpassed the 328Be @ 80Hz?? I don't think you have to worry all that much at sub frequencies and any rise there may be countered by a reduction above.

Rob

The distortion is already so low it's not even necessary to high pass the speakers -- unless if you're trying to avoid/minimize some SBIR related issues. You can overlap the subs and speakers and then linearize the overlapping area with some EQ.

 

[lovemusic]

If one can afford the Revel F328be, then perhaps one could get bigger & better subs to match.

Have been using it for few months now. I don't think these speakers need a sub, unless someone is into heart throbbing movies. For music, which is my primary use, these are outstanding standalone speakers.