Elac Uni-Fi 2.0 Review (bookshelf speaker)

[amirm]

My guess is really the midrange is rubbing on the tweeter since there's very little tolerance and a bump in shipping could dislodge it just enough. Try playing a test tone at the resonant frequency and then put your hand on the tweeter and gently push it in different directions to see if that's really what it is.

You can't guess what this is if you have not heard the artifact. That aside, did you say to put my hand on the tweeter? You should never put your hand on the tweeter much less pushing it in different directions. That is a recipe for damaging it. In this case it is an impossible exercise anyway because there is a rigid metal grill on top of it.

If you want to help figure out what this is, do as I explained with turning up the volume. Speculating without basis and making excuses regarding shipping and such is not helpful. Playing at low volume and make a recording with a mic at a few inches has nothing to do with the scenario I have explained that causes this artifact. Sit back in a far-field listening position. Keep playing that segment of the song and notch up the volume step by step. Do not stop unless the woofer bottoms out or something (which it never did for me).

You are also welcome to send me your sample and I can test if you can't. Do you own it or did Elac loan it to you?

 

[amirm]

So it is clear, if you listen in near-field as you were doing, it gets too loud potentially before the problem occurs. I also tried to record the artifact at close distance but it did not work. Problem there I think is where to put the mic relative to the acoustic surface that is creating it. In far field this is not an issue.

 

[amirm]

Someone asked about the owner. He could not replicate the problem. He has just a few days to return and get his money back so I have to pack and send the speaker to him right away.

 

[Shazb0t]

did you say to put my hand on the tweeter? You should never put your hand on the tweeter much less pushing it in different directions. That is a recipe for damaging it. In this case it is an impossible exercise anyway because there is a rigid metal grill on top of it.

I'm guessing he meant that it's possible the tweeter grill is contacting the inner edge of the midrange under certain conditions and that you can likely lightly press on the tweeter grill slightly to see if that's an issue during playback without causing damage. I've never tried it with my UB5s, I'm not sure exactly how rigidly the tweeter grills are mounted in these speakers or if the tweeter mounting itself can be bent around by hand.

 

[jackocleebrown]

Great a expert like you @jackocleebrown is part of ASR community. if I may, what do you think about coaxial mid-tweeter speaker on axis response vs 20-30 degree off axis?
I always see they perfom better out of axis above crossover point.
KEF, ELAC, SEAS (C18EN002)
thanks in advance for your opinion
best

Hi Lorenzo, this is somewhat off-topic of this thread so apologies to everyone else.

Yes, that's generally true. With any rotationally symmetric waveguide, response irregularities will tend to be seen worst along the axis of rotation. This is simply caused by the symmetry. For example, say you have some feature of the waveguide that disturbs the HF wave resulting in some reflection or diffraction. This reflected/diffracted sound is loudest on the axis of rotation because sound from then entire feature arrives at precisely the same time and constructively sums. With a coincident driver, like Uni-Q, avoiding such problems is one of the major challenges. We normally recommend listening at least 10 degrees off axis, but there's also no major issue with going further than this because the MF and HF directivity are well matched and the direct response reaching the listener remains smooth. This means that the toe-in can be used to adjust the amount of direct sound versus room sound the listener hears (like a DI control) and it can help to tune the system to the acoustics of the listening room.

All the best,
Jack.

 

[joentell]

It has nothing to do with the room as the artifact is well above bass frequencies where room is dominant. Regardless, I have given you a specific hint as to what to look for. You should be able to listen past other factors and detect it. Your youtube video actually makes it easier since you have a comparison of the original versus the one I submitted.

A critical listener can listen past extraneous factors and find impairments. While the fate of this speaker is not important in grand scheme of things, learning to hear these small artifacts is important.

I understand that it's past the transition frequency. But we all know that distance reduces the higher frequencies which you said it is when you said it was in her breath.

Hearing an issue in person allows you to detect it much more easily in a recording because you've been pre-conditioned to hear it already.

Yes, you've given a hint what to look for, but you've changed the specific point where it happens about 3 times since the review.

