KEF LS50 Meta Review (Speaker)

[amirm]

I think that you missunderstood me - the bass unit is modulating the tweeter depending on cone placement.
In Erins measurement theres only the tweeter sounding in this particular measurement. On the bass unit he has a battery, no amplifier.

And the difference in frequency response is big for the tweeter depending on where the bass cone is.

No I didn't misunderstand you. I explained that by the time the woofer is moving that much, the distortion it generates is much of a concern than whatever tweeter is doing. This was certainly the case in my listening tests. Once you see the cone moving any appreciable amount -- which is easy to observe given its shiny surface -- the bass is already getting distorted.

 

[amirm]

Maybe it isn't going away because the re-appraisal of the K3 is buried in a the subjective listening section of the review of the Revel F35!

Sorry about that. I just added a bolded note to the listening section of R3 to indicate the issue with room mode.

 

[amirm]

One note about comparing these to active speakers. Those active speakers have more optimal acoustic design but their amplification puts a severe limit on how loud they get. With these passive speakers and a powerful amplifier, that limitation is not there and you are just limited by the excursion of the woofer.

 

[dfuller]

One note about comparing these to active speakers. Those active speakers have more optimal acoustic design but their amplification puts a severe limit on how loud they get. With these passive speakers and a powerful amplifier, that limitation is not there and you are just limited by the excursion of the woofer.

I mean, fair enough, but I've not experienced speakers actually going into limiting at "normal person" volumes save for Genelecs, which are unusually aggressive in the limiter department.

 

[q3cpma]

I mean, fair enough, but I've not experienced speakers actually going into limiting at "normal person" volumes save for Genelecs, which are unusually aggressive in the limiter department.

I doubt any active speaker of that price will trigger their limiter when used nearfield (as intended). Add pair gain and any potential boundary gain to that.
Though I could envision issues for some specific models like the 8260A or 8351A in hard midfield use without subwoofers.

 

[muad]

Price isn't the only factor though. There's also size. It's a small speaker. Physics are physics - a 5" doughnut (woofer with a big hole for the tweeter and phase plug) can only move so much air. So logically one would expect bass capability to be at or below a good 4" woofer monitor, e.g. Neumann KH 80. In their size class what does have strong bass capability? JBL 705 is the only one that I can think of offhand that I know of. Triad has a 5" minimonitor with a ScanSpeak Revelator midwoofer, but that speaker also has a primitive flat waveguide for the tweeter. Devialet I don't have first-hand experience with, but @napilopez has shown they go low in a usable way.

That leaves out the question of audibility. I'm in the camp that says pair them with "flanking subs" crossed relatively high and you'd have really nice full range stereo speakers. Too bad KEF hasn't found a way to protect the coax, though. A round funky color grille a la Amphion would go such a long way...

If you do the math, the surface area lost in the middle of the driver is miniscule, and is still much larger than a 4" driver. It's all about the diameter. I had the og ls50s and they sounded great. Pushed up close to the wall the boundary gain filled out the sound nicely. No 5" speaker doesn't need a sub in a big room...

 

[Aperiodic]

Not good. Until proven wrong, I consider that only Genelec and KEF solved the coaxial question.

Thanks. I remember that the reviews I saw (and mfr recommendations) recommended crossing in front of the main listening position. I can believe it.

Tannoy has also gone down the coax path, but... $$$$

 

[Sancus]

I mean, fair enough, but I've not experienced speakers actually going into limiting at "normal person" volumes save for Genelecs, which are unusually aggressive in the limiter department.

IMO the main reason for this is that Genelec EQs their bass extension up as much as possible for the driver capabilities, considerably more than many other monitors do. I mean this LS50 is playing 85dB at 60hz in the "96dB" test. The Genelec 8030C is trying to play ~94dB and it still comes in with comparable distortion. It's not really close, the 8030C clearly has the better woofer performance by a big margin. If you want it to play super loud, you need to EQ the sub-bass down or get a sub.

Personally I feel this design decision is actually a smart one, most people are going to want more bass extension than they are going to want high SPLs, but I can also see the other side of the argument. In either case it's got more to do with EQ choices than capability though.

E: I'm not saying people should buy that instead or anything, just comparing woofer performance on a factual basis. Given the choice, I would buy the LS50 because I strongly prefer coaxials.

 

[nothingman]

Relative newbie to the audio world and truly amazed by the level of knowledge many of you exhibit.

