Elac BS U5 Slim 3-way Coaxial Speaker Review

[BurgerCheese]

I decided to do some casual distortion measurements on my pair of UB5 slim as I couldn't remember the distortion being as bad as in this review.

Tested with REW and UMIK-1. Observations:
At 86dB/1m, distortion peaks at:
- 270 hz, 1.5% THD, 2nd+ 3rd order. Appears to come from midrange.
- 550 hz, 1% THD, 2nd + 3rd order. Appears to come from enclosure (resonance)
- 700 hz, 1% THD, mostly 2nd order. Appears to come from enclusure (resonance). But only on right speaker!
- 1700hz, 1% THD, mostly 2nd order. Appears to come from enclusure (resonance). But only on left speaker!
- 2700hz 1.5% THD, mostly 3rd order. Appears to be midrange cone breakup. Distortion peak not present when measuring off-axis.
- 8000hz 0.5% THD, 2nd order. Tweeter breakup? But only on left speaker!

At 96db/1m, distortion peaks at:
- 270 hz, 3% THD, mostly 2nd order. Appears to come from midrange.
- 550 hz, 2.5% THD, 2nd + 3rd order. Appears to come from enclosure (resonance)
- 700 hz, 2.5% THD, mostly 2nd order. Appears to come from enclusure (resonance). But only on right speaker!
- 1700hz, 2% THD, mostly 2nd order. Appears to come from enclusure (resonance). But only on left speaker!
- 2700hz 2.5% THD, mostly 3rd order. Appears to be midrange cone breakup. Distortion peak not present when measuring off-axis.
- 8000hz 1.5% THD, 2nd order. Tweeter breakup? But only on left speaker!

Distortion figures are a bit lower than what amirm measured, but still pretty bad. The differences between right and left speaker might indicate poor QC.

 

[ROOSKIE]

300Hz might be pushing it, but as this paper showed, a crossover frequency of 120 Hz is not localizable, with half of the listeners not even being able to localize the subwoofers at the highest crossover frequency tested of 227 Hz.

Anecdotally, I've read that the often quoted standard crossover of 80 Hz was recommended by THX partially because it was two standard deviations below the lowest frequency localizable, just to be sure. It should also be considered that the LFE (Low Frequency Effects) channel on movie soundtracks goes up to 120 Hz, which is specifically for subwoofers, definitely not intended to be localizable, so this frequency was most likely chosen for a reason.

I've never really had the opportunity to test this blind but I have had a difficult time with this. I usually have ended up using a crossover for a single sub around 60hrz. Above that and I am able to localize the bass once I settle in and listen for awhile and that bothers me. It may be other factors such vibrations and obviously knowing where the woofer is.
Multiple subs and or placing the sub in the center in between the speakers seems to fix this though.
One sub just isn't enough I suppose.

 

[thefsb]

At this point speakers became "good." They provide a diffused image

Can you elaborate? I can't relate to the word image in listening to a single channel. And what's it like when it is diffused compered to not?

 

[Esskay]

I am surprised that the Elacs performed poorly- objectively in amirm's tests and subjectively according to many. I heard the UB5 Slims at a group demo with Andrew Jones. I thought they sounded great (for the price) and so did everyone I spoke with. But Jones picked nearly all the music and "audiophiles" picked the rest. This could be an example of us admiring the sound of the sources rather than that of the system.
SK

 

[981CS]

I think this should only matter if you listen really loud. Amplification shouldn't affect the sound, so if they're fatiguing and shouty on one amp, then that should be true for all amps.

Not necessarily true. It takes significantly more power to produce bass notes than it does mid/high frequencies. So even when listening at lower/medium volume, remember than this is a 4ohm speaker with very low sensitivity and it is capable of hitting down to 40Hz or so. It takes power to get those dynamics.

Again, Hoffman's Iron Law...