 

[joentell]

Having listened to your recording, the source, and Joe's video on 4 different speakers(one of those being the M105) now, I do think the distortion you're talking about is more on the "g" than it is the "h". It seems to me that there is actually some sort of distortion(echo, mic clip, something else?) going on in the track itself, but then the Elac is magnifying that distortion 10 fold, and/or maybe exciting some other sort of metallic sounding resonance. On all my speakers, there is something slightly unpleasant about that "g", but it's hard to describe. The distortion is similar to the distortion I hear on your recording, but much, much less loud. On my clearest speakers(haven't tried headphones), it almost sounds like 2 distinct sounds, first the "g" then the distortion tied to the reverb immediately after? On my less clear, more enveloping speakers, it kinda blends the distortion in with the "g", making it a little less audible.

@joentell , you may not have been playing loud enough in your test. Comparing the amount of reverb present in your recording to Amir's, it sounds like Amir's recording was recorded at a much higher volume. What spl(pink noise?) was your recording taken at? Also, I really liked the end of the video where you slowed all the tracks down. Good idea, and I found it quite helpful for hearing the difference.

I was playing it at 10dB down from reference at my listening position 13ft away according to my AVR. Those are set for 85dB with headroom for +20 dB peaks. I had my camera about a foot away. In other words, pretty freaking loud! If it wasn't going to make the sound at that volume, i don't think it would have without smoke eventually coming out of something. Lol. It sounds like my voice is about the same volume, but it wasn't, it was way lower. Remember I level-matched the entire video to -16 LUFS.

 

[BDWoody]

Man, this thread really captures the essence of "forest for the trees". LOL

What do you expect from this herd of cats?

 

[joentell]

You can't guess what this is if you have not heard the artifact. That aside, did you say to put my hand on the tweeter? You should never put your hand on the tweeter much less pushing it in different directions. That is a recipe for damaging it. In this case it is an impossible exercise anyway because there is a rigid metal grill on top of it.

If you want to help figure out what this is, do as I explained with turning up the volume. Speculating without basis and making excuses regarding shipping and such is not helpful. Playing at low volume and make a recording with a mic at a few inches has nothing to do with the scenario I have explained that causes this artifact. Sit back in a far-field listening position. Keep playing that segment of the song and notch up the volume step by step. Do not stop unless the woofer bottoms out or something (which it never did for me).

You are also welcome to send me your sample and I can test if you can't. Do you own it or did Elac loan it to you?

The tweeter has a hard grill so you wouldn't be touching the soft dome at all. The way that concentric driver is setup, the tweeter unit is placed inside of the midrange's voice coil which of course moves up and down. The tweeter unit is centered within the coil using a screw which allows for minor adjustments. If the tweeter gets unseated slightly, it can cause that squeak sound at various frequencies. The tolerances are very tight so even the slightest variation could cause it to rub.

I'm speaking from actual experience having had experience with a "squeaking issue" on the previous Uni-Fi 1.0 model. I've owned over 6-7 pairs so I have an idea. The problematic one was a used one I purchased.

I will not be sending mine in for review since they're my main speakers. Also, it's clear from my video according to you that it isn't audible on my set. Maybe reach out to Elac and see if they'll send you a pair.

 

[joentell]

So it is clear, if you listen in near-field as you were doing, it gets too loud potentially before the problem occurs. I also tried to record the artifact at close distance but it did not work. Problem there I think is where to put the mic relative to the acoustic surface that is creating it. In far field this is not an issue.

I don't have that issue with getting too loud because I can adjust the gain manually on my mic and camera so that it will not clip even at the loudest volume possible.

 

[joentell]

Hi Lorenzo, this is somewhat off-topic of this thread so apologies to everyone else.

Yes, that's generally true. With any rotationally symmetric waveguide, response irregularities will tend to be seen worst along the axis of rotation. This is simply caused by the symmetry. For example, say you have some feature of the waveguide that disturbs the HF wave resulting in some reflection or diffraction. This reflected/diffracted sound is loudest on the axis of rotation because sound from then entire feature arrives at precisely the same time and constructively sums. With a coincident driver, like Uni-Q, avoiding such problems is one of the major challenges. We normally recommend listening at least 10 degrees off axis, but there's also no major issue with going further than this because the MF and HF directivity are well matched and the direct response reaching the listener remains smooth. This means that the toe-in can be used to adjust the amount of direct sound versus room sound the listener hears (like a DI control) and it can help to tune the system to the acoustics of the listening room.