I currently have the LS50 Meta paired with a single KC62 and am pleased with their performance in a relatively small listening room. My question is what would the ideal Low Pass as well as High Pass setting be. I’m running a Parasound New Classic 200 integrated that gives me the option to set both.

Welcome! You’ll get a different answer from just about every person because a lot of it is “it depends.”

First, have you seen KEF’s recommendation for the WII and KC62? They suggest a very low crossover region. 45hz LPF (LR 24db/oct) and 70hz HPF (BW 12db/oct).

That’s way lower than you’re seeing people suggest here, but I’d try it as a starting point. Not least of why is because KEF seems to know what they’re doing. With the Metas maybe bump up both by 10hz since the WII has some DSP magic to get it to hit lower.

I’d guess KEF thinks you want to still get some bass out of the mains to round things out while the KC62 augments the true low end. This is at the expense of more aggressively relieving the Metas of bass duty and trying to squeeze at much SPL out of them as possible.

 

[Djdhifi]

Welcome! You’ll get a different answer from just about every person because a lot of it is “it depends.”

First, have you seen KEF’s recommendation for the WII and KC62? They suggest a very low crossover region. 45hz LPF (LR 24db/oct) and 70hz HPF (BW 12db/oct).

That’s way lower than you’re seeing people suggest here, but I’d try it as a starting point. Not least of why is because KEF seems to know what they’re doing. With the Metas maybe bump up both by 10hz since the WII has some DSP magic to get it to hit lower.

I’d guess KEF thinks you want to still get some bass out of the mains to round things out while the KC62 augments the true low end. This is at the expense of more aggressively relieving the Metas of bass duty and trying to squeeze at much SPL out of them as possible.

Appreciate the detailed response and the kind welcome! I actually did go with those recommendations by KEF for the WII but with all the talk of much higher crossovers in this thread figured I might be doing something wrong. I’ll definitely play around with both LP and HP a little more but probably won’t stray too far as it honestly sounds pretty good already. Thanks again!

 

[nothingman]

Appreciate the detailed response and the kind welcome! I actually did go with those recommendations by KEF for the WII but with all the talk of much higher crossovers in this thread figured I might be doing something wrong. I’ll definitely play around with both LP and HP a little more but probably won’t stray too far as it honestly sounds pretty good already. Thanks again!

Without two subs sitting very close to the mains you can kiss these ideas of 150-200hz crossover goodbye. I personally wouldn’t run a single KC62 above about 75-80hz on a normal LR 24db/oct just to make sure it isn’t localizable. Then on the HPF side you can play around depending on in-room response. If you’ve got the Metas up against a wall or have some room resonances in the 80-100hz range they’ll take a higher HPF fine. If you have them way out from the wall in big room then they’re gotta sound real thin real fast with a 100hz HPF and might not have enough response to meet the sub.

 

[aarons915]

Without two subs sitting very close to the mains you can kiss these ideas of 150-200hz crossover goodbye. I personally wouldn’t run a single KC62 above about 75-80hz on a normal LR 24db/oct just to make sure it isn’t localizable. Then on the HPF side you can play around depending on in-room response. If you’ve got the Metas up against a wall or have some room resonances in the 80-100hz range they’ll take a higher HPF fine. If you have them way out from the wall in big room then they’re gotta sound real thin real fast with a 100hz HPF and might not have enough response to meet the sub.

The studies I've seen on localization all place the maximum crossover frequency in the 120-150hz range assuming they are setup and integrated properly, that is an important step that many people seem to skip. Also remember most people use receivers so the crossover is 4th order on the lowpass and 2nd on the high pass, what that means is even if you go up to 120Hz, you're in-room or acoustic crossover is still around 100Hz which should be fine. I wouldn't want to go higher than 120 personally because I have to worry about localization in a 5.1 setup, but in a stereo setup 150Hz 2nd order high pass with a 4th order low pass would probably still be ok. I do agree the room can make a joke out of all of that theory.

 

[Maiky76]

This is a review and detailed measurements of the KEF LS50 Meta bookshelf coaxial speaker. It was kindly purchased new by a member and drop shipped to me and costs US $1,500 for a pair.

The LS50 Meta comes in different colors and I must say, it looks stunning in white:

View attachment 145854

You could sell it as a decoration piece and it would still sell strongly!