 

[amirm]

Can you elaborate? I can't relate to the word image in listening to a single channel. And what's it like when it is diffused compered to not?

It is whether the sound seems to be coming from the point source of the speaker, or a diffused image at and around it. This was the latter. Just turn off one of your channels and see what your perception is.

 

[thefsb]

It is whether the sound seems to be coming from the point source of the speaker, or a diffused image at and around it. This was the latter. Just turn off one of your channels and see what your perception is.

Thanks. I see what you mean. The Genelec near-field monitors on my desk sound rather like a point source individually. Not surprising.

 

[aarons915]

Not necessarily true. It takes significantly more power to produce bass notes than it does mid/high frequencies. So even when listening at lower/medium volume, remember than this is a 4ohm speaker with very low sensitivity and it is capable of hitting down to 40Hz or so. It takes power to get those dynamics.

Again, Hoffman's Iron Law...

Assuming the same sensitivity in the bass as the mids/highs, this just can't be true. An amplifier is a constant voltage device so if it's sending out 1 Volt to your speakers, then a 50Hz signal will take the same power as a 500Hz signal assuming the same impedance. Of course a speaker with low sensitivity is going to require more power than a higher sensitivity speaker to achieve the same SPL, most people who buy low sensitivity speakers are aware of that.

 

[981CS]

Assuming the same sensitivity in the bass as the mids/highs, this just can't be true. An amplifier is a constant voltage device so if it's sending out 1 Volt to your speakers, then a 50Hz signal will take the same power as a 500Hz signal assuming the same impedance. Of course a speaker with low sensitivity is going to require more power than a higher sensitivity speaker to achieve the same SPL, most people who buy low sensitivity speakers are aware of that.

Also remember how the human ear works with perception to bass, aka Fletcher Munson curve.

Mid/high's are also 6db-10db more efficient than woofers, perhaps more in this particular speaker. And we know that bass notes take more power than mid/highs. If the woofer is really inefficient in the UB5, this could explain some things.

I often wonder if there's something about the UB5 that's somewhat funky about how the mid/tweeter integrates with the woofer. Because all of my tests have the UB5 sounding "flat" on small amps, but with an amp that had some genuine "grunt" (like a 100wpc+ A/B with a proper torodial power supply and MOSFETs) the UB5's reward you with a huge dynamic performance, regardless of how loud you turn it up.

 

[WHO23]

Speaker Shootout (Left to Right):

1| Ascend Acoustics Sierra-2EX
+Clean bass & treble, biggest sound, best sub-bass, the attack and decay are astounding but not always noticeable
-Vocals lack presence, needs good positioning
*Currently using MiniDSP SHD with Dirac. These speakers sounded amazing at first but I found myself constantly changing the volume. I was struggling to hear the vocals at times and when I did increase the volume, the bass was overpowering. After EQ-ing, these speakers were by far the best of the bunch as I never found myself nitpicking about it and it just disappeared into the background. It could play loud enough for movies, and I never found myself needing a sub. During the shootout, it got two votes for the best speaker of the bunch. One person found it didn't sound "high-end."

2| Focal Aria 906
+Natural vocals, best mids, driver integration is seamless, best build
-Treble can get slightly sharp, bass has punch but no sub-bass rumble
*During the shootout, this was the best sounding for music right out of the box. Vocals on it was noticeably clearer than the others. However on some music there is a slight harshness or sibilance and detracted from the music. The lack of bass is not as noticeable on music, but for movies the lack of rumble on the sub-bass made explosions unrealistic. During the shootout, it got one vote for the best speaker of the bunch. One person found the harshness/sibilance annoying.

3| KEF Q150
+Point source imaging, easy to position, best when used near-field
-Doesn't play very loud, doesn't have out of room sound when spaced far apart (sounds the smallest), bass gets distorted easily
*During the shootout, this was instantly deemed the least impressive of the bunch. They sounded small, due to being spaced far apart. Also when level matched, this speaker struggled to play at moderate volume levels as we could hear the driver rattle. However, I have since moved them to a small desk to be used just beside my laptop and they sound phenomenal. When listening near-field, the center image and the bass appeared - they now sounded bigger than the box.