All the best,
Jack.

That's cool that you're here responding. I know Andrew Jones has a history with KEF having worked there in the past.

Andrew Jones also recommends listening/measuring approx. 10 degrees off-axis for the exact reasons you just stated. Great minds....

 

[JIW]

I have made spectrograms of the section in question. First, the mono of the original track followed by @amirm's recording. Then, the same for the waveforms. The peak in the waveforms at 56 s and 3.5 s, respectively, and the blob at 600 Hz on the spectrograms at 56 s and 3.5 s, respectively, are also the 'gone'. This can be seen if the track is trimmed to just before it (attached for download). This is shown in the bottom two figures.

 

[joentell]

It should be no surprise to anyone that when she says "he goes" on that track, that it causes the issue on @amirm's review unit. He measured the resonance at around 600hz and her voice is right there when she says that. I made a video to help illustrate it for anyone interested.

 

[richard12511]

It should be no surprise to anyone that when she says "he goes" on that track, that it causes the issue on @amirm's review unit. He measured the resonance at around 600hz and her voice is right there when she says that. I made a video to help illustrate it for anyone interested.

That video makes it really clear. You can clearly hear the metallic resonance in Amir's video, but it's absent in the other two. Perhaps it really is a defect with that particular unit.

 

[Rock Rabbit]

It should be no surprise to anyone that when she says "he goes" on that track, that it causes the issue on @amirm's review unit. He measured the resonance at around 600hz and her voice is right there when she says that. I made a video to help illustrate it for anyone interested.

Sure, she goes to D5 on the song...."away" y ends in D5 too, + low frequency artifacts...a nightmare for the 3 drivers (at -26 dB)
https://therangeplanet.proboards.com/thread/947/eva-cassidy
https://www.audiosciencereview.com/...-0-review-bookshelf-speaker.19216/post-632055

 

[mhardy6647]

I had auditioned the elac adante a couple of years ago and found the sound a bit dark, maybe they suffered from the same issue of the higher frequencies being shelved down

The inexpensive little ELAC "Debut 2.0 b6.2", also designed by Mr. Jones, manifests the same (well... at least similarly) reticent treble -- although its immediate predecessor (an example of which I have not heard) was reputed to be sort of bright.

 

[amirm]

I will not be sending mine in for review since they're my main speakers.

So the answer to my question is clear I will repeat the question: do you own these speakers or are they loaners from Elac?

If they are loaners, then I can ask Andrew to get your sample to me since you say it is working fine.

 

[amirm]

I don't have that issue with getting too loud because I can adjust the gain manually on my mic and camera so that it will not clip even at the loudest volume possible.

I don't know why this is so hard to communication. Earlier you told us your mic would clip if you play too loud:

There's probably enough power in here to blow the speaker with all channels driven. The mic will likely clip before the speaker does at a close distance.

So I suggested to avoid that altogether and listen and record at far field. Keep turning up the volume even if it is uncomfortable to listen to. Do not stop at whatever you think is the proper listening level. We are trying to see if your sample has this issue at all. This is the scenario under which the problem appears. Any other test you do is not helpful or related to this problem. If you don't want to do this, or loan the speaker for me to test, then really I am sure what help you are offering.

 

[mtmpenn]

These posts seems a bit hostile for “chief fun officer.”

 

[Lorenzo74]

I've forwarded this post to Andrew Jones so hopefully he'll chime in. He's not really the type to shy away from stuff like this from what I've seen, but who knows. That's up to him.

thank you @joentell , having Andrew Jones here would be great. this is the best thing you can actually do (beyond your efforts in recording “that” songs on your setup, or loan your speaker to Amir)

I met A.J. at CES2019 and he is exceptional in describing speaker design, objective data and subjective results. I really look forward to hear what he might tell us.
I also believe ELAC is packing a pair of ”pre-measured” UB5 2.0 right now to be delivered by a dedicated truck directly to Amir‘s house. This is the most visited website (just overtake stereophile afaik) globally.
Can they leave the new “concentric“ panther head-less? No...!
I belive @amirm will fix it again as he did for KH80 discovering Seattle garages are colder than Berlin labs.
my best to all
Lorenzo