Even the back panel oozes beauty and custom design:

View attachment 145855

Love the wide apart binding posts that are easy to tighten and loosen.

Speaker also feels quite dense and solid which is good.

Measurements that you are about to see were performed using the Klippel Near-field Scanner (NFS). This is a robotic measurement system that analyzes the speaker all around and is able (using advanced mathematics and dual scan) to subtract room reflections (so where I measure it doesn't matter). It also measures the speaker at close distance ("near-field") which sharply reduces the impact of room noise. Both of these factors enable testing in ordinary rooms yet results that can be more accurate than an anechoic chamber. In a nutshell, the measurements show the actual sound coming out of the speaker independent of the room.

I performed over 1000 measurement which resulted in error rate of about 1%. Clear high frequency response is responsible for ease of measurement in this regard.

Reference axis is approximately the center of the tweeter.

KEF LS50 Meta Measurements
Acoustic measurements can be grouped in a way that can be perceptually analyzed to determine how good a speaker is and how it can be used in a room. This so called spinorama shows us just about everything we need to know about the speaker with respect to tonality and some flaws:

View attachment 145856

Ah, close to perfection. The only issue is the roughness in the crossover region with some hills and valleys. Directivity which is a metric of how close off-axis sounds are to direct on-axis (what hits your ears first), is very good as well.

The above is substantially better than older LS50 which had an uneven frequency response.

Sensitivity is low at around 83 dB or so.

With both drivers co-located, I could not separate their response but could provide the port/cabinet contributions:

View attachment 145857

I can't figure out that bump around 1.5 to 2 kHz. Could be resonance from the woofer that we can't see.

Back to our spin measurements, here are the off-axis responses:

View attachment 145858

Put that together with on-axis and we get one well behaved speaker:

View attachment 145859

We can see the well managed directivity control better in beam width and horizontal directivity:
View attachment 145861

View attachment 145860

The price you pay for this is slightly narrow directivity of ± 50 degrees instead of the usual ±60 degrees. So I suggest pointing the speaker at you.

Vertically the coaxial driver cleans the clocks of any standard 2-way speaker:

View attachment 145862

So not very critical if you sit at the level of the tweeter or not.

Looking at the mid frequency 3-D directivity balloon, we see the best response since I started showing it (which hasn't been long as of this writing):
View attachment 145866

The globes are not very deformed and nicely project energy forward.

Company touts lower distortion for this speaker versus the old KEF LS50. Let's look at that:

View attachment 145863

View attachment 145864

Looks like distortion in from 200 Hz and up is excellent but down low, even at 86 dBSPL, we hit 100% THD. Unfortunately I don't have comparable measurements for LS50 as that was a long time ago before I standardized this way.

Impedance is quite low at 3.7 ohm and stays there for good bit of the spectrum:
View attachment 145865

Combined with low sensitivity, you need to have a beefy amplifier to drive them.

Finally, for the fans of timing graphs, here are the impulse and waterfall responses:

View attachment 145867

View attachment 145868

KEF LS50 Meta Speaker Listening Tests
I always test speakers with the same set of tracks and in the same order. Usually the first few seconds of the first track tells me most of what I need to know about the sound of the speaker and this situation was no different. The sound was "right" and very nice. For confirm I went through the rest of my test playlist and the answer stayed the same.

Wanting to see the effect of the dips in 1 to 3 kHz, I developed a single filter at 1189 Hz. Getting it to fill that gap requires a Q of something like 7. At that level, turning it on and off showed such basically non-existent difference as predicted by psychoacoustics. We just don't have that kind of disorientation in frequency in that range. I suppose if you wanted to be anal about it, you could fill the holes. It wouldn't make things worse and maybe the combination would make more of a difference. For me, it wasn't worth the time. I was happy with the speaker as is.

Was it all perfect? No. As I turned up bass heavy track, the low bass notes change their tonality and quickly become distorted. Notch the volume even higher and you are greeted with scary crackle. You can visually see this in the driver. It separates from the tweeter which is kind of disconcerting but that is how the coaxial driver works. By the time you see any significant separation/movement of the woofer, the bass starts to change. Push it to move more and you are in distortion territory. The driver is simply too small/lacks the excursion for high dynamic range.

That said, I had no trouble getting usable volume out of one speaker. With two speakers, it would be plenty for most people. Problem is, I am not most people. I don't want to know limits of equipment I use.