4| ELAC Uni-Fi UB5
+Warm sound, best center image, big sound
-Upper treble tapers off, can hear high frequencies distort compared to other speakers
*During the shootout, all three of us really enjoyed this speaker although it was lacking the "high-end" sound. It sounded big, fun and enjoyable. I suspect we just enjoyed hearing a speaker that's more noticeably warm compared to the others. Listening to some tracks however, we found that we couldn't hear the triangle instrument properly and it lacks sparkle.

Gear used: PC (FLAC) -> SMSL SU-8v2 -> Hattor Audio Mini Preamp -> Audiophonics Hypex NC252MP (MPA-252NC XLR) -> Speakers
Test conditions: Untreated bedroom, no stands but head level, no subwoofer, listen at 82dB max SPL@5m, measured 91dB max SPL@1m
I ordered an Asipilin LS3/5a (15 ohm) but it did not arrive in time for the shootout.

Bonus Round| Devialet Phantom Silver (Mono)
+Plays extremely loud, can fill a whole room with sound, plays well off-axis
-Frustrating UI, can't use own DAC/Amp, stereo setup issues
*This isn't part of the shootout as it's located in the living room but we listened to it on the same day. The Devialet sounds huge for such a small unit. The ample amount of sub-bass is the most noticeable compared to the speakers above. We could sit anywhere and be happy to have music played without distortion. After extended listening, the bass gets tiring and I'm considering using the MiniDSP SHD here as the Phantom only has an Optical In connection. Also note that I do not recommend using these on a stereo setup.

Early this year I did a shootout featuring ELAC UB5, KEF Q150, Focal Aria 906, and Ascend Acoustics Sierra-2EX <Results in the quote above>

Although we felt that the UB5 was the least analytical of the bunch, we felt the sound was very agreeable and enjoyed music out of it. However out of the list of speakers the shootout, the only one that I have sitting in my cupboard is the UB5.

Recalling the shootout, I didn't even notice the distortion in the mids but definitely felt the treble was more rolled off compared to the other speakers. As primarily a headphile, I can notice distortion easily on headphones but not as much on speakers, particularly since my room is small and untreated.

 

[rebbiputzmaker]

I am surprised that the Elacs performed poorly- objectively in amirm's tests and subjectively according to many. I heard the UB5 Slims at a group demo with Andrew Jones. I thought they sounded great (for the price) and so did everyone I spoke with. But Jones picked nearly all the music and "audiophiles" picked the rest. This could be an example of us admiring the sound of the sources rather than that of the system.
SK

They are not nearly as bad as they are being made out here. It’s always a good idea to listen and make ones own decision.

 

[rebbiputzmaker]

It is whether the sound seems to be coming from the point source of the speaker, or a diffused image at and around it. This was the latter. Just turn off one of your channels and see what your perception is.

Listening to one channel could sometimes be deceiving depending on how the music is recorded. One channel of a stereo recording may not have all the recorded information for proper single channel listening. I would say success is dependent on the recording that you are using for your evaluation.

 

[QMuse]

Listening to one channel could sometimes be deceiving depending on how the music is recorded. One channel of a stereo recording may not have all the recorded information for proper single channel listening. I would say success is dependent on the recording that you are using for your evaluation.

That can be easilly solved - with Audacity it is very easy to mix stereo down to mono.

 

[rebbiputzmaker]

That can be easilly solved - with Audacity it is very easy to mix stereo down to mono.

Not necessarily, there could still be issues.

 

[tecnogadget]

That can be easilly solved - with Audacity it is very easy to mix stereo down to mono.

