Elac Uni-Fi 2.0 Review (bookshelf speaker)

[Rock Rabbit]

I do. Unfortunately as the weather has gotten colder, the problem has gotten worse. So for now, I have stopped using that monoblock and this test and recent other ones have been with the other monoblock.

So no, it is not related to that. Indeed as I reported -- and I have repeated -- I have played the exact track at levels matching or exceeding what this speaker is doing with Revel M105 and there is zero hint of this problem there.

But with the Revel you use hpf @ 30 Hz?

 

[amirm]

But with the Revel you use hpf @ 30 Hz?

No, no. The test was with no EQ with either speaker. I just moved the speaker wire from one and connected to the other.

 

[ctrl]

Peaks from left D2 D3 D4 A4...for resonance very little energy and two narrow resonances is a hint of membrane breakup
Modes http://hyperphysics.phy-astr.gsu.edu/hbase/Music/cirmem.html
https://www.acs.psu.edu/drussell/Demos/MembraneCircle/Circle.html

Thanks for the links, I didn't know the websites.

Such modes also occur with cone drivers, but of course not in a frequency range like with drum membranes.
The advantage of metalized cones is their increased stiffness and thus such chassis behave longer (i.e. up to a higher frequency limit) "ideally" than e.g. paper or polypropylene cone chassis.

From THD graph there's evident problems in two resonance peaks starting at 585 Hz 4% next some 620 Hz 3%, these are not visible in FR and there's a small change in impedance ...
Generally narrow peaks with similar frequency can be associated with cone membrane breakup (metal cone), narrow or high Q is only possible from mechanical resonance.

I would be surprised if the midrange would already show break-up modes around 1200Hz (which then lead to increased HD2 around 600Hz).
Would rather expect a surround resonance in this frequency range as the cause for the increased HD2 around 600Hz.

Due to the crossover frequency at 2 kHz, the break-up modes of the metal midrange driver are suppressed very well, since no noticeable harmonic distortions show up in the 700-2000Hz range.

[some speculation]
The cause for the small peak in the impedance frequency response around 620Hz, could also be the rear vent of the pole piece (in which the tweeter sits) of the coax chassis, which acts like a Helmholz resonator.
Perhaps this is also the cause or is partly responsible for Amir's hearing impressions.
.

 

[JIW]

The track has DC component at the start of "goes away" !
View attachment 104955
Edit: midrange moving too? Better process the track with a 20 Hz hpf, it seems some DC is going with bass line (some effect?)

There is seemingly a fair bit of DC in the track intermittently. Y-axis is frequency in Hz. However, this seemingly goes all the way up the spectrum. Further, since the Revel did not misbehave and EQ ameliorated it, this does not seem to be he source of the resonance problem but probably the reason the woofer stood still.

Here is the same for the full track.

 

[MZKM]

I do. Unfortunately as the weather has gotten colder, the problem has gotten worse. So for now, I have stopped using that monoblock and this test and recent other ones have been with the other monoblock.

For the Infinity tower, did you test it on this other monoblock?

 

[posvibes]

Rookie question, what is a "shelf" and why/why not put one in?

 

[amirm]

For the Infinity tower, did you test it on this other monoblock?

I did not. I think it is safe to assume the issue there was the amplifier as well, not the speaker.

 

[Todd74]

FYI Eva Cassidy album is standard issue at all audio shows in multiple suites so it is not like it is some oddball track one never sees.

Have Freya Ridings & Halie Loren become standard issue female vocals yet?

 

[amirm]

Rookie question, what is a "shelf" and why/why not put one in?

A "shelf" is a type of filter where from one frequency on, the level is raised or lowered. That is, it looks like a shelf in frequency response measurements. There is usually no good reason for a speaker to permanently overlay such a thing on your music. On the other hand, if the speaker comes that way, then an inverse shelf filter can correct it.

 

[Rock Rabbit]

Thanks for the links, I didn't know the websites.

Such modes also occur with cone drivers, but of course not in a frequency range like with drum membranes.
The advantage of metalized cones is their increased stiffness and thus such chassis behave longer (i.e. up to a higher frequency limit) "ideally" than e.g. paper or polypropylene cone chassis.


I would be surprised if the midrange would already show break-up modes around 1200Hz (which then lead to increased HD2 around 600Hz).
Would rather expect a surround resonance in this frequency range as the cause for the increased HD2 around 600Hz.

Due to the crossover frequency at 2 kHz, the break-up modes of the metal midrange driver are suppressed very well, since no noticeable harmonic distortions show up in the 700-2000Hz range.

[some speculation]
The cause for the small peak in the impedance frequency response around 620Hz, could also be the rear vent of the pole piece (in which the tweeter sits) of the coax chassis, which acts like a Helmholz resonator.
Perhaps this is also the cause or is partly responsible for Amir's hearing impressions.
.

Metal treated cones are better in theory but resonance (breakup) is more complicated.
Port resonance is relevant in 2 way systems but in this case the problem seems the midrange (3 way system!) that is isolated from the box.
Finally there's a DC component issue in the music track that could be part of the problem (at least woofer movement), but the same track has no problem with other speakers.