Conclusions
KEF moves the bar it set with the LS50 with the Meta revision. I was not a fan of the original but they have won me over with near perfect measurements and listening test results. Make this speaker handle more dynamics in bass and I would kiss the ground it walks on. But that is not there so a notch lower for me. But really, this is an excellent speaker. No doubt about that.

I am happy to recommend the KEF LS50 Meta. Suggest pairing it with a subwoofer if you want to play loud bass though.

Edit: video review posted as well:

----------
As always, questions, comments, recommendations, etc. are welcome.

Any donations are much appreciated using: https://www.audiosciencereview.com/forum/index.php?threads/how-to-support-audio-science-review.8150/

Hi,

Here is my take on the EQ.

These EQ are anechoic EQ to get the speaker right before room integration. If you able to implement these EQs you must add EQ at LF for room integration, that usually not optional… see hints there: https://www.audiosciencereview.com/...helf-speaker-review.11144/page-26#post-800725

The raw data with corrected ER and PIR:

Score no EQ: 5.6
With Sub: 7.7

Spinorama with no EQ:

• Quite a bit different from the data published by KEF
• a bit ragged in the 1-3k range

KEF data:

Directivity:
Coaxial shinning here

EQ design:

I have generated one EQ. The APO config file is attached.

• Quite a few biquads
• Moderate gain

Score EQ Score: 6.4
with sub: 8.6

KEF LS50 Meta APO EQ Score 96000Hz
August092021-115640

Preamp: -2.3 dB

Filter 1: ON HPQ Fc 55.51,    0.00,    1.24
Filter 2: ON PK Fc 117.69,    -1.59,    1.70
Filter 3: ON PK Fc 510.76,    -1.44,    1.13
Filter 4: ON PK Fc 936.04,    -1.45,    5.55
Filter 5: ON PK Fc 1168.57,    2.67,    5.65
Filter 6: ON PK Fc 1869.86,    -1.12,    4.67
Filter 7: ON PK Fc 2684.42,    1.73,    3.35
Filter 8: ON PK Fc 4693.42,    -0.86,    3.40

Spinorama EQ Score

Zoom PIR-LW-ON

Regression - Tonal
Flat after EQ for ON

Radar no EQ vs EQ score
Small improvements to be confirmed (or not) during listening tests

The rest of the plots is attached.

 

[whazzup]

Appreciate the detailed response and the kind welcome! I actually did go with those recommendations by KEF for the WII but with all the talk of much higher crossovers in this thread figured I might be doing something wrong. I’ll definitely play around with both LP and HP a little more but probably won’t stray too far as it honestly sounds pretty good already. Thanks again!

You don't sound like you have a measuring mic, so alternatively you can use this excellent test tone video to gauge the smoothness of the crossover region by ear:

 

[Tangband]

The studies I've seen on localization all place the maximum crossover frequency in the 120-150hz range assuming they are setup and integrated properly, that is an important step that many people seem to skip. Also remember most people use receivers so the crossover is 4th order on the lowpass and 2nd on the high pass, what that means is even if you go up to 120Hz, you're in-room or acoustic crossover is still around 100Hz which should be fine. I wouldn't want to go higher than 120 personally because I have to worry about localization in a 5.1 setup, but in a stereo setup 150Hz 2nd order high pass with a 4th order low pass would probably still be ok. I do agree the room can make a joke out of all of that theory.

Receivers with 12dB HP filtering for the main speaker, is made for closed box main loudspeakers that naturaly rolls of 12/dB oct at their resonance frequency. 12 dB HP plus 12 dB natural roll off = 24 dB. It can then sum correctly to the subwoofers 24dB/oct LP filter. Meaning that in most cases a crossover about 80 Hz is the best if you use a AVR crossover. Only very small main speakers have their resonance frequency at 150 Hz.

Thats how the THX 24/12 filtering works.

 

[Tangband]

Without two subs sitting very close to the mains you can kiss these ideas of 150-200hz crossover goodbye. I personally wouldn’t run a single KC62 above about 75-80hz on a normal LR 24db/oct just to make sure it isn’t localizable. Then on the HPF side you can play around depending on in-room response. If you’ve got the Metas up against a wall or have some room resonances in the 80-100hz range they’ll take a higher HPF fine. If you have them way out from the wall in big room then they’re gotta sound real thin real fast with a 100hz HPF and might not have enough response to meet the sub.