And pretty much with any AVR stereo downmix to mono. So yes, you can try Amir's recommendation easily.
I know pretty much nobody listens to music in mono. But testing speakers in Mono is quite an interesting exercise if you know what to listen for and understand the parameters.

 

[tecnogadget]

Not necessarily, there could still be issues.

Mind that they are cheering you up to do mono just for testing purposes. Not for critical evaluation of the recording content as if you were to judge a music track as a sound engineer focusing on panning, etc. Its about listening the speaker.

 

[rebbiputzmaker]

Yes testing, listening to and evaluating a speaker. How could anyone think that’s acceptable? How do you know any aberrations you might hear are not within the recording as opposed to the speaker? Also I don’t believe Amir is mixing anything, but just using one channel of a recording. In my opinion not a good idea. Ymmv

 

[tecnogadget]

Yes testing, listening to and evaluating a speaker. How could anyone think that’s acceptable? How do you know any aberrations you might hear are not within the recording as opposed to the speaker? Also I don’t believe Amir is mixing anything, but just using one channel of a recording. In my opinion not a good idea. Ymmv

Have you given a try to the experiment? If you own more than 1 set of speakers I encourage you to just make for this fun test and then tell us if you suspect you are listening such aberrations.
The same doubt you have would apply to stereo, if there is any aberration how could you tell is not the speaker nor the recording? Common sense dictates you use good recording tracks that you already know too well, and make sure through different playback mediums, speakers, high-quality headphones, etc - that no issue is inherent to the recording.

This works best if your amp has A-B switch, you connect different speakers to Righ and Left and switch back and forth. The tonal balance between the two speakers will be very easy to compare. The only hassle is level matching, for what I'm lucky both my speaker sets are very quite similar sensitivity.

 

[GelbeMusik]

300Hz might be pushing it, but as this paper showed, a crossover frequency of 120 Hz is not localizable, with …

Ja, if the sub is actually limited to its dedicated frequency range. But there is "chuffing" from reflex ports, there is harmonic distortion galore.

Also remember how the human ear works ... If ... this could explain some things.
... but with an amp that had some genuine "grunt" (like a 100wpc+ A/B with a proper torodial power supply and MOSFETs) the UB5's reward you with a huge dynamic performance, regardless of how loud you turn it up.

You depart from the function of the human hearing to land at the function of toroidal power supplies. There is no connection I'm aware of. You question it yourself, in that You say "could explain some things" but without actually explaining it. It is a myth, this saying that "small speaker need much steam". Everything that can be calculated and measured tell the opposite, take it for granted. Small speakers cannot stand, let alone transfer into good sound, much power.

 

[rebbiputzmaker]

Have you given a try to the experiment? If you own more than 1 set of speakers I encourage you to just make for this fun test and then tell us if you suspect you are listening such aberrations.
The same doubt you have would apply to stereo, if there is any aberration how could you tell is not the speaker nor the recording? Common sense dictates you use good recording tracks that you already know too well, and make sure through different playback mediums, speakers, high-quality headphones, etc - that no issue is inherent to the recording.

This works best if your amp has A-B switch, you connect different speakers to Righ and Left and switch back and forth. The tonal balance between the two speakers will be very easy to compare. The only hassle is level matching, for what I'm lucky both my speaker sets are very quite similar sensitivity.

Been doing this for over 40 years, currently own nine pairs of speakers. Probably owned 50 pairs overall, built a number of them. Have had every technology except for Walsh. Own a bunch of headphones, and IMEs also. Unless you use a true mono recording, (preferably homemade), you have no true idea what you are listening to, when comparing channels.

Also nowhere in the reviews is there any mention of a left - right channel comparison of speakers. I understand you do that which is cool, but that wasn’t mentioned here at all.

If you are building a two channel system, part of Determining the quality of a pair of speakers imo is how they function as a pair.

There also seems to be an overall misunderstanding of concentric drivers and their attributes imo. This could be mis-application, coloring these discussions.