 

[hardisj]

A buddy and I were talking about this thread. I sent him the link and he, too, was perplexed by the issue Amir is noting ~600Hz. And, he made a really good point about this being an example of why accepting a single unit for test is problematic for not just the inability to check for QC error but... (his quote):

If he's asking for people to send him a single speaker to test, then for sure they aren't sending it in the original packaging designed for a pair of speakers (FYI the box is huge on the ELAC's because they have thick foam.)

And, honestly, I think he makes a very fair point here. Do we know how the speaker was shipped? Is it possible it was damaged or banged around in transit due to not being packaged in the OEM packaging? And, is something members should be mindful of when sending in a single sample? It just seems unlikely (not impossible, mind you) this is a QC issue. I'm not trying to make anyone look foolish for bad shipping... crap happens. I wrapped up an amplifier a few months ago like a TANK and it showed up with dented fins and I had to refund the buyer and deal with UPS over it.


(yes, I understand there is additional cost with sending a pair as well as the "hey, if it comes from the factory bad, it's not my issue". Totally get that. But if sending in the OEM packaging, it might not be as bad, especially for bookshelf speakers and would go a long way toward the science of things here in cases just like the one presented here)

 

[joentell]

A buddy and I were talking about this thread. I sent him the link and he, too, was perplexed by the issue Amir is noting ~600Hz. And, he made a really good point about this being an example of why accepting a single unit for test is problematic for not just the inability to check for QC error but... (his quote):




And, honestly, I think he makes a very fair point here. Do we know how the speaker was shipped? Is it possible it was damaged or banged around in transit due to not being packaged in the OEM packaging? And, is something members should be mindful of when sending in a single sample? It just seems unlikely (not impossible, mind you) this is a QC issue. I'm not trying to make anyone look foolish for bad shipping... crap happens. I wrapped up an amplifier a few months ago like a TANK and it showed up with dented fins and I had to refund the buyer and deal with UPS over it.


(yes, I understand there is additional cost with sending a pair but if sending in the OEM packaging, it might not be as bad, especially for bookshelf speakers and would go a long way toward the science of things here in cases just like the one presented here)

To add to that, shipping a single speaker in OEM packaging designed for two speakers in order to save on shipping costs will lower the integrity of the protectiveness of the packaging foam. In other words, it needs to have two speakers in there to be effective.

I plan on doing a recording of my Uni-Fi 2.0 left, center, and right speaker playing the "problem song" along with a few other speakers to compare them to.

 

[whazzup]

Amir should come clean. This is obviously a ploy with Eva's record labels to increase her posthumous recognition.

**In other news, multiple streaming companies have reported a surge in the streaming of the late Eva Cassidy's albums on their platforms. Some have tied this surge to a particular online review of speakers and her ability to cause said speaker to break down...**

 

[amirm]

And, he made a really good point about this being an example of why accepting a single unit for test is problematic for not just the inability to check for QC error but... (his quote):

In general, when speakers come as pairs in a box, they are sent to me as a pair even though I only test one. In this rare case, the owner managed to find a perfect box to fit the single speaker in, including its oversized Styrofoam. I see zero signs of any kind of damage or scuff marks. If there is a problem, it is not due to packing.

 

[amirm]

Amir should come clean. This is obviously a ploy with Eva's record labels to increase her posthumous recognition.

OK, I will come clean. This used to be one of my favorite tracks. Now, after listening to it 200 times. I am darn sick of it!

I was so tired it that I went to hit pause when the glitch happened with this track! A few seconds later and I would not have found it.

I have ruined a number of my favorite music tracks this way....

 

[MZKM]

I did not. I think it is safe to assume the issue there was the amplifier as well, not the speaker.

Would that affect the rating you would give it? You gave it a shrugging panther, would you now give it a laying down/relaxed panther (you still had to use EQ to take a distortion/resonance)?

 

[amirm]

Would that affect the rating you would give it? You gave it a shrugging panther, would you now give it a laying down/relaxed panther (you still had to use EQ to take a distortion/resonance)?

Maybe half a point.

 

[MZKM]

Maybe half a point.

Still weird that the Salon2’s didn’t make them go into protection, as even the EPDR of the R253 didn’t look terrible.

So, are you going to have it fixed or transition to some Purifi/Benchmark amp?

 

[amirm]

So, are you going to have it fixed or transition to some Purifi/Benchmark amp?

It is some 100 pounds. I dread even packing it let alone sending it to Harman for repair.

I have a Mark Levinson stereo No 532 Reference, 400 watt/channel amp (into 8 ohm) as "back up" so won't be going to Unifi route. She is small compared to the No 53 but a monster in its own right:

Weighs 120 pounds....

 

[amirm]

I plan on doing a recording of my Uni-Fi 2.0 left, center, and right speaker playing the "problem song" along with a few other speakers to compare them to.

Be sure you have good amount of amplification. This is an inefficient speaker and the problem becomes obvious at elevated volumes. So you need a lot of power. Once you hear the artifact, then you will hear it at lower volumes as well.