This is a correct statement .
You have a lot more freedom in crossover frequencys with two subwoofers in stereo, than with only one subwoofer.
In practice, even with two subwoofers placed at floor level and the kefs on loudspeakerstands, a crossover much higher than 120 Hz is gonna work less good.

 

[SS55]

Yeah probably not worth it. But definitely EQ! Use these settings for the og LS50 (by Maiky76):

Score after Optimization: 6.36

Type Freq Gain Q
PEQ 65.0, 3.00, 1.78,...
PEQ 166.5, -2.10, 0.88,...
PEQ 550.0, 1.40, 3.72,...
PEQ 830.0, -2.40, 1.83,...
PEQ 1224.0, -1.34, 9.45,...
PEQ 1609.0, 2.11, 2.65,...
PEQ 2764.0, -2.80, 1.35,...
PEQ 4747.0, -3.40, 3.00,...

(+ additional room correction for bass frequencies as well if you have a measurement mic)

Maybe get a sub instead if you don't have one yet?


BTW the Meta seems to have slightly wider directivity than the R3, true point source, smaller, arguably better looking, not as expensive. If you have a sub and/or don't need high SPL I'd argue it's the better option for most people.

Thanks so much, I will try these. Is it possible to skip the first two low frequency eq? Since every room is different it might not be the best for my room. And yes I do have a subwoofer.

My dealer has agreed to let me demo the meta for a few hours in my room so I can do a back to back comparison and see if the difference is worth the extra $500.

 

[napilopez]

Hi,

Here is my take on the EQ.

These EQ are anechoic EQ to get the speaker right before room integration. If you able to implement these EQs you must add EQ at LF for room integration, that usually not optional… see hints there: https://www.audiosciencereview.com/...helf-speaker-review.11144/page-26#post-800725

The raw data with corrected ER and PIR:

Score no EQ: 5.6
With Sub: 7.7

Spinorama with no EQ:

• Quite a bit different from the data published by KEF
• a bit ragged in the 1-3k range

View attachment 146439

KEF data:
View attachment 146454

Directivity:
Coaxial shinning here
View attachment 146444


EQ design:

I have generated one EQ. The APO config file is attached.

• Quite a few biquads
• Moderate gain

Score EQ Score: 6.4
with sub: 8.6

KEF LS50 Meta APO EQ Score 96000Hz
August092021-115640

Preamp: -2.3 dB

Filter 1: ON HPQ Fc 55.51,    0.00,    1.24
Filter 2: ON PK Fc 117.69,    -1.59,    1.70
Filter 3: ON PK Fc 510.76,    -1.44,    1.13
Filter 4: ON PK Fc 936.04,    -1.45,    5.55
Filter 5: ON PK Fc 1168.57,    2.67,    5.65
Filter 6: ON PK Fc 1869.86,    -1.12,    4.67
Filter 7: ON PK Fc 2684.42,    1.73,    3.35
Filter 8: ON PK Fc 4693.42,    -0.86,    3.40

View attachment 146442

Spinorama EQ Score
View attachment 146438

Zoom PIR-LW-ON
View attachment 146435

Regression - Tonal
Flat after EQ for ON
View attachment 146437

Radar no EQ vs EQ score
Small improvements to be confirmed (or not) during listening tests
View attachment 146436

The rest of the plots is attached.

Hi @Maiky76, I think there was a small mistake. The plot that you posted as 'KEF's data' is actually my spinorama =] either way, Amir's results are a bit different so the same comments apply.

Posted it earlier in the thread but here's KEF's spin from their whitepaper. Main difference is their measurement is noticeably tilted downward a bit more

 

[Eetu]

Thanks so much, I will try these. Is it possible to skip the first two low frequency eq? Since every room is different it might not be the best for my room. And yes I do have a subwoofer.

My dealer has agreed to let me demo the meta for a few hours in my room so I can do a back to back comparison and see if the difference is worth the extra $500.

Yes, good point. I would confirm whether the first two filters make an improvement or not by listening/measurements. As you said, depends on your room, how close you have the speakers to the front wall and of course your sub crossover freq.

That's a nice a nice opportunity and $500 extra sounds like a better deal (I believe you mentioned 800 previously). See @phoenixdogfan comment #279

 

[Music1969]

If KEF are reading this, I think it’s clear that there’s a market for a step up product from the LS50, something like an R3 Meta using the LS50 design language.

Using LS50 design language would be Baby Blades , no ?